Interview with John Hanes
by The Gearspace.com Community
4th October 2021
Virginia-based engineer John Hanes hung out with us for a few weeks in 2020 in what proved to be one of the most popular Gearspace Q&As of all time. Not only has he won an astonishing 13 Grammy® awards (and four TEC Awards too!) - he is also one of the nicest guys in the music industry. Working as a team with Serban Ghenea, John has done platinum-selling records with Taylor Swift, Adele, Katy Perry, Rod Stewart and Lorde to name just a few. He was extremely generous with his knowledge and stories, and he continues to occasionally drop into various places on GS to this day. Read on to learn some top tips from a cool dude!
My room is approximately 16' x 16'.
Ok sure. A bit of history; I worked for a while with the architecture firm Studio bau:TON in their TEC:ton division. I wasn't working with the architects directly on design, but was helping to put together equipment packages and overseeing installation of equipment. While there, I did absorb enough design knowledge to be dangerous.
My room is a 3rd floor attic room buildout space. Referring to the picture here; the back wall, not shown, is angled from full 8' height down to a 4' knee wall. My room also has an additional 8' x 8' "entry" area where the stairs enter.
On that back wall I've got some wooden dispersion and some absorptive cloth window blinds.
On the ceiling you can see we created a "cloud" This is an 8' square frame, hung from the ceiling at an angle. The top of the frame has 1" rigid fiberglass panels. the bottom side is faced with POAL (https://poal.net). This doubles the surface area of the fiberglass and any reflections have to go through it twice.
The front and side walls you can see fabric covered 2" rigid fiberglass panels. The corners have foam "corner killers". It is hard to see in the photo, but those panels are mounted on wooden frames that are angled off of the walls. Over the half room 8', each panel is angled out about 10" at the outside edge. Again this doubles the surface area of the rigid fiberglass as well as gives some non-square surfaces.
A couple extra corner killers in the back of the room, a big cloth covered futon behind me, carpeted floors, with a wooden floor cut in for chair rolling.
Because we are essentially only using the near field speakers, we were mostly going for absorption of reflections. We don't really have any isolation built-in except being in different rooms and floors, but we are not doing recording, so that isolation isn't a problem for us.
I've not isolated the equipment very well and do deal with some fan noise, just kind of gets blanked out at some point. If I need to hear detail over that noise, I'll go to headphones.
We try to do updates as little as possible for OS and for ProTools. Currently on High Sierra and ProTools Ultimate 2019.6 Before that we were on 12.7 (we skipped PT 2018).
When we do a major update like that, we'll clone the Disk, and then do a migrate upgrade. Always keeping one Bootable HD with the current working stable OS/PT configuration. Also will make a bootable backup of that stable configuration.
Then on the 3rd bootable disk, clone the stable configuration and migrate to the new Mac OS, or install the new ProTools, update all plugins, etc. When that disk is working perfectly and stable, it will be backed up to bootable back; one copy of the previous stable OS/PT is kept on a bootable drive.
We've got two Satellite Octo boxes hooked up to each system. Yes, I think that there are a lot of familiar faces that are comfortable there. Lexicon 224, 480l, AMS RMX-16; all have that familiarity. I tend to use less of the classic EQ's; there were never that many of them in the old hardware days to get used to using them a lot. It is definitely a must for us to have them to be able to open other people's sessions accurately.
To sound good and convincing as actual doubles, it should be done in production and recorded that way by the singer or BG vocalist. If I want to create some computerized synthy doubles and harmonies, I like the Isotope VocalSynth (check out the bridge on TxT "9 and Three Quarters")
To be honest, most of the productions we receive are very well done and these kinds of decisions are mostly made by the producer.
Now, I can take from what they are doing and offer some suggestions, as well as what I've done in this area where needed.
We talked already about song arrangement; save doubled and tripled parts, or save harmony parts for second and third chorus. Don't be afraid to mute parts earlier in the song so that you can bring them in later. (does that cowbell need to be in every chorus, or just in the last one!)
The song should be arranged like a triangle, add more parts as the song goes.
Your automation within the chorus can also follow the triangle model. Ramp up effects so they build through the chorus.
I would definitely be doing automation to make each chorus a bit bigger than the one before. This can be effects, volume rides. Also a lot of productions are using the EDM pumping kind of chorus these days to add excitement. You could try just a bit of this if it fits the genre.
You can do things like the drums in the chorus can hit the compressor harder than in verses, so they don't really go up in volume, but they crush and densify a bit more. Same can go for the whole song in the chorus on the master fader.
I'll convert sessions in the following circumstances;
There are times we'll mix in 96kHz if files come in that way; small sessions, more acoustic style music, songs with a lot of openness and space.
No different work flow or strategies I think. These things kind of become rote after a while, so I really have to question myself "am I doing anything different"?
I guess one thing would be not worrying so much about clean production and vocals. I think that part of the production and sound of the genre is some built-in dirt and grunge. Vocals can be mumbly and some noises can be left in; that becomes part of the performance. I'm not going to be cleaning up a lot of headphone bleed, mouth noises, etc on a Hip-Hop track.
We'll use these on occasion. UAD's SPL Transient Designer comes to mind, I've played around with the Waves Smack Attack which feels pretty powerful.
I think that we'll continue down the path of the shrinking of the recording studio business, the home studio will continue to grow and become more relevant to more and bigger artists and producers.
I think that probably traditional genre categories will continue to be shaken up; there is so much world music influence now, and cross-genre productions and artist experimentations. Globalization I guess is the word.
Mostly everything has gotten bigger and more complicated. I look at sessions from 15 to 20 years ago; say Musiq Soulchild, and we're talking about sessions with 24 tracks of music, maybe 8 tracks of vocals.
The power that is in the hands of today's artists, songwriters, and producers is pretty immense.
We get a variety. Many people still working at 44.1kHz (and some files received even 16bit!!! ) Large variety of files are 48kHz, 24bit. Some come in 32 bit Float. This would be most of the big guys. Occasionally even 88.2kHz and 96kHz.
Why? Because most of the music we're working on doesn't benefit from a high sample rate and it just makes everything bigger and session resources smaller. I would prefer that everything we receive be 48kHz, 24bit (or 32bit float). This is also one standard for Atmos mixing, so I don't need to do file conversions again.
For us it has been a gradual change, from console and tape to small DAW productions, to now huge DAW productions. The rough mix has now become good enough that people are falling in love with it a bit too much perhaps.
I can't really think of any techniques specifically right now. I'll have to think more about this question and see if I can come up with a blog.
We are not converting sessions to 44.1kHz. We are bouncing out final mix passes at 44.1kHz, 24bit.
Again, the reasoning is that since the mastering engineer is going to be sending 44.1kHz files out, we want to be the ones to make and control that conversion at the mix stage.
No hardware, this is done by ProTools during the bounce to disk. Tweakhead setting.
We don't use that multiple master faders technique here. I'm not sure what you mean by "clip files down prior to mixing". Generally we are adjusting volumes as we mix into the final Master Chain so that the effects of it are in control throughout the mixing process.
We're hard in the Yellow on the master meter at the end of the mix; but if your techniques are working, don't change.
Sorry, no plans for myself or Serban to do a MWTM.
Mac trashcan tower, High Sierra OS, ProTools Ultimate 2019 currently. Two HDX cards, Avid HD I\/O. (I've also got Focusrite Red 16Line and a Focusrite RedNet HD32R I/O's for Dante In/Out for Atmos mixes).
We will generally do the bulk of the mixing on Day 1. Then leave it overnight, and revisit for ourselves on Day 2. If it sounds like it is a good place then, it gets sent to the client, if it needs more work, it might be worked on on Day 2 as well, and then sent out after a fresh listen on Day 3. It is always good to check with fresh morning ears before sending out if you are unsure.
Their revisions can be spread over multiple days. The hard part here is that once you send to the client to listen, you can't touch it again even if you start hearing something that is bothering you, because once you've sent off, if you change it, you can throw off the whole process as the client will start to hear differences between revisions that they didn't ask for.
Touched on elsewhere, but I really try not to listen loud or for long periods of time. Frequent times during the work day while I'm not playing music at all. We don't have clients attending mix sessions, so I don't need to turn up the mains to impress anyone or spend time while someone else is listening.
Outside of work, I'm not going to concerts often, earplugs when I do, I'm using hearing protection with machinery, and I don't drive with the windows down (seriously!)
I've got a good chair for my back (currently All33 Backstrong), I try to stand up often, generally stay healthy, fit. It is an ongoing issue.
As a lot of sessions come from producers with their effects on them, I'll have to answer pretty generally because there are a lot of different people doing different things.
Where we are adding our own preferred / standard delay it is not ducked and not fed into a reverb as a general rule. Specific situations might call for that, but it is not part of the base preset. Mostly we'll be playing with the amount sent to the delay, and then muting or riding volume (either delay send or return) if we don't want delay in certain places.
Where some producers do like ducking, the Air Dynamic Delay is popular, as is just plain riding volume on the delay return track, and some are automating parameters within the delay plugin.
Some do add verb and a variety of other things to the delays.
I'm a really bad Music Historian, I have not consciously studied the trends, so going off the top of my head, here are my opinions. First let me say that we might be more setting mixing trends than following them; so we’re not really doing analysis and comparison to other mixes right now. We are not pulling up frequency spectrums, or making mix decisions based on technical analysis.What I’ve noticed over the years is that the production quality is both improving and turning to ****.
Many of the really experienced (old school?) producers, and the people that they are mentoring, educating, and training are really good. Their attention to detail is exacting, specific, and well thought out. Their technical chops are generally excellent and there is not much wasted production space.What do I mean by that? I’m defining wasted production space as using three plugins to do the work of one. Boosting 1kHz on one plugin and cutting 1kHz on the next. Nested routing decisions of no use (Kick Drum aux to Drum Aux to Beat Aux to Music Aux to Backing Track Aux to Master Bus) with plugins everywhere making it nearly impossible to breakdown a song.On the other hand, many of the self-trained Producers, “Bedroom Producers” (another upcoming blog topic, and I love you and think that you are the future of the business!) seem to be making it up as they go, may have quite poor technical chops, and have a lot of wasted production space and unnecessary complexity. This is kind of one reason that I’ve come to do this Q&A and the blog. I think that the self-trained Producers and “Bedroom Producers” are creating some really amazing, experimental, and leading-edge sounds; I hope that the influence of us “Old School” engineers and producers can help to encourage the experimentation while educating on easier work flows, compatibility and file delivery, and career longevity advice. As far as do an alternate or rock mix such as the Interpol song; I think that we approach it as any other mix. Serban is being sought out to bring his sound to it.
I think you've analyzed this as well as I can. It is another thing that becomes habit and rote, but one thing that I think that we are definitely looking for is vocal clarity and overall clarity which has a lot to do with midrange, frequency masking, etc.
So all of the above.
[top]Do you have any tips to get emotionally connected to a song? I really really like your mix and Serban Ghenea’s mixes on those records: Callum Scott - What I Miss Most
This is actually a topic for one of the blogs that I've written, which will be published here.
I'm not sure what the publication schedule is going to be, so I'll give you a preview.
If the song doesn't hit you right away, find something in it that you like. Maybe the percussion patterns are cool, maybe the bassline makes your head bounce. Find something and work out from there.
If a session does not come to us already subgrouped like this (about 50% probably) then I generally won't add a drum bus for programmed drums.
If it is more like a broken-out loop that has lots of interplay between the tracks, then it might be more useful to do.
I do like a drum bus for Live drum kits because there is a lot more interplay between the recorded tracks that you want to have global control over.
If we're mixing the whole album, then definitely we are thinking of a whole album feel. This is especially important if songs on the album are from a variety of producers.
It isn't necessarily that we are mixing differently than for a single, I think each song gets worked to the same standard, but we might be thinking if overall we are filling the same frequency spectrum, do vocals sound complementary song to song. I would definitely be referencing prior mixes as I go along.
I am constantly A/B’ing the rough mix. I put the rough mix on output A 3-4 which goes to a separate input on my Studio Comm monitor controller. One nice thing about this is that because both A 1-2 (My Mix) and A 3-4 (rough mix) are going to it analog, there is no delay when switching between them. So I’ll line it up pretty much sample accurate and can flip back and forth and hear clearly what the differences are.
With really good rough mixes that we are getting these days, the first step is to make sure that your mix sounds THE SAME as the rough mix. You have to start from where they left off.
Then look for areas that you can improve. Take apart their Master Fader chain if you are using their ProTools session and see what they are doing. Half the time just turning off or replacing bad “self mastering” products with your own will be an improvement.
Work on the technical corrections, make the vocals present and understandable. If I’m not sure if what I’m doing is helping; I’ll do a quick blind A/B test. Eyes closed, flip back and forth between the rough and your mix quickly enough times that you don’t know which you are landing on, and then ask yourself which one you like better and why. Hopefully it is your mix. If not, figure out what it is that is different and why you like it better.
These days you can’t ignore the rough mix. Spend more time with it and really analyse what you like, and what you don’t like and why.
I also sometimes feel a bit of that imposter syndrome. Sending off a first pass is still nerve-wracking. If notes are coming back positive, you’re doing fine!
I’ll have to tread lightly and generally here. There isn’t a secret or a special plugin to achieve the “Serban Vocal Sound”. It is all about his experience, taste, and skills.
Remember we are getting vocals from 100 (I didn’t count, I’m sure it is more) different producers, recorded by 100 different engineers, in 100 different studios, on 100 different microphones, 100 different mic-pres. There is no way to plug any standard anything into that and get a specific sound.
Don’t really do much parallel track mixing; again we’re “Old School” operatives. General effects sends are pretty standard. A bit of chorus, a bit of delay, a choice of reverb(s).
If a producer has sent us some specific effects, we’ll use / modify / work with it. If they don’t we’ll create it. It is maybe a 50/50 split. Remember that some of these producers could and have mixed and released on their own (Max Martin, Greg Wells, etc.) They are coming to Serban to make that final 5% to 10% or so polish. We’re not going to toss out what they have crafted and sent to us, but we will mold it to fit a little better if necessary.
No, not at all here. The mastering engineer should be the one to handle this. We have had issues with loud mixes not playing back at proper loudness on Spotify in the past. Long story short, after consulting with Mastering Engineers and Spotify's technical team about their codecs and processes, we determined that Mastering needs to supply a file to Spotify that conforms to essentially the Apple MFit standard. We never try to compensate for anything in the playback chain; a good mix should sound good on every system and medium. If it does not, the problem is in the end media and that should be corrected at their end.
Yes, when receiving projects not in ProTools, we'll ask for "consolidated" or "committed" audio files. Prefer it to be track by track for instruments (i.e. not a percussion stem, but all 10 percussion tracks). Blended BGV parts are usually ok (i.e. stereo doubles, stereo harmony 1, stereo harmony 2). Effects can be left on instrument tracks as you see fit.
Lead vocals should be "committed" with processing, but not effects, (i.e. Leave on your Compressor, EQ, but not your reverb and delay).If possible we prefer to have any effects on vocal printed separately (i.e. Lead vocal chamber reverb, lead vocal 1/4 delay, BGV 1 chamber reverb, BGV 1 1/4 delay, etc.) We might also ask for non-processed Lead vocals in case we want to undo or re-create something a bit differently; it depends on the producers and how married they are to the exact sound of their rough mix.So it can be a lot of printing parts for you to deliver to us.Producer Jesse Shatkin (Sia, Kelly Clarkson, Paloma Faith, etc.) works this way and his engineer Sam Dent sends us very detailed and very organized files.
Instruments folder, VoxWet folder, VoxProcessed folder, VocDry folder, and VocFX folder. He'll also do screenshots of certain plugins to put on some tracks. We can then pick which of the vocals to use. Usually the Processed vocals combined with the Vox FX, but sometimes better to go with the VoxWet for some parts.
Other times we'll receive parts that are closer to stems. In these cases we at least want to get Dry Lead Vocals and separate effects if possible. There have even been times we've received a stereo music track and three or four vocal tracks; in these cases we have on occasion turned down the project as not mixable. It is not our preference to try to "stem master" someone else's work.
I would definitely group snare mics or bass parts of the same performance. This way when I get a good blend that I like, I can do volume moves on the whole blend as is.
I'll also sometimes send those tracks to an Aux input as a subgroup if I like the blend but want to do an overall EQ or Compression insert or other overall processing, or an effect send to reverb for the whole blend.
I'll also group the same vocal parts (same notes) agains so I can get a blend I like and then ride the whole blend. Also I'll send vocal parts to an Aux Input again so I can do overall EQ or compression on the blend, or send the blend to an effect.
So by Group I'm talking about the ProTools Mix/Edit groups that allow you to solo, mute, and do volume controls (and other functions) across the tracks in the group.
These days, with such high track counts available, I don't do any group consolidations, sub mix bounces, or such things. This allows fine control of all the parts as needed as well as avoids having to undo such a sub-mix when a replacement part comes in. (producer sends a new snare to replace one track, etc).
I personally don't have a big presence on Social Media. No Instagram or website. My credits are here on Jaxsta.
But due to the way they are often copywritten at the label (often despite specific language that we give them), it might look like I'm actually mixing a lot of projects that are actually mixed by Serban Ghenea. (ie. Uptown Funk should be Mixed by Serban Ghenea and Engineer, or Mix Engineer John Hanes). I've highlighted some of my mixes at the top of that page; (Mixed by John Hanes).
In addition, I'm not credited at all on some things that I've mixed. Dotan Harpenau's "No Words", "Numb", and "Letting Go" for example. Proper credit and proper Credits are an ongoing issue.
All that said; I felt really good about the Dotan mixes. I'm really proud of the mixing on the upcoming release of Morgan Saint's EP where I mixed 6 songs. The TxT "9 and Three Quarters" mix was really challenging and I think it came out great.
copying my answer from another. I'd say that I don't tend to use a specific limiter on individual tracks often, though I might push a compressor into a fast attack mode when needed. Most of the time the Metric Halo ChannelStrip is in the MIO mode and mostly not using that or any other limiter.
I think mostly it is the rough mix and familiarity with various producer's overall style (and quirks) from past experience with them.With producers new to us, we'll sometimes talk a bit, get some pointers on direction, but often I think they've picked Serban specifically for his interpretation and input on that. I like to try to glean as much as I can out of the rough mix, figure out what they are shooting for by feel and what feels good for the song and production as I unravel it. Number of revisions can be anything, whatever it takes. From first pass approved to upwards of 15 or 16. Try to never leave a client unsatisfied, but also try to let them know when their requests are not improving the mix any longer.
I think that at this high level of the business we have to have a bit more openness to achieving the artist and producers vision of perfection than most.
It helps that we are working remotely. If you’ve got an artist or band sitting in on the mix that can take a whole day to mess with things on one song. We have the ability to slow the process down a bit by taking our time, sending out the new pass, and then waiting maybe another day for approval or more notes. I think that it removes some of the impulsive requests and makes them more thoughtful.
I think you do need to balance your time and set boundaries where you can.
I would say that in general, and I don’t want to sound elitist here, that our clients are highly experienced and educated in the processes involved.
Part of that education and experience would come from having those boundaries set on them earlier in their careers. It definitely helps to get consolidated and agreement on notes. It helps to have a strong producer who will oversee the direction and be the voice of reason in disagreements.
Through all of that we try to do everything asked but also will say that “this isn’t getting better” when needed.
Sometimes it comes down to trying what the artist wants because they just want to know they’ve exhausted every avenue they can think of. We’re not often going to say “no” to Beck, Bruno Mars, Ariana Grande, or Taylor Swift. When we do say “no” they understand and trust that opinion.
So overall I do think you should work on that balance. Make the clients feel that you are open to all of their ideas, while finding a way to not get run over. Sometimes we do get run over too, and feel taken advantage of with excessively "needy" clients. In the end a happy client is your best advertising.
I'd say that it is a natural growth cycle. However, some of my favorite projects that we've worked on are new artists, first records, and new styles. We don't "weed out" a lot; in fact our managers are often pushing to do more new acts, more trendsetters.Again going back to Music Soulchild and Jill Scott, and some work with The Roots; it was a lot of fun to be in on the beginning of that resurgent Philly Soul sound.My first Grammy win for LaRoux is another one. Not a polished act or technically great recordings, but it had that vibe and feel. My recent mixing of soon to be released Morgan Saint is again another new act that is a lot of fun to help to craft a sound and vibe for. I personally find it more satisfying to be in at the beginning of a new artist's career, a new sound and vibe, and to be a part of crafting that.We are definitely not just catering to established, well financed, and well promoted artists.
[top]How many tracks are in Taylor Swift's "Daylight" and "Death by a thousand Cuts"?
Ok, I'm pulling up my archived tracks for this one. "Daylight" has 109 tracks, not including Aux effects returns. Now that includes a lot of raw stacks for BGV's (the "daylight" verby bgv hit is 16 tracks). Jack Antonoff and his engineer Laura Sisk typically send us a PT session with vocals separated and not bounced together. There are 4 kick tracks, 4 snare tracks, 7 guitars, 27 BGV tracks not including that verby bgv hit. Lead vocals are spread over about 5 tracks (separate tracks for V1, Pre, Chorus, V2, Bridge, etc.). "Death By a Thousand Cuts" is 85 tracks not counting Aux effects returns. There are 24 tracks of music, the rest vocals. Again the Ld Vocal is spread over 5 tracks, there are 10 tracks in the Aahs bgvs.
Bloom is pretty typical size; about 35 tracks of music, 26 tracks of BGV parts. The producer, Oscar Holder, a master of his craft, created and printed for us most of the vocal effect tracks (computer voices, verby vocal hits) maybe 10 tracks of those things, Another 10 tracks or so of doubles, leads, and harmonies.
First, we try to maintain regular hours and schedule. No point in getting burned out on one project to the detriment of another. Nights and weekends are primarily no-studio time zones.
Second; there are usually enough things to do that if I'm not feeling creative, I can concentrate on some technical issues or other tasks like making stems, running backups, or playing around with and learning about a new plugin.
Because we are choosing our own workflow without clients present, if we don't feel like working on Song or Artist X today, we can work on Song or Artist Y. Luckily, being in demand enough, we probably have a bit more leeway in getting a mix out the door. Also, almost any day in front of music is a good day. I love what I do and look forward to coming to work almost every day.
I'll usually start with the whole song playing, listening down and making bigger volume moves as I learn the song. When I've got a pretty good feel, I'll turn everything off and start unmuting in the order from top to bottom of how I arrange the tracks. Drums, Percussion, FX, Bass, Guitars, Keys, etc. . . . BGV, Lead Vocals.
I'll generally do this adding each instrument as I go, i.e. not turning off drums when I start opening up percussion.
Of course I'll solo certain things to hear and work on specific issues. When I've gone through everything like this and the whole song is open again, I'll then usually do my vocal editing; soloing vocals and cleaning them up.
Finally after that, I'll tend to leave everything on as I do the meat of the mixing; still solo'ing to nail down specific parts of course. I think that this method lets me really learn the parts while spending the most time hearing drums and hearing where vocals relate to everything, which are often the key elements needing attention.
I guess I stick to this pretty regularly; but if something else catches my attention while I'm in the process, I'll go in and put that work in on that part as needed.
I'm feeding it from an AES/EBU mult (through a Z-Systems Digital Detangler) of the A 1-2 Output.
My DK meter came out before LUFS were a thing, so I've found that the NBC scale is essentially showing the same as LUFS, but I'm not exactly sure what the scale is technically showing.
Yes, looking at -6 on the output after all master fader plugins; this doesn't mean trying to hit it for every song. This is just my mark to watch out if it's getting pretty loud.
Nothing used across all tracks for full console emulation. We will use vintage emulated EQ, Channel Strips on certain sounds for a specific effect or purpose.
I'm mostly familiar with Waves, Soundtoys, and UAD offerings in these areas. so many to choose from so it isn't really a choice of which sounds best! or most! but which one I remember when I'm looking for something.
Currently using;2013 Trashcan Mac, High Sierra. I backup everything to a hard drive, and then the harddrive to an LTO tape.When hard drive backups are full, the hard drive is retired and stored. Nothing is ever deleted.
Current projects are on local disks and a secure cloud account. Projects are delivered via many types of FTP and Transfer services (i.e. WeTransfer, Dropbox).
When we have a stable OS and ProTools version; we'll not upgrade it for a year or more.
Note: John Hanes participates heavily in this thread, sharing further, detail knowledge about Atmos and Spatial mixing
I've been going by the UMG deliverable requirements for Atmos. I'm not sure if this document is in the public domain, so I'm not going to share it here.
Basics of delivery are, which are pretty much across the board for UMG, Netflix, Apple, Amazon, and Tidal:
We are not dithering the mix. Typically sending a 24bit file to mastering.
Yes, we do receive 32bit float. I'm sure it does have advantages, I must admit I have not carefully studied and listened for any differences.
Mostly I notice problems with phase when lots of drum samples are used. In these cases, I'll usually nudge tracks by 10's or 100's of samples to find a nice phase relationship rather than playing with a plugin.
I think it all comes back to experience, and to knowing the speakers that we are using pretty intimately. We are not really changing our mixes to fit the environment of listeners.
Basics of good mixing will apply anywhere; the listeners are going to modify their opinion of their playback equipment, and not modify their opinion of the quality of a good mix.
In other words; if someone likes the way music sounds on Beats Headphones, they should like the way any good mix sounds on them because they are normalized to that sound quality.
Because the majority of our work is Pop music, which is constantly changing by incorporating elements from more diverse and experimental genres, I would say that making that balance is inherent in Pop music mixing. The producers, artists, and engineers are taking those cues and incorporating them into their work and balancing or smoothing them out to fit a more mainstream vibe.
Yes I always mix into the master chain with limiting on from the beginning of the mix. If you try to put something on at the end of the process it will undo or change a bunch of the work that you already put in. Mix into the glue!
In this way, you can also control the amount of limiting going on as you work; mix getting too loud? Pull everything down a bit. You can do this during the mix because you'll be able to adjust for how it changes things as you mix. If you do it at the end of the mix, it can change sound of the mix. Some individual tracks will get limiters as well as necessary, or the volume of parts within the tracks can be ridden down as a sort of manual limiting.
I would say overall there is a lot more compression and volume rides being done than just hitting louder parts with a limiter. I think that when we were mixing on a console, back in the Teddy Riley days, it was different. We would do the mix and then put the Master Compressor on towards the end. But in those days mixes weren't so loud out of the console and we printed back to 1/2" tape so we needed to get things under control for that.
We create a mix that sounds good. If broadcast or streaming services down the line **** that up, then that is their problem; we're not going to try to guess what they are going to do and compensate for it. Funny story: We had one artist we mixed for who was traveling around the country on tour. In each city he would listen to his songs on the local radio. He called and complained that the mixes sounded different in different cities and wondered if we should create different mixes for different radio stations. We shut that **** down immediately!
We don't do anything after the limiter.
Thanks, that is the gist of what I want to share! We are all dealing with the same issues.
Yes, depending on the song. Often that comes built into the session from the producers if that is the sound that they are going for.
We might also use some clipping as a sound design tool where it sounds right. Lo-Fi is still a great plugin for this; most often used a bit on cymbals, but for aggressive songs I'll try it on drums, bass, 808. I've also tried out the Kilohearts Distortion plugin which I like. Soundtoys Decapitator or Devil-Loc is also a frequent choice.
One thing that I have mixed recently that I've used these aggressive techniques is an artist named Morgan Saint, I just wrapped up her EP which will be coming soon I hope. We wanted a bit of a wild, nasty, aggressive sound on some parts; so I used a lot of these kinds of clipping and distortions to push it hard.
Generally speaking it isn't going to be a ton. For a standard mid tempo maybe 1.5 to 2dB on a compressor and another 1 to 2dB at the end of the chain in combined compression and limiting. For a really aggressive song, it could be 3dB on the compressor and 2dB down the chain of compression and limiting. if I'm approaching 3dB at the end of the chain, I'd really start looking at my gain structure. I'm paying attention to gain staging and gain structure through the whole mix, it isn't really something that you can easily fix at the end of the process.
There is no separate mastering bus. The Master Fader for ProTools output 1&2 has multiple things inserted on it.
Not necessarily. Crush where needed if appropriate if you like the crushed sound on a particular instrument, use volume rides, light to medium compression where needed, and overall use good gain structure strategies.
No clipping plugin on the master bus, though the limiter can be pushed there as needed.
We don't do anything after the limiter, so I have not studied the issue of hotter with transparency as you are using it. But if it works for you then do it!
Thanks, that is the gist of what I want to share! We are all dealing with the same issues.
I corrected my mistake after you quoted me. To make sure I'm not giving false testimony, it should have been Soundtoys Decapitator.
There is no separate mastering bus. The Master Fader for ProTools output 1&2 has multiple things inserted on it.
Not necessarily. Crush where needed if appropriate if you like the crushed sound on a particular instrument, use volume rides, light to medium compression where needed, and overall use good gain structure strategies.
The first part is kind of hard to answer; it is basically the essence of mixing. Controlling elements of the song with volume, compression, EQ, to manage the overall aesthetic and gain structure throughout the song. I'm not sure of the mathematics involved, but I think that a first compressor doing 2dB of compression and then another doing 2dB of compression and limiting would add up to more than one compressor or limiter doing 4dB of work.The second part is easier to answer; it is due to the Loudness Wars. A songwriter will make a demo loud because louder sounds more impressive, right? The artist and producer then have to be louder than that on their production mix. They send to us, and we need to be louder than that because they are A/B'ing the mix to their rough and will say "it doesn't hit as hard". If we send mix that has more dynamic range and is not pushed hard, they will have mastering do it after the fact, because when they play their song next to the current hit, it needs to be as loud.So our theory here is that if it is going to be pushed that loud anyway at some point, it is better to do it in the mix where we have total control over the effects of loudness than to leave it to be done later with possible adverse effects. As I said before, this is kind of a "pro-move" and not everyone needs to be doing this if it is adversely affecting your mixes. It is a difficult position to be in to have to make a mix as loud as the rough mix but fix all of the problems that their loudness creates. Sometimes the rough mixes come in much louder than our final mixes can safely get to and at that point we just have to say "that is not going to sound good as a final product".
When I'm doing the Atmos mixes, I'll typically be dropping the volume by about 10dB. The reason for this is that first, the companies streaming Atmox mixes have guidelines for delivery volume. Second, the Atmos processor at the playback end has some leveling compensation built into it so that it is doing some of the work of equalizing volume across songs.
The trick here is to try to maintain the "crush" and "crunch" elements of the mix if they exist as part of the desired sound with a significant change in overall level.
No clipping plugin on the master bus, though the limiter can be pushed there as needed.
I have a template that I use as a starting place, and then will create special things for certain songs as I feel the need for them. I'm not going to give specific settings or plugins that we use; not my place to do that.
Basic template (generalized)
First in the general category I choose delay settings that have a bit of spread (groove in the ModDelay), and the feedback might have a heavy LPF to clear some room for the main vocal. Adding effects after the delay would be one of those sound design features that I would create as needed for a specific effect.
Definitely adding effects as the mix comes together. I would never mix with dry tracks; basic effects would go on first and then revised or specialized as the mix goes along; especially on vocals.
Same with drums, instruments, etc. Adding effects as I'm working on those sections, and revising the effects as I see fit as the mix goes on.
I don’t think it is a conscious thought or plan These days. As I said I’ve got three standard reverbs set that I’ll send things to, so anything going there would be in that same space. After that is just making choices to what seems to sound good together.I’ll find a really clean reverb such as Altiverb if the situation calls for it, and other times a gritty machine like Rmx-16.
No tips on that sorry. We also seem to be constantly buying plugins and upgrading others.I think the best thing you can do is to make your computer hardware last as long as possible. Get a stable configuration and hang on as long as possible.We don't actually update the main ProTools system very often. Still working on 2013 trashcan Mac. I think we’re still running High Sierra OS. Maybe next year we’ll try the new tower Mac and new OS.
8 core, 64Gb ram
Yes, lots of back and forth notes and revisions. We'll keep working on the revisions as long as they have notes. Sometimes it gets to the point where we say it is going in the wrong direction and enough is enough.Yes, using streaming for real-time review. We're actually using a product that is obsolete and no longer supported for this, so no sense in me naming it I think. It is preventing us from upgrading the OS on the computer that is running it, so we'll be looking for another solution.Mostly nothing specifically on the vocals for a bit of distortion or grit; the Waves CLA-76 can add a bit, overall Mix Buss chain decisions can add a bit. Sometimes if distortion is specifically called for Lo-Fi is great for that.What brings clients like Max Martin back to you again and again?
First, of course, is giving them good mixes, so let's move beyond that. I believe that there is also a comfort level for them that a second set of ears that they trust has double checked their work.I think a big part of this is that I approach this as a service industry. Each client gets one-on-one attention. If they have an important deadline, we'll make room for it.I do everything I can to provide ongoing support. If an artist has a show coming up on SNL and needs a special version of the mix by Friday, I'll make it happen. Need a radio edit for Italy? Done. Need a TV track with the Lead vocals only in the chorus, but down 10dB, Adlibs removed except for that one that goes "ahhhhh" at 3:15? Done, you'll have it in 15 minutes.Need stems for a mix we did 10 years ago? If I've got them printed already, you'll have it in an hour. If I didn't print stems back then, give me a day.The mix is done, but now the artist is changing the lyrics and sending a completely new set of vocals? Sigh! ok, send it over. What your deadline is tomorrow! I'll get right on it.
There are a lot of producers out there who can make a mix good enough to "release". If you heard their demo mix, you would say "that is a great mix!. Why do they then go to someone like Serban?" It is true that sometimes Serban might do little things to the mix. Over 100 tracks, lots of little things add up. Adding that final 10%, 5%, or even 1% of polish is what they are after.
They could probably spend the time and effort and do that themselves as well. But is their time better spent writing and producing a new song, or just taking a well deserved break; or should they be sitting in the studio agonizing over the perfect reverb decay, the last bit of noise on the vocal tracks, or the last .5dB of balance between the kick and bass.
When these very busy and in demand people find someone that they trust who can do what they want as well as, or better than they can do, they have found a valuable ally. When they can trust someone to take the burden of finishing the mix; dealing with pages of notes from producers, artists, A&R, management, and sometimes the artist's boyfriend or girlfriend (for real!) that is a major resource.
Sometimes barely touching something, but still making it sound 2% better is the greatest of skills. I would say that much of the mixing that Serban does is among the hardest that there is to do; because it comes in so well done, you have to find a way to make improvements, not lose the thread, and satisfy lots of people with sometimes very different opinions.
Mixing isn't about just pushing around EQ's and dynamics, it is dealing with the opinions involved that is the hardest part.
Yes, generally we try to stick to M-F, 9am or 10am to 6pm. We like to have some time for normal life stuff and because we don't have clients visiting it is much easier to be in control of our own schedule. Certainly there are times when we need to come in on weekends, or come back in the evening after dinner and family time to catch up or knock something out. Yes, lots of ear breaks. Also we aren't necessarily working on one song all day long. Might pull up and work on 5 or 8 songs throughout the day. Some might be starting a new mix, some might be in the critical listening nail it down phase, and others might be little tweaks for producer's notes and comments.It can be a lot of pressure, I definitely try to eat healthy, workout (my gym opening here this week!) I try to listen to my Apple Watch when it tells me to stand up.
Yes, did that when I was younger and luckily survived.
Can't plan for how long a mix will take, but if there is a deadline, it is not ours to enforce. Can't really base it on the demo either; you might get an amazing demo mix, nearly complete, and then they'll start adding this, redoing that, sending new vocals. Every project and every song is a unique beast.Generally we are sending out a mix for review and will get emails or texts back over the next day or two. We'll be able to pull up that mix, knock out revisions, and send a new pass out. Let's say that part of the process is a half hour. Now it's another day for revision notes to come in again; so that song is put away for the day and we'll do notes for another.Some clients prefer a Facetime video chat with mix streaming. These will be scheduled as they are available. Some take 5 minutes, some might take a couple of hours of try this, try that, what if we . . . How about . . . .We somehow find a way to get each client the time that they need; it is just a balancing act. While waiting for a video chat session, can knock out a couple of little tweaks on something else.On occasion you get a v1 mix accepted as the final release. You either feel great! or suspicious that they're going to come around a week later and have notes. Revisions can be anywhere from v1 - v3, some going up to V15. These can be for major reasons, or there might be 7 passes of nudging various adlibs up a little here and there.
Some of both. Serban mixes until it feels right, of course with suggestions and notes from artists and producers. When I'm mixing on my own, I'll sometimes reference similar genres or artist suggested material. Of course I have a wealth of pre-mastered files to choose from.
If I'm doing something new to me like a metal mix, I'll find a good representation of that genre. I don't try to sound exactly like it, but I'll try to get in the ballpark and then use my own judgement and feel. I'll just be A/Bing and working overall to mix to that standard. Nothing special like automatic EQ matching or 'one click' mastering that is now available in Ozone for example. I don't use those types of tools.
Yes I did one as a spec mix! I did a mix of Saint Ansonia Feat. Sully Erna "The Hunted". Here's how it went down from my perspective. My manager got the word that they wanted to test out about 6 mixers. One test mix of "Hunted" to see who would get the whole album project.I got the files, did the mix. This was a one-pass deal, no notes or back and forth from artist or producer. They wanted to evaluate each mix on its own with no feedback or input. The story that I got (and who really knows, maybe they told each mixer the same thing) is that in separate completely blind judging, the management, label, and (I think) producer all picked my mix.
When it came down to the final say, the band wasn't really comfortable sending off the songs to be mixed remotely and wanted to be hands-on in the mixing. Unfortunately I can't release my mix for you. I've had feedback from a very metal friend that he liked it better than the final release; though they did a bunch of additional production kinds of decisions in the final mix, he had the benefit of band and producer feedback and input. On listening to the final release, I can say the my mix doesn't necessarily conform to the general treatment of the genre.
I would say that my mix sounds much more like what it would be like if they were playing down the song in front of you without a lot of post production. Not like a live show mix, but let's say an authentic performance capture.I really enjoyed doing this mix, and I think that if a Metal artist ever wants to find a bit of crossover audience, my mix would stand up for that.
Monitor path - Avid HD I/O analog out to Studio Comm 69A monitor controller. ProAc Studio 100 powered by Bryston 4BSST, no sub. Quested VS3208b and 15” Quested subwoofer are self powered.
Headphones PSB M4U1 through Oppo HA-1 headphone amp. AES/EBU from Avid I/O to Oppo. I chose the PSB because they sound kind of close to the ProAc. Mixing on ProAc 91.5% of the time, headphone mixing 8%, Quested .5%.
I would like to re-work my MidField setup at some point so I would use it more. It is not tuned. I kind of inherited the Quested’s from a friend who needed to raise some cash way back when.
I’ve not blown too many ProAc woofers recently; I’m the only one who controls the volume now and I know to start soft and clean up extreme low end before getting loud.
Listening levels: about 85dB for critical listening. 95dB is loud listening for me to feel it, 65-70dB is softer listening for longer periods of time for basic levels, learning the song, editing, printing parts. Almost never doing any mono listening, but I am looking (correlation meter) for and listening for phase issues.
Headphones for bass checks, listening for noise, another perspective. We designed and installed the room treatments ourselves. We have not had them tested and calibrated. Working almost exclusively on the nearfields and having long experience in the rooms with them is the key to consistency. I don’t find the ProAc overly bright.
I feed the DK meter AES/EBU signal from a mult of the ProTools main output. I don't have a microphone hooked up for loudness.
I mainly use it on the NBC scale, trying not to get too far above -6, some hot tracks peak above this, ballads might fall more into the -8 range on average. These are levels post all processing, on the output. I also will use the 1/3 Octave mode, occasionally the FFT, and I watch the correlation meter a lot.
The Grace m908 monitor controller that I use for Atmos work has a built-in microphone for SPL which is really nice for checking listening levels.
The car checks are great when transitioning to a new studio, or a new speaker system. For many years after we moved into our current studio, we would do car checks. After a long period of time that faded out. For anyone working out of different rooms on a regular basis it is a great additional tool.
Earbuds only for casual listening for me. Computer speakers are really for auditioning files quickly by hitting the spacebar. They are on the output of the computer, not connected to ProTools.
As pointed out to me in a separate question, I may be using some technical terms incorrectly.I said "I mainly use it on the NBC scale, trying not to get too far above -6dBfs, some hot tracks peak above this, ballads might fall more into the -8dBfs range on average. These are levels post all processing, on the output."
I should not have used the term dBfs here. I'm not actually sure what the scale for NBC is, but it is closest to the currently used LUFS.
On the DK Meter, if I switch over to DMU1 I'm pretty much bouncing between 0 and -1. On the 40C scale, it would be around +14, VU is crushed to the top of the scale. These are all on a Loud mix, most would not be pushed quite this hard.
Nothing I can think of is specifically done for male vocals. It uses the same tools; Compression, EQ, De-Esser. I always try to mix vocals for intelligibility and artistic presentation. I think it would be hard for me to mix my own vocals (if I was a singer). Does anyone like the way their own voice sounds on a recording? Again, no special specific treatments on Bruno's vocals. Great vocalist, great recording engineer (if you don't know, Charles Moniz not only records all of Bruno in studio, but also does his Front of House in concert), great production, and great direction and collaboration with Bruno in the mixing process.
Yes, I've touched on this in a few other answers. I try hard to be aware of listening levels, I don't really time myself. Frequent breaks of not listening or listening low are normal. I don't use the big monitors much at all, use the Headphones if I need to really concentrate or pay attention to very soft details.
Outside of work, I protect my hearing when using machinery, power tools. I don't go to concerts often, and when I do I always have earplugs with me. I don't listen to music on EarPods very often, and at low levels when I do.
I encourage everyone to look for other sources for technical recommendations.
I don't have any special insight specifically. I think we knew at the time that it was great and special. First Katy is a special singer and lyricist. You put that talent in the hands of Dr Luke, Max Martin, Ammo, Benny Blanco, Greg Wells, Bonnie McKee, and all of the other talent there; you're going to have something special.
I don't want to say it was just another job, because we were excited and happy and thrilled to be working with all of these guys and girls, but I don't want to come off too callous here.
Looking back there are a number of projects that hold a special place and that I consider to be part of history and not just a part of our body of work. Katy Perry "Teenage Dream" certainly is one. Some others that stand out for me.
I also thought that Jewel "0304" was really special but it didn't get the same critical acclaim.
Country stuff, I’m still sent a lot of live drums, pop not so much anymore. Even the country stuff is often layering samples and loops in production.
Occasionally we will augment a kick, snare, or cymbal with a sample, pretty much never doing drum replacement. If we are sent real drums we will work with them and use them. Any decision to do any drum replacement would be done before tracks are sent to us.
In my mixing projects specifically, I don’t like to put in my own samples. I know that the producers like something they are getting with the sounds they have so I make it my mission to figure out how to make what I’ve been given the best that it can be. I like the challenge of working with the sounds I’ve been sent as opposed to trying to make a quick fix and drop in a sample that I like. I don’t want everything to start to sound the same.
I like to process a live drum kit through an aux buss with a channel strip on it overall; also doing things on the individual channels. Mostly this is because when you’ve got good balances on a live kit it is easier to then do overall changes to the aux. I usually try to start with the OverHeads, and then add in the kick, snare, hat. I'll often do a lot of editing and compression on the toms to just capture the meat and mute the bleed of other drums. Much of that sound is otherwise in the overheads. I don’t tend to use much parallel compression tactics unless it comes that way from the producer.
I generally prefer to treat drum tracks separately rather than bussing all kicks, all snares together. If it is useful for a specific reason, then summing is fine as well. If there are a number of samples, try playing with their phase instead of just trying EQ and Compression. (If I slide this kick sample 100 samples left or right does it sound better or worse?)
Drums are usually the next thing after vocals that require a lot of attention and mind melding with the producers. Everyone is careful about their drum sound, so If I can see that a producer has put a lot of thought and energy into their drum production, I’ll tend to leave it mostly intact.
A lot of the drum vibe is created during production. on Dua Lipa “Blow Your Mind” those drums are very well done by Jon Levine. I just looked at that session and it is about 36 stereo tracks of drums and percussion. The goal during mixing of these kinds of productions is to maintain all of the work that the producer has put into it; we might not be touching much in the drums at all in this case; a bit of low roll-off here, a bit of high here, small level and balance adjustments. We might be doing a bit of panning changes to spread the drums and percussion, adding or tweaking some reverb sends.
The bulk of the drums mixing in this case would be polishing the work already done, fixing any problems, and mixing the rest of the song and the vocals so that all of that work and detail is not lost.
Mostly nothing special as far as plugins; I’m setting up and working like I used to on a console. There is an EQ, Gate, Compressor, Limiter, on each track. I’ll start with using those tools; if anything needs more specific effect or sound, I’ll start digging through plugins.
After I've got to the good rough mix stage, I might mute everything and then start bringing in parts starting with drums and working through all of the instruments to vocals; just to hear how everything builds and adds together.
After that, mostly everything on. There will be times to solo all drums and listen, or solo all vocals and listen of course; but I'll not mix separately as an instrumental and an acapella for any long period of time.
It is difficult to do in a small room. Look at the woofers themselves and visually see what the low end looks like as the woofers bounce around on a mix that sounds good elsewhere. You can see some of the low end information. Check low end on good, not hyped, headphones.
Generally we ignore everything downstream of mastering and just work to create the best sounding mix we can. There is only one situation that I can think of where we have intentionally created different mixes for radio is with Country Radio. They seemed to really mess with the Bass so much that we have done different mixes for Keith Urban to send to radio than were used for other releases. His music is very pop influenced, so it doesn’t always blend with the typical Country Radio preset.
I don’t really love it or hate it, I just kind of ignore it. I think that most people only listen to broadcast radio in cars these days , which is far from an ideal listening environment anyway.
This is a really hard topic. I think that I've been really lucky in the path that my career has taken. Let's say I've gone in all-in several times and been lucky to have a winning hand. There can't be a plan for that.
There have been several times during my career when I've been near broke, not getting any work, getting work and not being paid for it, and wondering if it will ever get better.
I've seized hold of every opportunity I've been given and tried to make myself indispensable in every situation. My advice is that it will be really hard, you might get used and abused, it is a grind. You might get lucky, like I have, and make a career out of it, but you need a backup plan and a paying job until that happens.
If you get a job as an assistant engineer, absorb everything, learn everything. Make yourself available anytime, anywhere to be ready to work. Be kind, be humble, understand that you are easily replaceable and find a way to make yourself not easily replaceable. Know how to learn on your own; don't ask a lot of questions when people are busy, figure out the why and how by watching and listening. Ask questions later when everyone is relaxed.
Don't make any enemies; you never know who might be the next A&R or head of a label for example.
Protect your hearing; carry earplugs, wear them in noisy environments. Don't go to a lot of loud concerts. Don't be afraid to step out of the room or put in earplugs if someone cranks up the mains.
Practice mixing; maybe you can share work with other young engineers. You mix theirs, they mix yours. Feel what it is like to work on someone else's songs and sessions. Practice the communication and understanding of trying to put mixing notes into words.
Practice recording; find a way to get studio time if you have any access. Maybe after you've documented and are breaking down a session, have someone play the drums while you move microphones and listen. Have someone play guitar while you move mics around the speaker cabinet.
The most important thing getting a foothold for me starting out was making sure I was impressing people. Put the clients first. Show up early, stay late, anticipate problems and solutions. Know how to do things; how quickly can you change a blown woofer, what do you do if a mic cable is bad and crackling, etc. You will not be judged first on your abilities with sound. You will be judged first on your reliability. You'll be judged on your ability with sound later, but you have to get to that place first.
I'm using the PSB M4U1 Headphones. I like these in particular because they sound pretty similar to my ProAc speakers and they are not heavy. I have an Oppo HA-1 headphone amplifier that I use for headphone listening. When monitoring the binaural downmix of an Atmos mix, I'm using the headphone output of the Grace m908 monitor controller.
I'm not a producer or writer, so I don't have any real-world experience here to give you.
I might suggest some divisions of labor though. When you start mixing, try to use what you've already recorded, try not to fix a mix issue with adding new parts. Take some time between your production phase and your mixing phase. Take a few days, maybe a week of not listening to your song; then approach it with a clean perspective.
I don't do any additional production in mixing; I am not really a musician these days (played trumpet from grade-school through college) I don't play keys or guitar.
I think I've seen Serban throw down a guitar part a few times in all of the years we've been working together.
Additional production is not part of our scope of work. We would never replace an element with one of our own. Producers will often send a replacement part if something isn't feeling right to them.
I’ve got a control surface; a mouse is the control, and a mouse pad is the surface.
Seriously, pretty much all automation is drawn in on the edit screen automation lanes by mouse; occasionally I’ll turn on automation parameters and ride a “knob” live via mouse. I like the precision that drawing in via mouse gives; instant square volume changes between words, or completely smooth point to point rides.
This is just how my interaction with ProTools evolved; when we started there were no real control surfaces, and when you’re comfortable with a mouse it seemed silly to buy a Pro Control or an S6 for thousands, tens of thousands of dollars to replace a mouse.
I will get basic balances going and pretty quickly start doing the obvious volume automation moves that need to be done as I’m learning the song. These would mainly involve leveling out vocals, as they have the most variability within the song.
As the mix proceeds, the automation will get more and more detailed. Smaller volume moves, pans, effects sends rides or mutes. I might turn on and off some EQ bands for different song parts of the same track. Say a high hat is poking through a snare track or looped drums, I might ride every hit down with automation rather than messing with trying to EQ, set a compressor, or other “fix”.
Automation on the master, yes sometimes. A good way to make the chorus have more energy and drive than the pre-chorus is just pop it up 1dB on the master volume.
After all of this, though, probably not as much automation going on as you would think. I’m doing automation during the whole mixing process, but mostly small moves or setting big moves for sections of the song (i.e. panning wider in choruses, or bypassing an effect in the chorus). A lot of the “tweaks” that artists and producers ask for become small automation moves; turn this down, turn this up, for certain sounds or certain parts of the song. More reverb here, less delay there kinds of things.
The interesting thing that I’ve gotten into recently with Atmos mixing is that I’m doing a LOT of automation there for panning (there are 28 lanes of pan automation on a stereo track going out to a 7.1.2 Atmos output).
So in Atmos mixing, I find myself doing a lot of selecting automation types (write, latch or touch), setting automation modes (write on stop), and dragging handles on the surround panner, and then going back and grabbing different handles. So there I might do one pass where I’m panning back to front, and then a second pass where I’m panning top to bottom. So overall the sound sweeps from top back to front bottom. I wish I had a motorized 3-D surround panner for this.
All of the above. If the problem is too much low frequency, then don't boost the highs. Identify the problem and apply the simplest solution. Boosting is more likely to drive the track to distortion, and everyone likes to add more of this and more of that, but try to avoid only boosting.
I run into the same problems when ProTools sessions come in from some producers that have complex and nested aux tracks.
There is not a pain-free solution and there are not perfect stems. The purpose of the mix is to make the Stereo track sound the best; if the stems suffer because of this than too bad! If you want me to mix for perfect stems, then don't ask for a perfect Stereo mix.
A smart mix template for *near perfect* stems is to not use any sub-mixed aux busses, not put any effects on track inserts, and not do very much Master Bus processing. This is not realistic and is not the goal of a Stereo mix.
Just mixing these days, no tracking. I think the last project that we fully tracked and mixed was Kenna "New Sacred Cow" or NERD "In Search Of"Our studio is a private facility, there is no good information available in regards to the studio.
A bit of both Plugin De-Esser on the tracks, but also manually pulling down anything that still stands out too much. I don't want to crank the De-Esser so hard to smash the big ones that it makes the whole track lisp-y. (Yes, I read the thread about The Weeknd! Glad that our mixes passed inspection). So a lighter touch on the Plugin and manual riding.
It kind of comes naturally after a while, but a few things that I do that could become pointers.
I Always mix with the vocals on. I don't mix an instrumental and then add the vocals. First step after getting to the "good rough mix" stage is get the vocals sitting right, and then go and mix the instruments with the vocals in the mix.
Listen really soft; can you hear and understand the words? If not, vocals are too low or muddy. Listen without concentrating; play the mix down while doing something else. What sticks out, what is missing when you are not in full concentration mode. Any particular EQ, compressor, verbs, etc. that I use and rely on are not because they are in any way special or amazing, but because I am comfortable with them.
Mostly by what sounds good in the mix; I can't really solo a track and sweep for bad frequencies because they might only be bad when interacting with all of the mix. Only after identifying which track I might do a bit of sweeping to find the sweet spot to cut.Mainly I'm identifying problem areas by simple listening as well as looking at a 1/3 octave analyzer, and now we've got really good frequency graphs in many EQ plugins.
We have one standard mix buss chain that is used on everything. We mix through it, meaning it goes on at the beginning of the mixing process and the whole process of mixing is done with it in place.
It is mostly “set and forget”, but there are times when small adjustments might be made to overall EQ, compassion, limiting, etc.
When we get sessions from producers that have their own mix bus chains in place, we’ll review what they are doing with each plugin. Some things we will keep in place because it has become an integral part of their rough mix. Other things can be removed or replaced by our own tools. Quite often just removing some things from the rough mix chain will make the mix better right away.
I won’t share exactly what we use here, as that chain was developed and perfected by Serban and is his “proprietary” information to keep to himself if he wishes. We generally leave little for the mastering engineer to do. It takes a great skill to take it to that “sounds like it’s already mastered” place, so definitely a pro move.
So on a typical loud mix, at the loudest part of the song, on my DK Audio meter, set to NBC scale which is AES full scale, we’ll be sitting around -6, with peaks to -4. On the Waves WLM Meter, which I normally look at just for Atmos mixing, it is pretty crushed at about -6LUFS at the loudest part of the song.
On occasion the mastering engineer will ask us to bring it down, and rightfully so. Other times, the artist or producer will say we like the mix better than the mastering, can you ask them to pass it straight through. The theory here is do everything that will affect the sound inside the mix, which the artist, producer, and mixer have tweaked to perfection.
If a ProTools session comes in from the producer with separate Vocal Bus and Mix Bus, I'm not going to remove it. They created it with that processing in place and it will change things too much to rework it after the fact.But even in these cases, everything is still going to go through the Master Bus chain and output 1&2. The Master Bus plugins and settings might need to back off some as a majority of that compression is being done already.I'm not going to create a session that is set up that way, I definitely prefer what you call Old School! I guess I am getting old.
No trim plugin. Consistency comes from experience and listening to the volume, looking at the meters, and knowing about what levels I should be at to hit the Master Bus about where I want constantly. If things start to get too hot, I'd rather make a group of everything that is feeding the Master Bus and trim all of the tracks down equally rather than using a plugin on the master to do it. Always trying to maintain proper gain structure and not just compensate for bad gain structure at the end of the chain.
Generally no. Again if the producer has sent us his ProTools session and crafted his mix with this in place on his Master Bus, We will evaluate it and see if it is needed to maintain the production values he has set up.In my preferred world, low end should be crafted and nurtured on the individual tracks where you have much more control over it; just adding a bunch to the whole mix feels very imprecise to me. Chances are you're just going to be boosting a bunch of low mud that has not been filtered out of individual tracks.
I think other than loudness, it is doing light touches. As I said before, we are mixing through this chain from the beginning of the process, so all of the mixing decisions and tweaks are being done on the individual tracks but heard through the whole chain. So if I want to widen a sound, I'm not widening on the Master Bus, but I am hearing the consequences of widening through the Master Bus. Both full range and multi-band processes are used.
Subgroups and busses are created as needed for things like BGV's to apply processing to all equally. We don't have, for example, a Bass Buss or Drum Bus that has stock presets. If one is needed, it is created for that mix.
I wouldn't generally mix into a high/low pass eq. You're just going to compensate elsewhere by boosting or cutting those frequencies on the individual tracks. I would prefer to fix these things at the source. Nothing else to add on the Mix Buss!
If the Vocals are driving the compressor too much, then they are too loud or you are using too much bus compression. If I take my bus compressor and crush the mix ridiculously, it doesn't change the balance of music and vocals, it only changes the overall volume and of course distorts.
I think that we all have grown used to music sounding as if it has come off of tape. Even music created when tape machines are a distant memory are influenced by the past benchmarks of music created on tape. It just sounds good, right? Apply as you see fit.
Stems are a part of our standard delivery to the label. These days I offline bounce all of my stems.
I don't bounce the stems right away as the mix passes are delivered to mastering; it is something I take care of as I find time between more urgent requests. One benefit of 100% in the box is that I can open and close things as needed to do these tasks; there is no reset needed to go into stems mode or mix mode.
While I'm bouncing stems, I'll be doing paperwork, catching up on billing, or just taking a hearing break.
If you're talking about doing a side-chain to pump the whole mix; we don't do that. We don't mix with side-chains pumping the master bus. Any side-chain is within the tracks or groups on an aux bus.
If you're talking about generally the overall processing, limiting, compression of the master bus not hitting the same, I don't worry about it. I print all the stems through the master bus, they do come out hotter than the mix because of the difference in the overall bus compression, etc., I'm not doing any side chain or compensation for stems.
I do just solo the parts for the stems and print what plays, reverbs, effects, and all.
The purpose of the stems is not to re-create the mix perfectly when they are all played back together. It is not going to happen. We do sometimes get complaints about this and I will tell them to drop everything by 6dB, put a bit of master bus compression on the whole thing and you're 99% there. If a special mix is needed, we'll do it here, don't try to do it from the stems.
The purpose of the stems is for the artist to get remixes done where they might use some of the elements in a new production, so it doesn't matter if they perfectly re-create our mix.
The other purpose is for the artist to be able to do live shows and use elements from the mix where again it doesn't need to be a perfect reproduction of the mix. The FOH engineer is going to be making their own adjustments and probably blending in live instruments.
So one thing that goes on here is that we work with so many different producers and artists using so many different plugins, that we are buying plugins regularly to be compatible with the ProTools sessions as they are sent to us.
My AAX plug-ins folder has over 1000 items in it (when all folders are open).
As the producers are constantly experimenting with new plugins, we seem to collect more and more of them. To narrow that down to a Top-5 of what I choose when I am choosing the plugins comes up with a list of things that mostly I have used for a long time and are familiar with.
Some producers will have taken care of enough of this before they send that I don't need to do this extensively on every song that comes in.
Here is how I clean vocals: First I want a good rough mix going; I'll get the vocals at least with basic settings on compression, EQ, and de-esser. I'll get the volumes and balances close. Much of the time this is done for me as the producer has sent the ProTools session of his rough mix. I don't run a complete top to bottom pass of RX-7. I'll open up a few Audio Suite plugins (turn target off so you can open multiple plugin windows).
For mouth noises; I’ll select maybe one second of the audio around the click I hear and zap it with Mouth De-Click. I’ll listen back to what it has done, and adjust if necessary.
Some clicks are better resolved with standard De-Click. Plosives I’ll hit with the EQ3 filter, big breaths and some excessive sibilance I’ll hit with the Gain. I’ll also take this time to cut (region separate and mute) and fade the beginning and end of phrases as needed, and cut any noise where there are no vocals.
I have to be careful here of room noise appearing and disappearing as the vocals come in and out, so sometimes it might need to be left in or faded out slowly. Other times if it is really annoying I’ll go into full RX-7 Connect / RX-7 Monitor mode and deal with room noise, hums, or other, deeper issues.
I’ll do the same to each BGV part, but here I’ll often completely mute breaths and make a quick fade in on the vocal. Also sibilance on BGV’s can be treated much more harshly.
If voices in the BGV’s stack are out of time with each other in the attack or release, I’ll fade back the ones that are early or held long. Often this is to make sure that the Left and Right panned vocals are beginning and ending at the same time, and timing with the Lead vocal is also suitable. I’ll do this for each stack of Background Vocal parts, and any other vocals (vocoder parts might need a lot of breath reduction, gang vocals might need a lot of dead space mutes).Again, as I’m doing this, I’m listening back and adjusting everything I’m doing as I go.
So it can be a very tedious process in a session with tons of stacked BGV’s, or with bands with a lot of lead vocalists (boy band, K-Pop).
No shortcuts. Especially on vocals, because we are naturally so attuned towards vocal comprehension, each fix needs to be consciously made.
Yes, this sometimes happens; usually towards the end of the mix they start to regret that as those bad edits or clicks suddenly become much more prominent because everything else in the mix is nice and clear and clean and present.
There are some producers that we work with that I know don't want things fixed or changed much, so I'll take a lighter hand with those.
Also, if the producer puts a vinyl crackle track on the song, I'm much less inclined to go in and spend a lot of time fixing vocal clicks.
What you do is hire Max Martin and his WolfCousins to do production! Seriously, this is not something that is created in the mix; it is the producer's choices and skills.
See also new mix by Serban for Tones & I "You're So ****ing Cool". Beautiful Steve Mac production and wide!
Don't be afraid to pick a pad that is a bit out of phase for this.
To be honest, mono compatibility isn't really a top concern. I don't listen down to the mix in mono.I do pay attention to it as I'm working. I am listening for phase and watching the phase meter on the DK Audio meter that sits directly in my line of sight.If I hear or see something going out of phase, I'll find which track is doing it an see if flipping the phase on one side of the stereo channel makes it better, or if it changes the sound too much. I won't sacrifice the sound for mono compatibility. Some choices by a producer to have a super wide sound in a mix might make it mono incompatible, but I don't think I've ever heard anyone really care about it these days.
I don't normalize rough tracks. If the recorded signal is so faint that I can't even see the waveform in the standard view, I might AudioSuite Gain the tracks up to that I can at least see waveforms.
Other than that it is just using regular fader volume or gain settings on any compressors to bring the level up. Also the Master Fader volume can be brought up to get the gain structure set correctly before it hits the master bus plugins.
Achieving "perfection" with quick turnarounds and tight deadlines
Several levels going on here. First, do we 'create' chart topping mixes? Or do the songs top the charts and drag the mix along? Would a 'bad' mix of a Taylor Swift or Bruno Mars single not top the charts? It is kind of a chicken and egg problem isn't it.
Obviously when topping the charts we are blessed to work with really good artists, usually very well done recordings and productions, great songwriting. I think a great mix makes the artist, producers, and A&R more comfortable that they have a great product to release.
Dealing with the demands of clients, deadlines, and timelines is a universal issue. First, deadlines are mostly meant to be broken. How many times we do a mix under a deadline, submit it, it is approved, even mastered. Then two weeks later; “we’re going to send you a new bass-line, and the singer has some new ad libs to drop in.”
So you do what you can by the deadline, and leave it up to the people who enforce the deadlines to say if the song is done or not. Some deadlines are hard and real. In those cases it might not be the best mix that can be, but obviously in something so rushed, the perfect mix is not the goal of the project.
Mixing songs for TV releases is usually the hardest deadline. In the mixes for Songland that Serban mixed; Jonas Brothers “Greenlight”, Julia Michaels “Give It To You” the files arrived about 3 days ahead and had to be ready for TV broadcast with an absolute hard deadline. For Ariana Grande’s “Thank You, Next” single files arrived in the morning and had to be out to mastering by that evening.
In these cases you have to fall back to basics. Make the mix sound good overall, don’t nitpick fine details. Fix things that stick out. Don’t get panicked and spend every hour working on the mix; get a good mix going, then walk away. Work on something else for a while; come back with a fresh perspective. Do the mixing in many shorter sessions rather than one long one; it is totally counter-intuitive when a deadline is crushing you; but your first 30 minutes of mixing might be your most productive; so create those first mix revelations over and over again. Don't concentrate too hard, let you subconscious hear things that aren't hitting right. Vary the volume, listen loud for a bit, listen soft for a bit, don't burn out your hearing trying to get the chest-thump on the kick. Of course being really efficient and organized in general helps in these situations.
Here is how our workflow here operates. When a song comes in, I deal with the files. My job at that point is to make sure that technically it is ready to mix. First, do the files sent match the rough mix? Same production? These files are saved on a drive called “Original Files” Next, “Save Copy In” and copy it all to you “Prepped Mixes” drive. NEVER work on the original files. You will, at some point, want to refer back to the original session or files untouched.
Then I will get the session organized in the way we like it here; drums at the top, percussion, bass, guitars, keys, synths, BGV, Adlibs, Lead Vocals. Everything put in order, labelled, and groups made to easily identify parts. I’ll also do any cleaning of clicks, pops, bad edits, noise.
I’ll simplify the session if necessary and if I am able to. Clean up the I/O Setup and get rid of all unused busses, inputs, outputs. Rename and move busses that interfere with our “effects template”.
Drums going to an aux bus with no plugins? I can copy the volume moves to the tracks and send it all out the master bus. Lead Vocals and BGV’s feeding the same Vocal bus? I will duplicate and create an aux for just the Leads and one for just the BGV’s. the entire goal here is to get something into Serban’s hands that is organized, cleaned, and easy to understand. If he hears a tamborine, he knows where in the session it can be found, it is labelled, and grouped with the Percussion.
He should then only need to worry about the artistic decisions of mixing and not any technical details that take him out of the mixing headspace.
If you don’t have someone to do this work for you; as I don’t when I’m mixing my own projects, you can approach it in the same way. Do the work in two steps. First act as the technical engineer and do all of the organizing and cleaning tasks. You might identify things to hit later in the mix, but don’t spend time mixing at this point. When you have the mix organized and technically ready, then copy the files to you “Mix Sessions” drive. Again, don’t work on your cleaned up session or files; you will want to refer back to them at some point.
Also back up everything regularly. Have multiple copies and backups. Stay organized that you are working on the current file, but have backups ready. Nothing kills momentum faster than “oops my drive crashed and I lost everything”.
We work with so many different producers, who are working in so many different ways, that we just need to take whatever they send. Sometimes it is a full ProTools session, complete as they left it. Sometimes people will commit or remove some or all plugins (removing just makes it harder for us). Maybe half the time it is not a ProTools session at all, but a "pile of files" multitrack. This can range from a producer's "stems" (we always ask for at least dry vocals), to some producers who send a couple of hundred files with every instrument and vocal track separate as well as printed effects for each vocal track (hopefully separated by folders such as BGV Wet, BGV Dry, Lead Wet, Lead Dry, BGV printed effects, Lead printed effects). That's when I put on my white lab coat and goggles and get to work!
We don't have any assistants here; it is a two-man operation. We have had assistants in the past, but as we don't do any recording and are mixing in the box, there was little for them to do. Ultimately they got frustrated, we understood and helped them to move forward elsewhere. There was just not much for them to do.
If we set up another room, spent another, let's say $75k for another full ProTools setup, all the plugins, etc; maybe they would be able to start to prep mixes and print stems. As businessmen, we have to weigh the pro's and con's of that and the cost/benefit. So far it would not be a good financial move to spend a bunch of money on equipment, set up payroll, etc. so that we can get stems out a bit faster. We can't mix any more quantity than we are now, so any additional prepping help would not be useful.
Yes! 100% in the box mixing. The 'bounce to disk' is what is sent to clients and to mastering. Never touches analog in the mix, no external summing, no console. Everything is routed out the A 1-2 Master Fader.
Now, on occasion, and this is getting very rare, there might be an insert on the master fader. That is what the Neve 33609/J has been used for. But before anyone goes out and buys one because we have it; I think that last time we used it on the insert was maybe 4 years ago!
We are also using the Avid HD I/O, and internal clock. Because there are no D/A and A/D conversions happening, aside from going to the monitors and external meter, we don't need any fancier digital converters or clocking.
[top]What are the measurements of your mixing room? -victoredmundo
My room is approximately 16' x 16'.
[top]Could you talk a bit more about what you did with the acoustic treatment? I mean, more in detail? You mentioned that you guys did it yourself. - Oroz
Ok sure. A bit of history; I worked for a while with the architecture firm Studio bau:TON in their TEC:ton division. I wasn't working with the architects directly on design, but was helping to put together equipment packages and overseeing installation of equipment. While there, I did absorb enough design knowledge to be dangerous.
My room is a 3rd floor attic room buildout space. Referring to the picture here; the back wall, not shown, is angled from full 8' height down to a 4' knee wall. My room also has an additional 8' x 8' "entry" area where the stairs enter.
On that back wall I've got some wooden dispersion and some absorptive cloth window blinds.
On the ceiling you can see we created a "cloud" This is an 8' square frame, hung from the ceiling at an angle. The top of the frame has 1" rigid fiberglass panels. the bottom side is faced with POAL (https://poal.net). This doubles the surface area of the fiberglass and any reflections have to go through it twice.
The front and side walls you can see fabric covered 2" rigid fiberglass panels. The corners have foam "corner killers". It is hard to see in the photo, but those panels are mounted on wooden frames that are angled off of the walls. Over the half room 8', each panel is angled out about 10" at the outside edge. Again this doubles the surface area of the rigid fiberglass as well as gives some non-square surfaces.
A couple extra corner killers in the back of the room, a big cloth covered futon behind me, carpeted floors, with a wooden floor cut in for chair rolling.
Because we are essentially only using the near field speakers, we were mostly going for absorption of reflections. We don't really have any isolation built-in except being in different rooms and floors, but we are not doing recording, so that isolation isn't a problem for us.
I've not isolated the equipment very well and do deal with some fan noise, just kind of gets blanked out at some point. If I need to hear detail over that noise, I'll go to headphones.
[top]I have a question about how you deal with Mac OS, Pro Tools and plugin updates. What OS and Pro Tools version do you guys use? - musicmixer04
We try to do updates as little as possible for OS and for ProTools. Currently on High Sierra and ProTools Ultimate 2019.6 Before that we were on 12.7 (we skipped PT 2018).
[top]Do you have some quick workarounds when it comes to Pro Tools, or OS upgrades. Do you just upgrade the OS and copy the old plugin folder to the new PT version, and upgrade plugins when they come along in the session? Or do you use some days to do all the upgrades so everything is working perfectly without having issues that would ruin the creative flow? - musicmixer04
When we do a major update like that, we'll clone the Disk, and then do a migrate upgrade. Always keeping one Bootable HD with the current working stable OS/PT configuration. Also will make a bootable backup of that stable configuration.
Then on the 3rd bootable disk, clone the stable configuration and migrate to the new Mac OS, or install the new ProTools, update all plugins, etc. When that disk is working perfectly and stable, it will be backed up to bootable back; one copy of the previous stable OS/PT is kept on a bootable drive.
[top]As you mentioned quite a few times here being "Old School" in your approach, do you use a lot of UAD plugins yourself to get familiar sounds and GUIs from your analog days in the computer? Or do you have them just because you have to be able to open them all up when a session comes your way? How many UAD Cores/DSP do you have in your system? - bobbl
UAD-2 Satellite Octo
[top]I am wondering how you create realistic modern vocal "doubles", or stereo width on vocals when perhaps you only have one track to begin with? - CanadaSC1
To sound good and convincing as actual doubles, it should be done in production and recorded that way by the singer or BG vocalist. If I want to create some computerized synthy doubles and harmonies, I like the Isotope VocalSynth (check out the bridge on TxT "9 and Three Quarters")
[top]If you have a static mix which sounds ok but a bit boring. What are some of your moves to get the song exciting? - mattias78
To be honest, most of the productions we receive are very well done and these kinds of decisions are mostly made by the producer.
Now, I can take from what they are doing and offer some suggestions, as well as what I've done in this area where needed.
We talked already about song arrangement; save doubled and tripled parts, or save harmony parts for second and third chorus. Don't be afraid to mute parts earlier in the song so that you can bring them in later. (does that cowbell need to be in every chorus, or just in the last one!)
The song should be arranged like a triangle, add more parts as the song goes.
Your automation within the chorus can also follow the triangle model. Ramp up effects so they build through the chorus.
I would definitely be doing automation to make each chorus a bit bigger than the one before. This can be effects, volume rides. Also a lot of productions are using the EDM pumping kind of chorus these days to add excitement. You could try just a bit of this if it fits the genre.
You can do things like the drums in the chorus can hit the compressor harder than in verses, so they don't really go up in volume, but they crush and densify a bit more. Same can go for the whole song in the chorus on the master fader.
[top]Do you work at whatever format the files come in or always work at the sample rate you prefer? Also if you find that printing in 32bit gives better transient detail. - Prog
I'll convert sessions in the following circumstances;
- 44.1kHz, 16 bit will be converted to 44.1kHz 24bit (better resolution in the box)
- 88.2kHz will be converted to 44.1kHz (better use of computer resources)
- 96kHz will be converted to 48kHz (better use of computer resources)
- 32 bit 48kHz left as is.
There are times we'll mix in 96kHz if files come in that way; small sessions, more acoustic style music, songs with a lot of openness and space.
[top]Hey John, just wondering if there's any things you always do differently for eg a song with a hip-hop lead production (808s etc) against a four-on-the-floor pop-dance kinda track? - GROVER
No different work flow or strategies I think. These things kind of become rote after a while, so I really have to question myself "am I doing anything different"?
I guess one thing would be not worrying so much about clean production and vocals. I think that part of the production and sound of the genre is some built-in dirt and grunge. Vocals can be mumbly and some noises can be left in; that becomes part of the performance. I'm not going to be cleaning up a lot of headphone bleed, mouth noises, etc on a Hip-Hop track.
Waves Smack Attack
[top]Hi John, a short question - do you make much use of Transient Designer on the rhythm section or other instrumentation? - TLS
We'll use these on occasion. UAD's SPL Transient Designer comes to mind, I've played around with the Waves Smack Attack which feels pretty powerful.
[top]How do you see the mixing/recording/production aspect of the music in the next 3 or 4 decades? - Young.baws
I think that we'll continue down the path of the shrinking of the recording studio business, the home studio will continue to grow and become more relevant to more and bigger artists and producers.
I think that probably traditional genre categories will continue to be shaken up; there is so much world music influence now, and cross-genre productions and artist experimentations. Globalization I guess is the word.
[top]I'm wondering how your mixing has changed over time - do you do things or think about things in a certain way now that you may not have 10 years ago? - jah279
Mostly everything has gotten bigger and more complicated. I look at sessions from 15 to 20 years ago; say Musiq Soulchild, and we're talking about sessions with 24 tracks of music, maybe 8 tracks of vocals.
The power that is in the hands of today's artists, songwriters, and producers is pretty immense.
[top]What sampling rates are used in productions like Max Martin and other big guys? - Blue
We get a variety. Many people still working at 44.1kHz (and some files received even 16bit!!! ) Large variety of files are 48kHz, 24bit. Some come in 32 bit Float. This would be most of the big guys. Occasionally even 88.2kHz and 96kHz.
Why? Because most of the music we're working on doesn't benefit from a high sample rate and it just makes everything bigger and session resources smaller. I would prefer that everything we receive be 48kHz, 24bit (or 32bit float). This is also one standard for Atmos mixing, so I don't need to do file conversions again.
For us it has been a gradual change, from console and tape to small DAW productions, to now huge DAW productions. The rough mix has now become good enough that people are falling in love with it a bit too much perhaps.
I can't really think of any techniques specifically right now. I'll have to think more about this question and see if I can come up with a blog.
[top]Could you tell us which Sample Rate Conversion you are using? - Coldvodka
We are not converting sessions to 44.1kHz. We are bouncing out final mix passes at 44.1kHz, 24bit.
Again, the reasoning is that since the mastering engineer is going to be sending 44.1kHz files out, we want to be the ones to make and control that conversion at the mix stage.
[top]Can you tell us what converters and hardware you are using for that process ? - RightOnRome
No hardware, this is done by ProTools during the bounce to disk. Tweakhead setting.
[top]Do you use multiple master faders (masters for «final» aux busses) for an ultimate control over ProTools headroom/gain stage? Or do you clip the files down prior to mixing? Or are you doing both? By the way, are you planning to do a Mix With the Masters one day? - Young.baws
We don't use that multiple master faders technique here. I'm not sure what you mean by "clip files down prior to mixing". Generally we are adjusting volumes as we mix into the final Master Chain so that the effects of it are in control throughout the mixing process.
We're hard in the Yellow on the master meter at the end of the mix; but if your techniques are working, don't change.
Sorry, no plans for myself or Serban to do a MWTM.
Focusrite RedNet HD32R
[top]Hello John, wanted to know what version of Pro Tools do you guys work off off? Is it HDX or just Native. Also, are you guys Mac or PC? If Mac, what version do you both have. - M2E
Mac trashcan tower, High Sierra OS, ProTools Ultimate 2019 currently. Two HDX cards, Avid HD I\/O. (I've also got Focusrite Red 16Line and a Focusrite RedNet HD32R I/O's for Dante In/Out for Atmos mixes).
[top]How long do you typically spend on a mix? Is it a case of doing the mix in a single session and then deal with client revisions as necessary, or do you tend to do a 'first pass' and then revisit it on another day (perhaps multiple days) before sending it to the client? I guess I'm wondering if you are more of the 'go with your initial gut instinct' philosophy, or more of the 'gradually tweak to perfection' philosophy? - J-S-Q
We will generally do the bulk of the mixing on Day 1. Then leave it overnight, and revisit for ourselves on Day 2. If it sounds like it is a good place then, it gets sent to the client, if it needs more work, it might be worked on on Day 2 as well, and then sent out after a fresh listen on Day 3. It is always good to check with fresh morning ears before sending out if you are unsure.
Their revisions can be spread over multiple days. The hard part here is that once you send to the client to listen, you can't touch it again even if you start hearing something that is bothering you, because once you've sent off, if you change it, you can throw off the whole process as the client will start to hear differences between revisions that they didn't ask for.
[top]I wanted to know how do you keep your ears healthy by working so much? How do you take care of a healthy back while leading a sedentary lifestyle? - Postykish
Touched on elsewhere, but I really try not to listen loud or for long periods of time. Frequent times during the work day while I'm not playing music at all. We don't have clients attending mix sessions, so I don't need to turn up the mains to impress anyone or spend time while someone else is listening.
Outside of work, I'm not going to concerts often, earplugs when I do, I'm using hearing protection with machinery, and I don't drive with the windows down (seriously!)
I've got a good chair for my back (currently All33 Backstrong), I try to stand up often, generally stay healthy, fit. It is an ongoing issue.
[top]I notice you use a lot of 1/4 note delays on vocals. I was wondering if you are using any ducking on the delays to keep them out of the vocal or running them into a reverb? - LiamMcCluskey
As a lot of sessions come from producers with their effects on them, I'll have to answer pretty generally because there are a lot of different people doing different things.
Where we are adding our own preferred / standard delay it is not ducked and not fed into a reverb as a general rule. Specific situations might call for that, but it is not part of the base preset. Mostly we'll be playing with the amount sent to the delay, and then muting or riding volume (either delay send or return) if we don't want delay in certain places.
Where some producers do like ducking, the Air Dynamic Delay is popular, as is just plain riding volume on the delay return track, and some are automating parameters within the delay plugin.
Some do add verb and a variety of other things to the delays.
[top]I'm wondering what you've noticed in recent years when you receive projects, and how you adapt to trends. How technical are you and Serban when analyzing how a mix compares to other popular mixes right now? When you're doing a rock mix like the alternate mix of Interpol's "If You Really Love Nothing," how do you adapt to that? Is it the same as any other mix, or do you have to rethink your approach? - porgporgporg
I'm a really bad Music Historian, I have not consciously studied the trends, so going off the top of my head, here are my opinions. First let me say that we might be more setting mixing trends than following them; so we’re not really doing analysis and comparison to other mixes right now. We are not pulling up frequency spectrums, or making mix decisions based on technical analysis.What I’ve noticed over the years is that the production quality is both improving and turning to ****.
Many of the really experienced (old school?) producers, and the people that they are mentoring, educating, and training are really good. Their attention to detail is exacting, specific, and well thought out. Their technical chops are generally excellent and there is not much wasted production space.What do I mean by that? I’m defining wasted production space as using three plugins to do the work of one. Boosting 1kHz on one plugin and cutting 1kHz on the next. Nested routing decisions of no use (Kick Drum aux to Drum Aux to Beat Aux to Music Aux to Backing Track Aux to Master Bus) with plugins everywhere making it nearly impossible to breakdown a song.On the other hand, many of the self-trained Producers, “Bedroom Producers” (another upcoming blog topic, and I love you and think that you are the future of the business!) seem to be making it up as they go, may have quite poor technical chops, and have a lot of wasted production space and unnecessary complexity. This is kind of one reason that I’ve come to do this Q&A and the blog. I think that the self-trained Producers and “Bedroom Producers” are creating some really amazing, experimental, and leading-edge sounds; I hope that the influence of us “Old School” engineers and producers can help to encourage the experimentation while educating on easier work flows, compatibility and file delivery, and career longevity advice. As far as do an alternate or rock mix such as the Interpol song; I think that we approach it as any other mix. Serban is being sought out to bring his sound to it.
[top]Do you find yourself taming mid range, or boosting spots to add excitement more? Do you spend a lot of time looking for masking happening between tracks in the mid range, or trying to fit more mids in rather than take away? - Aidanthillmann
I think you've analyzed this as well as I can. It is another thing that becomes habit and rote, but one thing that I think that we are definitely looking for is vocal clarity and overall clarity which has a lot to do with midrange, frequency masking, etc.
So all of the above.
[top]Do you have any tips to get emotionally connected to a song? I really really like your mix and Serban Ghenea’s mixes on those records: Callum Scott - What I Miss Most
Katty Perry - Chained to the Rhythm - Young.baws
This is actually a topic for one of the blogs that I've written, which will be published here.
I'm not sure what the publication schedule is going to be, so I'll give you a preview.
If the song doesn't hit you right away, find something in it that you like. Maybe the percussion patterns are cool, maybe the bassline makes your head bounce. Find something and work out from there.
[top]Regarding pop music with electronic programmed drums, do you tend to send all drums to a drum bus for some final compression/glue between elements or do you leave kick, 808's etc... free and send them straight to the mix bus after some individual tweakings on the track? - jakelorenz
If a session does not come to us already subgrouped like this (about 50% probably) then I generally won't add a drum bus for programmed drums.
If it is more like a broken-out loop that has lots of interplay between the tracks, then it might be more useful to do.
I do like a drum bus for Live drum kits because there is a lot more interplay between the recorded tracks that you want to have global control over.
[top]I wondered whether you consider the cohesiveness of the tracks as a whole whilst mixing an album or whether you're just concerned about the individual songs/singles and leave mastering to sort out the cohesiveness? - Ollieneedham
If we're mixing the whole album, then definitely we are thinking of a whole album feel. This is especially important if songs on the album are from a variety of producers.
It isn't necessarily that we are mixing differently than for a single, I think each song gets worked to the same standard, but we might be thinking if overall we are filling the same frequency spectrum, do vocals sound complementary song to song. I would definitely be referencing prior mixes as I go along.
[top]What role does the rough play in your mix process? Are you regularly a/b'ing? If so, are there particular things you're paying attention to/comparing as you go? Or, is it more of a gut check? Should I be spending less time with the rough and trusting my instincts more? - Libertine
I am constantly A/B’ing the rough mix. I put the rough mix on output A 3-4 which goes to a separate input on my Studio Comm monitor controller. One nice thing about this is that because both A 1-2 (My Mix) and A 3-4 (rough mix) are going to it analog, there is no delay when switching between them. So I’ll line it up pretty much sample accurate and can flip back and forth and hear clearly what the differences are.
With really good rough mixes that we are getting these days, the first step is to make sure that your mix sounds THE SAME as the rough mix. You have to start from where they left off.
Then look for areas that you can improve. Take apart their Master Fader chain if you are using their ProTools session and see what they are doing. Half the time just turning off or replacing bad “self mastering” products with your own will be an improvement.
Work on the technical corrections, make the vocals present and understandable. If I’m not sure if what I’m doing is helping; I’ll do a quick blind A/B test. Eyes closed, flip back and forth between the rough and your mix quickly enough times that you don’t know which you are landing on, and then ask yourself which one you like better and why. Hopefully it is your mix. If not, figure out what it is that is different and why you like it better.
These days you can’t ignore the rough mix. Spend more time with it and really analyse what you like, and what you don’t like and why.
I also sometimes feel a bit of that imposter syndrome. Sending off a first pass is still nerve-wracking. If notes are coming back positive, you’re doing fine!
[top]I am curious about how you achieve the sound of your vocals which have that trademark consonant "pop", tonal clarity and forwardness, width, and effect splash excitement. How much of the vocal effect splash excitement is something you're added and how much has been already provided by the producer? - TLS
I’ll have to tread lightly and generally here. There isn’t a secret or a special plugin to achieve the “Serban Vocal Sound”. It is all about his experience, taste, and skills.
Remember we are getting vocals from 100 (I didn’t count, I’m sure it is more) different producers, recorded by 100 different engineers, in 100 different studios, on 100 different microphones, 100 different mic-pres. There is no way to plug any standard anything into that and get a specific sound.
Don’t really do much parallel track mixing; again we’re “Old School” operatives. General effects sends are pretty standard. A bit of chorus, a bit of delay, a choice of reverb(s).
If a producer has sent us some specific effects, we’ll use / modify / work with it. If they don’t we’ll create it. It is maybe a 50/50 split. Remember that some of these producers could and have mixed and released on their own (Max Martin, Greg Wells, etc.) They are coming to Serban to make that final 5% to 10% or so polish. We’re not going to toss out what they have crafted and sent to us, but we will mold it to fit a little better if necessary.
[top]Do you use codec auditioning tools such as Sonnox Pro Codec, or something like that from Nugen or iZotope and mix through it to make sure that the mix will translate well on lossy codecs? - Coldvodka
No, not at all here. The mastering engineer should be the one to handle this. We have had issues with loud mixes not playing back at proper loudness on Spotify in the past. Long story short, after consulting with Mastering Engineers and Spotify's technical team about their codecs and processes, we determined that Mastering needs to supply a file to Spotify that conforms to essentially the Apple MFit standard. We never try to compensate for anything in the playback chain; a good mix should sound good on every system and medium. If it does not, the problem is in the end media and that should be corrected at their end.
[top]I'm wondering what the process is like for mixing songs that were produced in another DAW. If I were to hire you or Serban to mix a song I made using something other than Pro Tools and both parties agreed it already sounded good and simply needed that last 10% would it then be on me to print everything and deliver the song as audio? - TeleMA50c
Yes, when receiving projects not in ProTools, we'll ask for "consolidated" or "committed" audio files. Prefer it to be track by track for instruments (i.e. not a percussion stem, but all 10 percussion tracks). Blended BGV parts are usually ok (i.e. stereo doubles, stereo harmony 1, stereo harmony 2). Effects can be left on instrument tracks as you see fit.
Lead vocals should be "committed" with processing, but not effects, (i.e. Leave on your Compressor, EQ, but not your reverb and delay).If possible we prefer to have any effects on vocal printed separately (i.e. Lead vocal chamber reverb, lead vocal 1/4 delay, BGV 1 chamber reverb, BGV 1 1/4 delay, etc.) We might also ask for non-processed Lead vocals in case we want to undo or re-create something a bit differently; it depends on the producers and how married they are to the exact sound of their rough mix.So it can be a lot of printing parts for you to deliver to us.Producer Jesse Shatkin (Sia, Kelly Clarkson, Paloma Faith, etc.) works this way and his engineer Sam Dent sends us very detailed and very organized files.
Instruments folder, VoxWet folder, VoxProcessed folder, VocDry folder, and VocFX folder. He'll also do screenshots of certain plugins to put on some tracks. We can then pick which of the vocals to use. Usually the Processed vocals combined with the Vox FX, but sometimes better to go with the VoxWet for some parts.
Other times we'll receive parts that are closer to stems. In these cases we at least want to get Dry Lead Vocals and separate effects if possible. There have even been times we've received a stereo music track and three or four vocal tracks; in these cases we have on occasion turned down the project as not mixable. It is not our preference to try to "stem master" someone else's work.
[top]When you get several tracks of the same thing, such as 4 different drum rooms, or a snare consisting of 3 different microphones, do you treat them separately, group them together or even both? - Spankjam
I would definitely group snare mics or bass parts of the same performance. This way when I get a good blend that I like, I can do volume moves on the whole blend as is.
I'll also sometimes send those tracks to an Aux input as a subgroup if I like the blend but want to do an overall EQ or Compression insert or other overall processing, or an effect send to reverb for the whole blend.
I'll also group the same vocal parts (same notes) agains so I can get a blend I like and then ride the whole blend. Also I'll send vocal parts to an Aux Input again so I can do overall EQ or compression on the blend, or send the blend to an effect.
[top]Would you still treat the sources individually but have them connected on the fader? Like VCA? - Spankjam
So by Group I'm talking about the ProTools Mix/Edit groups that allow you to solo, mute, and do volume controls (and other functions) across the tracks in the group.
These days, with such high track counts available, I don't do any group consolidations, sub mix bounces, or such things. This allows fine control of all the parts as needed as well as avoids having to undo such a sub-mix when a replacement part comes in. (producer sends a new snare to replace one track, etc).
[top]What is the song that you mixed that you are most proud of? If you have a website or instagram page where I can look at your discography, could you let me know please? - Coldvodka
I personally don't have a big presence on Social Media. No Instagram or website. My credits are here on Jaxsta.
But due to the way they are often copywritten at the label (often despite specific language that we give them), it might look like I'm actually mixing a lot of projects that are actually mixed by Serban Ghenea. (ie. Uptown Funk should be Mixed by Serban Ghenea and Engineer, or Mix Engineer John Hanes). I've highlighted some of my mixes at the top of that page; (Mixed by John Hanes).
In addition, I'm not credited at all on some things that I've mixed. Dotan Harpenau's "No Words", "Numb", and "Letting Go" for example. Proper credit and proper Credits are an ongoing issue.
All that said; I felt really good about the Dotan mixes. I'm really proud of the mixing on the upcoming release of Morgan Saint's EP where I mixed 6 songs. The TxT "9 and Three Quarters" mix was really challenging and I think it came out great.
Metric Halo ChannelStrip
[top]Do you use the MH CS3 limiter for anything, or would you use a dedicated plugin? - b0se
copying my answer from another. I'd say that I don't tend to use a specific limiter on individual tracks often, though I might push a compressor into a fast attack mode when needed. Most of the time the Metric Halo ChannelStrip is in the MIO mode and mostly not using that or any other limiter.
[top]I would like to know your opinion on getting the feel/emotions right and moving in the right direction with the mix. Do you call the client and ask for directions? Or is the rough mix enough information? What do you pay attention to? Also how many revisions do you guys usually do until the client is happy? - Tommiiii
I think mostly it is the rough mix and familiarity with various producer's overall style (and quirks) from past experience with them.With producers new to us, we'll sometimes talk a bit, get some pointers on direction, but often I think they've picked Serban specifically for his interpretation and input on that. I like to try to glean as much as I can out of the rough mix, figure out what they are shooting for by feel and what feels good for the song and production as I unravel it. Number of revisions can be anything, whatever it takes. From first pass approved to upwards of 15 or 16. Try to never leave a client unsatisfied, but also try to let them know when their requests are not improving the mix any longer.
[top]Do you think it's the wrong mindset to have limits or rules on revisions? - Andersmv
I think that at this high level of the business we have to have a bit more openness to achieving the artist and producers vision of perfection than most.
It helps that we are working remotely. If you’ve got an artist or band sitting in on the mix that can take a whole day to mess with things on one song. We have the ability to slow the process down a bit by taking our time, sending out the new pass, and then waiting maybe another day for approval or more notes. I think that it removes some of the impulsive requests and makes them more thoughtful.
I think you do need to balance your time and set boundaries where you can.
I would say that in general, and I don’t want to sound elitist here, that our clients are highly experienced and educated in the processes involved.
Part of that education and experience would come from having those boundaries set on them earlier in their careers. It definitely helps to get consolidated and agreement on notes. It helps to have a strong producer who will oversee the direction and be the voice of reason in disagreements.
Through all of that we try to do everything asked but also will say that “this isn’t getting better” when needed.
Sometimes it comes down to trying what the artist wants because they just want to know they’ve exhausted every avenue they can think of. We’re not often going to say “no” to Beck, Bruno Mars, Ariana Grande, or Taylor Swift. When we do say “no” they understand and trust that opinion.
So overall I do think you should work on that balance. Make the clients feel that you are open to all of their ideas, while finding a way to not get run over. Sometimes we do get run over too, and feel taken advantage of with excessively "needy" clients. In the end a happy client is your best advertising.
[top]You're obviously working with a whole different level of people (on a daily basis) than most of us. Did that kind of growth naturally happen for you, or do you remember taking steps to (and this is definitely going to come off as elitist..) "weed" out a lot of bands? - Andersmv
I'd say that it is a natural growth cycle. However, some of my favorite projects that we've worked on are new artists, first records, and new styles. We don't "weed out" a lot; in fact our managers are often pushing to do more new acts, more trendsetters.Again going back to Music Soulchild and Jill Scott, and some work with The Roots; it was a lot of fun to be in on the beginning of that resurgent Philly Soul sound.My first Grammy win for LaRoux is another one. Not a polished act or technically great recordings, but it had that vibe and feel. My recent mixing of soon to be released Morgan Saint is again another new act that is a lot of fun to help to craft a sound and vibe for. I personally find it more satisfying to be in at the beginning of a new artist's career, a new sound and vibe, and to be a part of crafting that.We are definitely not just catering to established, well financed, and well promoted artists.
[top]How many tracks are in Taylor Swift's "Daylight" and "Death by a thousand Cuts"?
Daylight doesn't sound too busy but Thousand Cuts does sound like there quite a bit going on. - JanetB
Ok, I'm pulling up my archived tracks for this one. "Daylight" has 109 tracks, not including Aux effects returns. Now that includes a lot of raw stacks for BGV's (the "daylight" verby bgv hit is 16 tracks). Jack Antonoff and his engineer Laura Sisk typically send us a PT session with vocals separated and not bounced together. There are 4 kick tracks, 4 snare tracks, 7 guitars, 27 BGV tracks not including that verby bgv hit. Lead vocals are spread over about 5 tracks (separate tracks for V1, Pre, Chorus, V2, Bridge, etc.). "Death By a Thousand Cuts" is 85 tracks not counting Aux effects returns. There are 24 tracks of music, the rest vocals. Again the Ld Vocal is spread over 5 tracks, there are 10 tracks in the Aahs bgvs.
[top]What was the track count for Troye Sivan's Bloom, my favourite mix of all time. Such a beautiful landscape! Must be a pleasure to mix a song like that. - NdK
Bloom is pretty typical size; about 35 tracks of music, 26 tracks of BGV parts. The producer, Oscar Holder, a master of his craft, created and printed for us most of the vocal effect tracks (computer voices, verby vocal hits) maybe 10 tracks of those things, Another 10 tracks or so of doubles, leads, and harmonies.
[top]How do you stay fresh and inspired? How do you avoid falling into periods that are low on creative energy? Thoughts appreciated on how you deal with the highs and lows of a creative profession. - gainreduction
First, we try to maintain regular hours and schedule. No point in getting burned out on one project to the detriment of another. Nights and weekends are primarily no-studio time zones.
Second; there are usually enough things to do that if I'm not feeling creative, I can concentrate on some technical issues or other tasks like making stems, running backups, or playing around with and learning about a new plugin.
Because we are choosing our own workflow without clients present, if we don't feel like working on Song or Artist X today, we can work on Song or Artist Y. Luckily, being in demand enough, we probably have a bit more leeway in getting a mix out the door. Also, almost any day in front of music is a good day. I love what I do and look forward to coming to work almost every day.
[top]I'm curious if you have a typical workflow when you're starting on a mix. - Kaleida
I'll usually start with the whole song playing, listening down and making bigger volume moves as I learn the song. When I've got a pretty good feel, I'll turn everything off and start unmuting in the order from top to bottom of how I arrange the tracks. Drums, Percussion, FX, Bass, Guitars, Keys, etc. . . . BGV, Lead Vocals.
I'll generally do this adding each instrument as I go, i.e. not turning off drums when I start opening up percussion.
Of course I'll solo certain things to hear and work on specific issues. When I've gone through everything like this and the whole song is open again, I'll then usually do my vocal editing; soloing vocals and cleaning them up.
Finally after that, I'll tend to leave everything on as I do the meat of the mixing; still solo'ing to nail down specific parts of course. I think that this method lets me really learn the parts while spending the most time hearing drums and hearing where vocals relate to everything, which are often the key elements needing attention.
I guess I stick to this pretty regularly; but if something else catches my attention while I'm in the process, I'll go in and put that work in on that part as needed.
DK-Technologies DK1 Meter
[top]Wondering how you are feeding signals to the DK Audio Meter from Pro Tools. - ryno1
I'm feeding it from an AES/EBU mult (through a Z-Systems Digital Detangler) of the A 1-2 Output.
My DK meter came out before LUFS were a thing, so I've found that the NBC scale is essentially showing the same as LUFS, but I'm not exactly sure what the scale is technically showing.
Yes, looking at -6 on the output after all master fader plugins; this doesn't mean trying to hit it for every song. This is just my mark to watch out if it's getting pretty loud.
[top]Do you like to use plugins that emulate consoles? - Oroz
Nothing used across all tracks for full console emulation. We will use vintage emulated EQ, Channel Strips on certain sounds for a specific effect or purpose.
I'm mostly familiar with Waves, Soundtoys, and UAD offerings in these areas. so many to choose from so it isn't really a choice of which sounds best! or most! but which one I remember when I'm looking for something.
[top]Could you share with us some info about the computer(s) you use (hardware, OS)? - Jetam
Currently using;2013 Trashcan Mac, High Sierra. I backup everything to a hard drive, and then the harddrive to an LTO tape.When hard drive backups are full, the hard drive is retired and stored. Nothing is ever deleted.
Current projects are on local disks and a secure cloud account. Projects are delivered via many types of FTP and Transfer services (i.e. WeTransfer, Dropbox).
When we have a stable OS and ProTools version; we'll not upgrade it for a year or more.
[top]Re: Dolby Atmos music mixing - As you mentioned in your recent Mix article, a current problem is that few people (including the labels) are able to approve the final ADM BWF mix. I am wondering if you are being tasked with embedding metadata into the Atmos "master" as the mix engineer, and as you sort of put it in the article, the de facto Atmos mastering engineer? - Bias Mike
Note: John Hanes participates heavily in this thread, sharing further, detail knowledge about Atmos and Spatial mixing
I've been going by the UMG deliverable requirements for Atmos. I'm not sure if this document is in the public domain, so I'm not going to share it here.
Basics of delivery are, which are pretty much across the board for UMG, Netflix, Apple, Amazon, and Tidal:
- Minimum 7.1.4 mix room.
- Near field mix
- Target loudness follows ITU-R BS.1770.4; -18LKFS +/- 2LU
- Peaks not to exceed -1dBFS True Peak
- Binaural render mode is set and metadata is included.
- File delivery is ADM/BWF export with BIMD (binaural meta data)
[top]I'd like to know if you dither your mix 24bit mix prior to sending it to mastering. - Young.baws
We are not dithering the mix. Typically sending a 24bit file to mastering.
[top]Hi John, do you receive sessions in 32 bit-floating? Do you think that working at 32 has its advantages? - Oroz
Yes, we do receive 32bit float. I'm sure it does have advantages, I must admit I have not carefully studied and listened for any differences.
[top]I would really appreciate it if you could share your technique/philosophy concerning phase. - Young.baws
Mostly I notice problems with phase when lots of drum samples are used. In these cases, I'll usually nudge tracks by 10's or 100's of samples to find a nice phase relationship rather than playing with a plugin.
[top]What are the techniques you've developed that help you know when a mix will translate well on all speakers? - MattyJoe
I think it all comes back to experience, and to knowing the speakers that we are using pretty intimately. We are not really changing our mixes to fit the environment of listeners.
Basics of good mixing will apply anywhere; the listeners are going to modify their opinion of their playback equipment, and not modify their opinion of the quality of a good mix.
In other words; if someone likes the way music sounds on Beats Headphones, they should like the way any good mix sounds on them because they are normalized to that sound quality.
[top]One thing that strikes me with records you've worked on is an incredible balance (or maybe intentional imbalance) between boldness and cleanliness/naturalness. I'm wondering if you can share any sort of philosophy here, or workflows between you the artists/producers/etc on finding that sweet spot? - RyanC
Because the majority of our work is Pop music, which is constantly changing by incorporating elements from more diverse and experimental genres, I would say that making that balance is inherent in Pop music mixing. The producers, artists, and engineers are taking those cues and incorporating them into their work and balancing or smoothing them out to fit a more mainstream vibe.
[top]Do you think it is useful to mix into a limiter from the very beginning of the process, especially when judging critical tracks (Vocals, Bass, Kick, Snare)? Do you like to control the peaks of these tracks with limiters before they hit your mix-bus processors? - Mix Dome
Yes I always mix into the master chain with limiting on from the beginning of the mix. If you try to put something on at the end of the process it will undo or change a bunch of the work that you already put in. Mix into the glue!
In this way, you can also control the amount of limiting going on as you work; mix getting too loud? Pull everything down a bit. You can do this during the mix because you'll be able to adjust for how it changes things as you mix. If you do it at the end of the mix, it can change sound of the mix. Some individual tracks will get limiters as well as necessary, or the volume of parts within the tracks can be ridden down as a sort of manual limiting.
I would say overall there is a lot more compression and volume rides being done than just hitting louder parts with a limiter. I think that when we were mixing on a console, back in the Teddy Riley days, it was different. We would do the mix and then put the Master Compressor on towards the end. But in those days mixes weren't so loud out of the console and we printed back to 1/2" tape so we needed to get things under control for that.
[top]Considering that loudness normalisation is becoming the norm in broadcast and streaming, can you imagine changing the way you use the limiter on the mix bus or you consider the limiter to be an irreplaceable "glue" tool and aesthetically beneficial for what you do? - Jetam
We create a mix that sounds good. If broadcast or streaming services down the line **** that up, then that is their problem; we're not going to try to guess what they are going to do and compensate for it. Funny story: We had one artist we mixed for who was traveling around the country on tour. In each city he would listen to his songs on the local radio. He called and complained that the mixes sounded different in different cities and wondered if we should create different mixes for different radio stations. We shut that **** down immediately!
[top]I read your mixes hit at about -6LUFS at the loudest part. Do you reach such a high pressure "simply" with your mix-bus chain into the limiter or do you raise the level after the limiter, too? - mix_dome
We don't do anything after the limiter.
[top]The biggest takeaway for me so far in this Q&A is that there is no magic, no special hardware, etc, it's just 2 normal guys with tons of experience who have developed (and keep refining) a process that meets the tastes of their clients. - bambamboom
Thanks, that is the gist of what I want to share! We are all dealing with the same issues.
[top]What about clipping on individual tracks or drum busses? Do you use that technique? If so, what plugin do you use? There are a couple of old interviews where Serban mentions the use of Lo-Fi, do you still use it? Could you share which tracks you used it the most on? - Oroz
Yes, depending on the song. Often that comes built into the session from the producers if that is the sound that they are going for.
We might also use some clipping as a sound design tool where it sounds right. Lo-Fi is still a great plugin for this; most often used a bit on cymbals, but for aggressive songs I'll try it on drums, bass, 808. I've also tried out the Kilohearts Distortion plugin which I like. Soundtoys Decapitator or Devil-Loc is also a frequent choice.
One thing that I have mixed recently that I've used these aggressive techniques is an artist named Morgan Saint, I just wrapped up her EP which will be coming soon I hope. We wanted a bit of a wild, nasty, aggressive sound on some parts; so I used a lot of these kinds of clipping and distortions to push it hard.
[top]You mentioned before that MixStar mixes hit at -6LUFS at loudest parts. How many dBs of limiting are being applied on those loud parts? Could you talk a little about your gain staging to be able to hit so loud with so much clarity? Also...totally understand not disclaiming the stereo bus...but any hints on the limiter choice at the end of the chain? L2? Pro-L? - Tiagoderrico
Generally speaking it isn't going to be a ton. For a standard mid tempo maybe 1.5 to 2dB on a compressor and another 1 to 2dB at the end of the chain in combined compression and limiting. For a really aggressive song, it could be 3dB on the compressor and 2dB down the chain of compression and limiting. if I'm approaching 3dB at the end of the chain, I'd really start looking at my gain structure. I'm paying attention to gain staging and gain structure through the whole mix, it isn't really something that you can easily fix at the end of the process.
[top]So if I’m reading this right you have two compressors on the mix, one on the mix buss and then one on the mastering buss? - JB872
There is no separate mastering bus. The Master Fader for ProTools output 1&2 has multiple things inserted on it.
[top]This means crushing very hard on all the tracks? - SPORT
Not necessarily. Crush where needed if appropriate if you like the crushed sound on a particular instrument, use volume rides, light to medium compression where needed, and overall use good gain structure strategies.
[top]I wanted to ask if you only use the limiter on the master bus or even a clipping plugin. - frums
No clipping plugin on the master bus, though the limiter can be pushed there as needed.
[top]I read your mixes hit at about -6LUFS at the loudest part. Do you reach such a high pressure "simply" with your mix-bus chain into the limiter or do you raise the level after the limiter, too? I know we are not supposed to but I really like to clip the limited mix a dB or two and then lightly limit it again to less than a dB to get the final "mastering" level. Do you think that processing after the limiter can help to hit hotter levels in a more transparent way? - mix_dome
We don't do anything after the limiter, so I have not studied the issue of hotter with transparency as you are using it. But if it works for you then do it!
[top]The biggest takeaway for me so far in this Q&A is that there is no magic, no special hardware, etc, it's just 2 normal guys with tons of experience who have developed (and keep refining) a process that meets the tastes of their clients. - Bambamboom
Thanks, that is the gist of what I want to share! We are all dealing with the same issues.
Soundtoys Decapitator
[top]What version of Distressor? Arousor, UAD, Slate? - jakelorenz
I corrected my mistake after you quoted me. To make sure I'm not giving false testimony, it should have been Soundtoys Decapitator.
[top]So if I’m reading this right you have two compressors on the mix, one on the mix buss and then one on the mastering buss? - JB872
There is no separate mastering bus. The Master Fader for ProTools output 1&2 has multiple things inserted on it.
[top]Very interesting. This mean crushing very hard on all the tracks? - SPORT
Not necessarily. Crush where needed if appropriate if you like the crushed sound on a particular instrument, use volume rides, light to medium compression where needed, and overall use good gain structure strategies.
[top]This is really fascinating to me. Can I ask, beyond compression and limiting are there any other processes on the mixbus they are reducing dynamic range? Are there any tips you could share about what contributes to shaping the dynamics of the mix so that you are getting such a high rms with only a dB or 2 of limiting? On the subject of loudness, do you go for such a high rms level just because you like the way it sounds that way, or are there other considerations (label/ artist expectations for example)? - Mr XY
The first part is kind of hard to answer; it is basically the essence of mixing. Controlling elements of the song with volume, compression, EQ, to manage the overall aesthetic and gain structure throughout the song. I'm not sure of the mathematics involved, but I think that a first compressor doing 2dB of compression and then another doing 2dB of compression and limiting would add up to more than one compressor or limiter doing 4dB of work.The second part is easier to answer; it is due to the Loudness Wars. A songwriter will make a demo loud because louder sounds more impressive, right? The artist and producer then have to be louder than that on their production mix. They send to us, and we need to be louder than that because they are A/B'ing the mix to their rough and will say "it doesn't hit as hard". If we send mix that has more dynamic range and is not pushed hard, they will have mastering do it after the fact, because when they play their song next to the current hit, it needs to be as loud.So our theory here is that if it is going to be pushed that loud anyway at some point, it is better to do it in the mix where we have total control over the effects of loudness than to leave it to be done later with possible adverse effects. As I said before, this is kind of a "pro-move" and not everyone needs to be doing this if it is adversely affecting your mixes. It is a difficult position to be in to have to make a mix as loud as the rough mix but fix all of the problems that their loudness creates. Sometimes the rough mixes come in much louder than our final mixes can safely get to and at that point we just have to say "that is not going to sound good as a final product".
When I'm doing the Atmos mixes, I'll typically be dropping the volume by about 10dB. The reason for this is that first, the companies streaming Atmox mixes have guidelines for delivery volume. Second, the Atmos processor at the playback end has some leveling compensation built into it so that it is doing some of the work of equalizing volume across songs.
The trick here is to try to maintain the "crush" and "crunch" elements of the mix if they exist as part of the desired sound with a significant change in overall level.
[top]I wanted to ask if you only use the limiter on the master bus or even a clipping plugin? - Frums
No clipping plugin on the master bus, though the limiter can be pushed there as needed.
[top]Do you have a delay template? (Quarter note, eighth note, slap, etc). - thepilgrimsdream
I have a template that I use as a starting place, and then will create special things for certain songs as I feel the need for them. I'm not going to give specific settings or plugins that we use; not my place to do that.
Basic template (generalized)
- 3 second, 30m space, like a cathedral
- 1.8 second, 20m space, like a wood room
- EMT-250
- Quarter delay
- Eighth delay
- Dotted Eighth delay
- Half delay
[top]I was wondering if you have any modulation or eq after the delays? - LiamMcCluskey
First in the general category I choose delay settings that have a bit of spread (groove in the ModDelay), and the feedback might have a heavy LPF to clear some room for the main vocal. Adding effects after the delay would be one of those sound design features that I would create as needed for a specific effect.
[top]Would you typically be getting a rough mix together dry first, and then adding reverbs and effects later? Or is it a vibe thing where you’re dealing with spatial fx, balances and things all at the same time? - Animesh Raval
Definitely adding effects as the mix comes together. I would never mix with dry tracks; basic effects would go on first and then revised or specialized as the mix goes along; especially on vocals.
Same with drums, instruments, etc. Adding effects as I'm working on those sections, and revising the effects as I see fit as the mix goes on.
[top]Do you have any tips for getting the various spatial effects on the different elements all gelling and working together rather than masking or mushing things up? - Animesh Raval
I don’t think it is a conscious thought or plan These days. As I said I’ve got three standard reverbs set that I’ll send things to, so anything going there would be in that same space. After that is just making choices to what seems to sound good together.I’ll find a really clean reverb such as Altiverb if the situation calls for it, and other times a gritty machine like Rmx-16.
[top]What has been the most useful investment for your career so far? - Young.baws
No tips on that sorry. We also seem to be constantly buying plugins and upgrading others.I think the best thing you can do is to make your computer hardware last as long as possible. Get a stable configuration and hang on as long as possible.We don't actually update the main ProTools system very often. Still working on 2013 trashcan Mac. I think we’re still running High Sierra OS. Maybe next year we’ll try the new tower Mac and new OS.
[top]Which processor are you running on that mac pro? the quad, six core or eight core? - JanetB
8 core, 64Gb ram
[top]A) Do you send mixes to clients to review and they send notes for rounds of revisions or B) Do you use some sort of online mix streaming platform to play mixes for clients in real time to review? Also wondering if there is any sort of distortion on vocals to have them cut through the mix And pump. Your vocal tracks always sound exciting with movement. - ekedmo
Yes, lots of back and forth notes and revisions. We'll keep working on the revisions as long as they have notes. Sometimes it gets to the point where we say it is going in the wrong direction and enough is enough.Yes, using streaming for real-time review. We're actually using a product that is obsolete and no longer supported for this, so no sense in me naming it I think. It is preventing us from upgrading the OS on the computer that is running it, so we'll be looking for another solution.Mostly nothing specifically on the vocals for a bit of distortion or grit; the Waves CLA-76 can add a bit, overall Mix Buss chain decisions can add a bit. Sometimes if distortion is specifically called for Lo-Fi is great for that.What brings clients like Max Martin back to you again and again?
[top]What is it about your approach and philosophy towards mixing that brings clients like Max Martin back to you again and again? - MattyJoe
First, of course, is giving them good mixes, so let's move beyond that. I believe that there is also a comfort level for them that a second set of ears that they trust has double checked their work.I think a big part of this is that I approach this as a service industry. Each client gets one-on-one attention. If they have an important deadline, we'll make room for it.I do everything I can to provide ongoing support. If an artist has a show coming up on SNL and needs a special version of the mix by Friday, I'll make it happen. Need a radio edit for Italy? Done. Need a TV track with the Lead vocals only in the chorus, but down 10dB, Adlibs removed except for that one that goes "ahhhhh" at 3:15? Done, you'll have it in 15 minutes.Need stems for a mix we did 10 years ago? If I've got them printed already, you'll have it in an hour. If I didn't print stems back then, give me a day.The mix is done, but now the artist is changing the lyrics and sending a completely new set of vocals? Sigh! ok, send it over. What your deadline is tomorrow! I'll get right on it.
There are a lot of producers out there who can make a mix good enough to "release". If you heard their demo mix, you would say "that is a great mix!. Why do they then go to someone like Serban?" It is true that sometimes Serban might do little things to the mix. Over 100 tracks, lots of little things add up. Adding that final 10%, 5%, or even 1% of polish is what they are after.
They could probably spend the time and effort and do that themselves as well. But is their time better spent writing and producing a new song, or just taking a well deserved break; or should they be sitting in the studio agonizing over the perfect reverb decay, the last bit of noise on the vocal tracks, or the last .5dB of balance between the kick and bass.
When these very busy and in demand people find someone that they trust who can do what they want as well as, or better than they can do, they have found a valuable ally. When they can trust someone to take the burden of finishing the mix; dealing with pages of notes from producers, artists, A&R, management, and sometimes the artist's boyfriend or girlfriend (for real!) that is a major resource.
Sometimes barely touching something, but still making it sound 2% better is the greatest of skills. I would say that much of the mixing that Serban does is among the hardest that there is to do; because it comes in so well done, you have to find a way to make improvements, not lose the thread, and satisfy lots of people with sometimes very different opinions.
Mixing isn't about just pushing around EQ's and dynamics, it is dealing with the opinions involved that is the hardest part.
[top]What sort of hours and work schedule do you guys keep to stay so consistently productive and balanced? - stealthbalance
Yes, generally we try to stick to M-F, 9am or 10am to 6pm. We like to have some time for normal life stuff and because we don't have clients visiting it is much easier to be in control of our own schedule. Certainly there are times when we need to come in on weekends, or come back in the evening after dinner and family time to catch up or knock something out. Yes, lots of ear breaks. Also we aren't necessarily working on one song all day long. Might pull up and work on 5 or 8 songs throughout the day. Some might be starting a new mix, some might be in the critical listening nail it down phase, and others might be little tweaks for producer's notes and comments.It can be a lot of pressure, I definitely try to eat healthy, workout (my gym opening here this week!) I try to listen to my Apple Watch when it tells me to stand up.
[top]So I think what you might be suggesting is 18 hours a days , 7 days a week may not be the best way to roll ? - stealthbalance
Yes, did that when I was younger and luckily survived.
[top]As for working on producer notes, you said that you keep working on a mix as long as there are notes. I can imagine mix notes are a factor that's difficult to plan in advance in a boxed timeframe within the busy schedule you guys must have, especially if you want to stick to working hours and have a good balance. How do you plan mixes while keeping other projects running? - NdK
Can't plan for how long a mix will take, but if there is a deadline, it is not ours to enforce. Can't really base it on the demo either; you might get an amazing demo mix, nearly complete, and then they'll start adding this, redoing that, sending new vocals. Every project and every song is a unique beast.Generally we are sending out a mix for review and will get emails or texts back over the next day or two. We'll be able to pull up that mix, knock out revisions, and send a new pass out. Let's say that part of the process is a half hour. Now it's another day for revision notes to come in again; so that song is put away for the day and we'll do notes for another.Some clients prefer a Facetime video chat with mix streaming. These will be scheduled as they are available. Some take 5 minutes, some might take a couple of hours of try this, try that, what if we . . . How about . . . .We somehow find a way to get each client the time that they need; it is just a balancing act. While waiting for a video chat session, can knock out a couple of little tweaks on something else.On occasion you get a v1 mix accepted as the final release. You either feel great! or suspicious that they're going to come around a week later and have notes. Revisions can be anywhere from v1 - v3, some going up to V15. These can be for major reasons, or there might be 7 passes of nudging various adlibs up a little here and there.
[top]Do you have a system for referencing material during the mixing process or do you guys just mix until it feels right? Are you listening to commercial releases A/Bing them in Pro Tools or any other similar practice for balance and brightness etc? - Glamdring
Some of both. Serban mixes until it feels right, of course with suggestions and notes from artists and producers. When I'm mixing on my own, I'll sometimes reference similar genres or artist suggested material. Of course I have a wealth of pre-mastered files to choose from.
If I'm doing something new to me like a metal mix, I'll find a good representation of that genre. I don't try to sound exactly like it, but I'll try to get in the ballpark and then use my own judgement and feel. I'll just be A/Bing and working overall to mix to that standard. Nothing special like automatic EQ matching or 'one click' mastering that is now available in Ozone for example. I don't use those types of tools.
[top]Have you ever done a metal mix? Can we listen to it? - bobbl
Yes I did one as a spec mix! I did a mix of Saint Ansonia Feat. Sully Erna "The Hunted". Here's how it went down from my perspective. My manager got the word that they wanted to test out about 6 mixers. One test mix of "Hunted" to see who would get the whole album project.I got the files, did the mix. This was a one-pass deal, no notes or back and forth from artist or producer. They wanted to evaluate each mix on its own with no feedback or input. The story that I got (and who really knows, maybe they told each mixer the same thing) is that in separate completely blind judging, the management, label, and (I think) producer all picked my mix.
When it came down to the final say, the band wasn't really comfortable sending off the songs to be mixed remotely and wanted to be hands-on in the mixing. Unfortunately I can't release my mix for you. I've had feedback from a very metal friend that he liked it better than the final release; though they did a bunch of additional production kinds of decisions in the final mix, he had the benefit of band and producer feedback and input. On listening to the final release, I can say the my mix doesn't necessarily conform to the general treatment of the genre.
I would say that my mix sounds much more like what it would be like if they were playing down the song in front of you without a lot of post production. Not like a live show mix, but let's say an authentic performance capture.I really enjoyed doing this mix, and I think that if a Metal artist ever wants to find a bit of crossover audience, my mix would stand up for that.
[top]Can you please talk to us about monitoring? - diegua
Monitor path - Avid HD I/O analog out to Studio Comm 69A monitor controller. ProAc Studio 100 powered by Bryston 4BSST, no sub. Quested VS3208b and 15” Quested subwoofer are self powered.
Headphones PSB M4U1 through Oppo HA-1 headphone amp. AES/EBU from Avid I/O to Oppo. I chose the PSB because they sound kind of close to the ProAc. Mixing on ProAc 91.5% of the time, headphone mixing 8%, Quested .5%.
I would like to re-work my MidField setup at some point so I would use it more. It is not tuned. I kind of inherited the Quested’s from a friend who needed to raise some cash way back when.
I’ve not blown too many ProAc woofers recently; I’m the only one who controls the volume now and I know to start soft and clean up extreme low end before getting loud.
Listening levels: about 85dB for critical listening. 95dB is loud listening for me to feel it, 65-70dB is softer listening for longer periods of time for basic levels, learning the song, editing, printing parts. Almost never doing any mono listening, but I am looking (correlation meter) for and listening for phase issues.
Headphones for bass checks, listening for noise, another perspective. We designed and installed the room treatments ourselves. We have not had them tested and calibrated. Working almost exclusively on the nearfields and having long experience in the rooms with them is the key to consistency. I don’t find the ProAc overly bright.
Grace m908 Monitor Controller
[top]I just picked up a DK meter and was wondering, do you use it to measure loudness or are you only doing that for Atmos work? When set to the NBC scale, is there a level you’re aiming for, post master chain / limiter? - ryno1
I feed the DK meter AES/EBU signal from a mult of the ProTools main output. I don't have a microphone hooked up for loudness.
I mainly use it on the NBC scale, trying not to get too far above -6, some hot tracks peak above this, ballads might fall more into the -8 range on average. These are levels post all processing, on the output. I also will use the 1/3 Octave mode, occasionally the FFT, and I watch the correlation meter a lot.
The Grace m908 monitor controller that I use for Atmos work has a built-in microphone for SPL which is really nice for checking listening levels.
[top]What are your thoughts on the famous car check, listening on (apple) earbuds or your little computer speakers next to your screen? - Bobbl
The car checks are great when transitioning to a new studio, or a new speaker system. For many years after we moved into our current studio, we would do car checks. After a long period of time that faded out. For anyone working out of different rooms on a regular basis it is a great additional tool.
Earbuds only for casual listening for me. Computer speakers are really for auditioning files quickly by hitting the spacebar. They are on the output of the computer, not connected to ProTools.
As pointed out to me in a separate question, I may be using some technical terms incorrectly.I said "I mainly use it on the NBC scale, trying not to get too far above -6dBfs, some hot tracks peak above this, ballads might fall more into the -8dBfs range on average. These are levels post all processing, on the output."
I should not have used the term dBfs here. I'm not actually sure what the scale for NBC is, but it is closest to the currently used LUFS.
On the DK Meter, if I switch over to DMU1 I'm pretty much bouncing between 0 and -1. On the 40C scale, it would be around +14, VU is crushed to the top of the scale. These are all on a Loud mix, most would not be pushed quite this hard.
[top]Can you share some tips on how to make vocals work in a mix? It would be great to have specific tips on mixing male vocals. For example: how did you treat Bruno Mars's voice on 24K Magik? - Thedberg
Nothing I can think of is specifically done for male vocals. It uses the same tools; Compression, EQ, De-Esser. I always try to mix vocals for intelligibility and artistic presentation. I think it would be hard for me to mix my own vocals (if I was a singer). Does anyone like the way their own voice sounds on a recording? Again, no special specific treatments on Bruno's vocals. Great vocalist, great recording engineer (if you don't know, Charles Moniz not only records all of Bruno in studio, but also does his Front of House in concert), great production, and great direction and collaboration with Bruno in the mixing process.
[top]Are you conscious of your monitoring levels? - H-Rezz
Yes, I've touched on this in a few other answers. I try hard to be aware of listening levels, I don't really time myself. Frequent breaks of not listening or listening low are normal. I don't use the big monitors much at all, use the Headphones if I need to really concentrate or pay attention to very soft details.
Outside of work, I protect my hearing when using machinery, power tools. I don't go to concerts often, and when I do I always have earplugs with me. I don't listen to music on EarPods very often, and at low levels when I do.
I encourage everyone to look for other sources for technical recommendations.
[top]Teenage Dream is an iconic generational album that has set many records (including the big 5 #1s on the same album. At the time did you sense you were working on a special project? And can you share a special insight from those sessions? - s wave
I don't have any special insight specifically. I think we knew at the time that it was great and special. First Katy is a special singer and lyricist. You put that talent in the hands of Dr Luke, Max Martin, Ammo, Benny Blanco, Greg Wells, Bonnie McKee, and all of the other talent there; you're going to have something special.
I don't want to say it was just another job, because we were excited and happy and thrilled to be working with all of these guys and girls, but I don't want to come off too callous here.
Looking back there are a number of projects that hold a special place and that I consider to be part of history and not just a part of our body of work. Katy Perry "Teenage Dream" certainly is one. Some others that stand out for me.
- Musiq Soulchild : "Aijuswanaseing", "Juslisen", "Soulstar"
- Jill Scott: "Who is Jill Scott?"
- Kelly Clarkson: "Breakaway"
I also thought that Jewel "0304" was really special but it didn't get the same critical acclaim.
[top]Any interesting/unusual insight into how you guys approach drums? - shuchoco
Country stuff, I’m still sent a lot of live drums, pop not so much anymore. Even the country stuff is often layering samples and loops in production.
Occasionally we will augment a kick, snare, or cymbal with a sample, pretty much never doing drum replacement. If we are sent real drums we will work with them and use them. Any decision to do any drum replacement would be done before tracks are sent to us.
In my mixing projects specifically, I don’t like to put in my own samples. I know that the producers like something they are getting with the sounds they have so I make it my mission to figure out how to make what I’ve been given the best that it can be. I like the challenge of working with the sounds I’ve been sent as opposed to trying to make a quick fix and drop in a sample that I like. I don’t want everything to start to sound the same.
I like to process a live drum kit through an aux buss with a channel strip on it overall; also doing things on the individual channels. Mostly this is because when you’ve got good balances on a live kit it is easier to then do overall changes to the aux. I usually try to start with the OverHeads, and then add in the kick, snare, hat. I'll often do a lot of editing and compression on the toms to just capture the meat and mute the bleed of other drums. Much of that sound is otherwise in the overheads. I don’t tend to use much parallel compression tactics unless it comes that way from the producer.
I generally prefer to treat drum tracks separately rather than bussing all kicks, all snares together. If it is useful for a specific reason, then summing is fine as well. If there are a number of samples, try playing with their phase instead of just trying EQ and Compression. (If I slide this kick sample 100 samples left or right does it sound better or worse?)
Drums are usually the next thing after vocals that require a lot of attention and mind melding with the producers. Everyone is careful about their drum sound, so If I can see that a producer has put a lot of thought and energy into their drum production, I’ll tend to leave it mostly intact.
A lot of the drum vibe is created during production. on Dua Lipa “Blow Your Mind” those drums are very well done by Jon Levine. I just looked at that session and it is about 36 stereo tracks of drums and percussion. The goal during mixing of these kinds of productions is to maintain all of the work that the producer has put into it; we might not be touching much in the drums at all in this case; a bit of low roll-off here, a bit of high here, small level and balance adjustments. We might be doing a bit of panning changes to spread the drums and percussion, adding or tweaking some reverb sends.
The bulk of the drums mixing in this case would be polishing the work already done, fixing any problems, and mixing the rest of the song and the vocals so that all of that work and detail is not lost.
Mostly nothing special as far as plugins; I’m setting up and working like I used to on a console. There is an EQ, Gate, Compressor, Limiter, on each track. I’ll start with using those tools; if anything needs more specific effect or sound, I’ll start digging through plugins.
[top]After getting a good vocal sound and now working on the drums, are the rest of the tracks/music/instruments muted or always on? - bobbl
After I've got to the good rough mix stage, I might mute everything and then start bringing in parts starting with drums and working through all of the instruments to vocals; just to hear how everything builds and adds together.
After that, mostly everything on. There will be times to solo all drums and listen, or solo all vocals and listen of course; but I'll not mix separately as an instrumental and an acapella for any long period of time.
[top]The bottom end of songs mixed at Mixstar is always incredible. Could you give some tips on how to better judge the bottom end for lesser rooms? As an example, the synth sound at the beginning of Taylor Swift's "Don’t Blame Me" has a beautiful bottom end and the chorus just explodes. - octopi
It is difficult to do in a small room. Look at the woofers themselves and visually see what the low end looks like as the woofers bounce around on a mix that sounds good elsewhere. You can see some of the low end information. Check low end on good, not hyped, headphones.
[top]On some radio stations multiband gain reduction can be as much as 25 to 35 db with lots of stereo widening and EQ coloration to push it in a certain energetic sound signature. Do you keep that in mind when doing a mix? - NdK
Generally we ignore everything downstream of mastering and just work to create the best sounding mix we can. There is only one situation that I can think of where we have intentionally created different mixes for radio is with Country Radio. They seemed to really mess with the Bass so much that we have done different mixes for Keith Urban to send to radio than were used for other releases. His music is very pop influenced, so it doesn’t always blend with the typical Country Radio preset.
I don’t really love it or hate it, I just kind of ignore it. I think that most people only listen to broadcast radio in cars these days , which is far from an ideal listening environment anyway.
[top]Wwhat is your advice for young engineers? - citytape
This is a really hard topic. I think that I've been really lucky in the path that my career has taken. Let's say I've gone in all-in several times and been lucky to have a winning hand. There can't be a plan for that.
There have been several times during my career when I've been near broke, not getting any work, getting work and not being paid for it, and wondering if it will ever get better.
I've seized hold of every opportunity I've been given and tried to make myself indispensable in every situation. My advice is that it will be really hard, you might get used and abused, it is a grind. You might get lucky, like I have, and make a career out of it, but you need a backup plan and a paying job until that happens.
If you get a job as an assistant engineer, absorb everything, learn everything. Make yourself available anytime, anywhere to be ready to work. Be kind, be humble, understand that you are easily replaceable and find a way to make yourself not easily replaceable. Know how to learn on your own; don't ask a lot of questions when people are busy, figure out the why and how by watching and listening. Ask questions later when everyone is relaxed.
Don't make any enemies; you never know who might be the next A&R or head of a label for example.
Protect your hearing; carry earplugs, wear them in noisy environments. Don't go to a lot of loud concerts. Don't be afraid to step out of the room or put in earplugs if someone cranks up the mains.
Practice mixing; maybe you can share work with other young engineers. You mix theirs, they mix yours. Feel what it is like to work on someone else's songs and sessions. Practice the communication and understanding of trying to put mixing notes into words.
Practice recording; find a way to get studio time if you have any access. Maybe after you've documented and are breaking down a session, have someone play the drums while you move microphones and listen. Have someone play guitar while you move mics around the speaker cabinet.
The most important thing getting a foothold for me starting out was making sure I was impressing people. Put the clients first. Show up early, stay late, anticipate problems and solutions. Know how to do things; how quickly can you change a blown woofer, what do you do if a mic cable is bad and crackling, etc. You will not be judged first on your abilities with sound. You will be judged first on your reliability. You'll be judged on your ability with sound later, but you have to get to that place first.
PSB M4U 1
[top]Do you have a pair of known trusted pair of Headphones you can mix with when you’re travelling around, or do you just use them for editing and session prep? - ryno1
I'm using the PSB M4U1 Headphones. I like these in particular because they sound pretty similar to my ProAc speakers and they are not heavy. I have an Oppo HA-1 headphone amplifier that I use for headphone listening. When monitoring the binaural downmix of an Atmos mix, I'm using the headphone output of the Grace m908 monitor controller.
[top]What mixing tips do you have for producers/artists who self-release music? - MattyJoe
I'm not a producer or writer, so I don't have any real-world experience here to give you.
I might suggest some divisions of labor though. When you start mixing, try to use what you've already recorded, try not to fix a mix issue with adding new parts. Take some time between your production phase and your mixing phase. Take a few days, maybe a week of not listening to your song; then approach it with a clean perspective.
[top]I have a question about whether you and Serban ever get into doing "additional production and mix" for clients, or is it just purely mixing? - Aidyhall
I don't do any additional production in mixing; I am not really a musician these days (played trumpet from grade-school through college) I don't play keys or guitar.
I think I've seen Serban throw down a guitar part a few times in all of the years we've been working together.
Additional production is not part of our scope of work. We would never replace an element with one of our own. Producers will often send a replacement part if something isn't feeling right to them.
[top]How do you approach automation? Do you use a controller and do it "live" (as the session is running)? Do you draw it? - Oroz
I’ve got a control surface; a mouse is the control, and a mouse pad is the surface.
Seriously, pretty much all automation is drawn in on the edit screen automation lanes by mouse; occasionally I’ll turn on automation parameters and ride a “knob” live via mouse. I like the precision that drawing in via mouse gives; instant square volume changes between words, or completely smooth point to point rides.
This is just how my interaction with ProTools evolved; when we started there were no real control surfaces, and when you’re comfortable with a mouse it seemed silly to buy a Pro Control or an S6 for thousands, tens of thousands of dollars to replace a mouse.
I will get basic balances going and pretty quickly start doing the obvious volume automation moves that need to be done as I’m learning the song. These would mainly involve leveling out vocals, as they have the most variability within the song.
As the mix proceeds, the automation will get more and more detailed. Smaller volume moves, pans, effects sends rides or mutes. I might turn on and off some EQ bands for different song parts of the same track. Say a high hat is poking through a snare track or looped drums, I might ride every hit down with automation rather than messing with trying to EQ, set a compressor, or other “fix”.
Automation on the master, yes sometimes. A good way to make the chorus have more energy and drive than the pre-chorus is just pop it up 1dB on the master volume.
After all of this, though, probably not as much automation going on as you would think. I’m doing automation during the whole mixing process, but mostly small moves or setting big moves for sections of the song (i.e. panning wider in choruses, or bypassing an effect in the chorus). A lot of the “tweaks” that artists and producers ask for become small automation moves; turn this down, turn this up, for certain sounds or certain parts of the song. More reverb here, less delay there kinds of things.
The interesting thing that I’ve gotten into recently with Atmos mixing is that I’m doing a LOT of automation there for panning (there are 28 lanes of pan automation on a stereo track going out to a 7.1.2 Atmos output).
So in Atmos mixing, I find myself doing a lot of selecting automation types (write, latch or touch), setting automation modes (write on stop), and dragging handles on the surround panner, and then going back and grabbing different handles. So there I might do one pass where I’m panning back to front, and then a second pass where I’m panning top to bottom. So overall the sound sweeps from top back to front bottom. I wish I had a motorized 3-D surround panner for this.
[top]Do you generally boost to make things stand out, or do you carve at things that might obscure what you want to stand out? - Lei
All of the above. If the problem is too much low frequency, then don't boost the highs. Identify the problem and apply the simplest solution. Boosting is more likely to drive the track to distortion, and everyone likes to add more of this and more of that, but try to avoid only boosting.
[top]I have a question regarding building a mix template to have a pain-free stem mixdown. A lot of labels ask for stems and they check if stems are summarized perfectly to sound like a printed mix. But with excessive mix buss processing and parallel aux processing it can be quite tricky to bounce separate stems so they would sound the same as print. Any thoughts on building a smart mix template to have the ability to bounce perfect stems? - EvgenyStudio
I run into the same problems when ProTools sessions come in from some producers that have complex and nested aux tracks.
There is not a pain-free solution and there are not perfect stems. The purpose of the mix is to make the Stereo track sound the best; if the stems suffer because of this than too bad! If you want me to mix for perfect stems, then don't ask for a perfect Stereo mix.
A smart mix template for *near perfect* stems is to not use any sub-mixed aux busses, not put any effects on track inserts, and not do very much Master Bus processing. This is not realistic and is not the goal of a Stereo mix.
[top]Are you primarily a mix engineer or do you track a lot of projects these days as well? If and when tracking, are you working out of MixStar in Virginia Beach, VA? - Wilkinswp
Just mixing these days, no tracking. I think the last project that we fully tracked and mixed was Kenna "New Sacred Cow" or NERD "In Search Of"Our studio is a private facility, there is no good information available in regards to the studio.
[top]I want to ask how you generally treat esss's and t's as they always seem to sit tight but never disappear. Do you more often than not de-es manually or how do you generally approach this? - Lei
A bit of both Plugin De-Esser on the tracks, but also manually pulling down anything that still stands out too much. I don't want to crank the De-Esser so hard to smash the big ones that it makes the whole track lisp-y. (Yes, I read the thread about The Weeknd! Glad that our mixes passed inspection). So a lighter touch on the Plugin and manual riding.
[top]Are there any pointers to getting voices to sit so perfectly in a mix? Are there any particular go-to compressors, reverbs, eqs etc, that you tend to rely on for vocals? - Coldsnow
It kind of comes naturally after a while, but a few things that I do that could become pointers.
I Always mix with the vocals on. I don't mix an instrumental and then add the vocals. First step after getting to the "good rough mix" stage is get the vocals sitting right, and then go and mix the instruments with the vocals in the mix.
Listen really soft; can you hear and understand the words? If not, vocals are too low or muddy. Listen without concentrating; play the mix down while doing something else. What sticks out, what is missing when you are not in full concentration mode. Any particular EQ, compressor, verbs, etc. that I use and rely on are not because they are in any way special or amazing, but because I am comfortable with them.
[top]Do you find that you often sweep around for bad frequencies by increasing volume or do you feel like you typically know what frequencies need to be attenuated by what sounds good in the mix? - Coldsnow
Mostly by what sounds good in the mix; I can't really solo a track and sweep for bad frequencies because they might only be bad when interacting with all of the mix. Only after identifying which track I might do a bit of sweeping to find the sweet spot to cut.Mainly I'm identifying problem areas by simple listening as well as looking at a 1/3 octave analyzer, and now we've got really good frequency graphs in many EQ plugins.
[top]Do you find you gravitate to various mix bus chains depending on the style of music, example mixes for Lana Del Rey or Camila Cabello, or do you have one general chain you know just works? Also, how many db's do you normally leave for the mastering engineer to do their thing? - H-Rezz
We have one standard mix buss chain that is used on everything. We mix through it, meaning it goes on at the beginning of the mixing process and the whole process of mixing is done with it in place.
It is mostly “set and forget”, but there are times when small adjustments might be made to overall EQ, compassion, limiting, etc.
When we get sessions from producers that have their own mix bus chains in place, we’ll review what they are doing with each plugin. Some things we will keep in place because it has become an integral part of their rough mix. Other things can be removed or replaced by our own tools. Quite often just removing some things from the rough mix chain will make the mix better right away.
I won’t share exactly what we use here, as that chain was developed and perfected by Serban and is his “proprietary” information to keep to himself if he wishes. We generally leave little for the mastering engineer to do. It takes a great skill to take it to that “sounds like it’s already mastered” place, so definitely a pro move.
So on a typical loud mix, at the loudest part of the song, on my DK Audio meter, set to NBC scale which is AES full scale, we’ll be sitting around -6, with peaks to -4. On the Waves WLM Meter, which I normally look at just for Atmos mixing, it is pretty crushed at about -6LUFS at the loudest part of the song.
On occasion the mastering engineer will ask us to bring it down, and rightfully so. Other times, the artist or producer will say we like the mix better than the mastering, can you ask them to pass it straight through. The theory here is do everything that will affect the sound inside the mix, which the artist, producer, and mixer have tweaked to perfection.
[top]Do you have a separated Vocal Bus and Music Bus to avoid the vocals driving the main bus compression or do you use the old school approach of sending everything, vocals and music, to the same bus compressor? - Jakelorenz
If a ProTools session comes in from the producer with separate Vocal Bus and Mix Bus, I'm not going to remove it. They created it with that processing in place and it will change things too much to rework it after the fact.But even in these cases, everything is still going to go through the Master Bus chain and output 1&2. The Master Bus plugins and settings might need to back off some as a majority of that compression is being done already.I'm not going to create a session that is set up that way, I definitely prefer what you call Old School! I guess I am getting old.
[top]When you add the mix buss chain into the session. Do you have a trim plugin on the beginning of the chain so the music will hit the mixbus chain at the same volume for all the songs you mix? (some sort of consistency I guess). - musicmixer04
No trim plugin. Consistency comes from experience and listening to the volume, looking at the meters, and knowing about what levels I should be at to hit the Master Bus about where I want constantly. If things start to get too hot, I'd rather make a group of everything that is feeding the Master Bus and trim all of the tracks down equally rather than using a plugin on the master to do it. Always trying to maintain proper gain structure and not just compensate for bad gain structure at the end of the chain.
[top]Is there any type of Pultec low end style boost going on the entire mix where you would add a bunch of bottom end to the entire session? Or is it mainly just limiting and small EQ moves / stereo m/s? - Spankjam
Generally no. Again if the producer has sent us his ProTools session and crafted his mix with this in place on his Master Bus, We will evaluate it and see if it is needed to maintain the production values he has set up.In my preferred world, low end should be crafted and nurtured on the individual tracks where you have much more control over it; just adding a bunch to the whole mix feels very imprecise to me. Chances are you're just going to be boosting a bunch of low mud that has not been filtered out of individual tracks.
[top]How much of the overall sound would you say comes from the mix bus chain? Like if you bypassed it all, other than loudness, would the mix sound drastically different or is it all doing quite light touches. Is the compression and or limiting full range or is there a multi-band process in use? - Mr XY
I think other than loudness, it is doing light touches. As I said before, we are mixing through this chain from the beginning of the process, so all of the mixing decisions and tweaks are being done on the individual tracks but heard through the whole chain. So if I want to widen a sound, I'm not widening on the Master Bus, but I am hearing the consequences of widening through the Master Bus. Both full range and multi-band processes are used.
[top]Is the same true for subgroups and busses, as in do you generally work within standard chains for these groups as well that feed into the mix buss? - ULA
Subgroups and busses are created as needed for things like BGV's to apply processing to all equally. We don't have, for example, a Bass Buss or Drum Bus that has stock presets. If one is needed, it is created for that mix.
I wouldn't generally mix into a high/low pass eq. You're just going to compensate elsewhere by boosting or cutting those frequencies on the individual tracks. I would prefer to fix these things at the source. Nothing else to add on the Mix Buss!
[top]Can I ask if you use very minimal compression on the master bus? - jakelorenz
If the Vocals are driving the compressor too much, then they are too loud or you are using too much bus compression. If I take my bus compressor and crush the mix ridiculously, it doesn't change the balance of music and vocals, it only changes the overall volume and of course distorts.
[top]Most of your mixes have some kind of grit/saturation only noticeable on Headphones. And it seems that any genre of music you mix goes through this process. I would really appreciate If you could share your philosophy concerning this process. - Young.baws
I think that we all have grown used to music sounding as if it has come off of tape. Even music created when tape machines are a distant memory are influenced by the past benchmarks of music created on tape. It just sounds good, right? Apply as you see fit.
[top]Delivery/stems bouncing - Could you share some insights on how you approach this? - Lei
Stems are a part of our standard delivery to the label. These days I offline bounce all of my stems.
I don't bounce the stems right away as the mix passes are delivered to mastering; it is something I take care of as I find time between more urgent requests. One benefit of 100% in the box is that I can open and close things as needed to do these tasks; there is no reset needed to go into stems mode or mix mode.
While I'm bouncing stems, I'll be doing paperwork, catching up on billing, or just taking a hearing break.
[top]I find it difficult to bounce with side chaining a master buss processor so that every stem is reacting the same way it would for the mix on each individual stem. Do you do this? - Dias
If you're talking about doing a side-chain to pump the whole mix; we don't do that. We don't mix with side-chains pumping the master bus. Any side-chain is within the tracks or groups on an aux bus.
If you're talking about generally the overall processing, limiting, compression of the master bus not hitting the same, I don't worry about it. I print all the stems through the master bus, they do come out hotter than the mix because of the difference in the overall bus compression, etc., I'm not doing any side chain or compensation for stems.
I do just solo the parts for the stems and print what plays, reverbs, effects, and all.
The purpose of the stems is not to re-create the mix perfectly when they are all played back together. It is not going to happen. We do sometimes get complaints about this and I will tell them to drop everything by 6dB, put a bit of master bus compression on the whole thing and you're 99% there. If a special mix is needed, we'll do it here, don't try to do it from the stems.
The purpose of the stems is for the artist to get remixes done where they might use some of the elements in a new production, so it doesn't matter if they perfectly re-create our mix.
The other purpose is for the artist to be able to do live shows and use elements from the mix where again it doesn't need to be a perfect reproduction of the mix. The FOH engineer is going to be making their own adjustments and probably blending in live instruments.
[top]I was wondering if there are some plugins that you use on every mix...or could you list your top 3-5 indispensable plugins? - Scubaman 6000
So one thing that goes on here is that we work with so many different producers and artists using so many different plugins, that we are buying plugins regularly to be compatible with the ProTools sessions as they are sent to us.
My AAX plug-ins folder has over 1000 items in it (when all folders are open).
As the producers are constantly experimenting with new plugins, we seem to collect more and more of them. To narrow that down to a Top-5 of what I choose when I am choosing the plugins comes up with a list of things that mostly I have used for a long time and are familiar with.
- Metric Halo ChannelStrip 3: This is no secret; it is the first go-to EQ/channelstrip picked for anything. This is mainly due to comfort level. When you are comfortable working on a console like an SSL 4000, you grab the eq knob or the compressor knob without really looking at it and turn it to what feels right. You don’t need to look at the printing to see if that is 1.5kHz or 3kHz, if the Q is 2 or .5; you have muscle memory by experience and feel. It is that same familiarity with this plugin; I can just grab the handles in the EQ window, or quickly click buttons or type in values and set it the way I want. No hunting for functions or hidden settings.
- Waves CLA-76: the first thing I try on vocals. Blacky, 4:1 and hit about -3dB compression generally. Bluey, 4:1 and -10dB compression for in your face.
- Waves Renaissance DeEsser: Another old friend. Frequency 5500, hit about -6dB and adjust to taste. Vocal channels generally get Compressor, then EQ, then De-Esser. If I need to clean out low-free on it, I’ll put an Avid EQ3 1-band EQ and filter the lows before compressor.
- Ozone RX-7: really quick to clean up track noise, mouth noise, repairs.
- Valhalla Vintage Verb: very popular on sessions sent to us, so is has become a go-to tool.
- Waves H-Delay or Avid Mod Delay III: good basic delays; if I’m looking for more specialized effects there are a world of them to choose from.
- Avid Lo-Fi: add a tiny bit of distortion (.1-.5) or saturation (.1-.4) to smooth synth cymbals.
- Kilohearts: I’ve been falling in love with these, simple utility interfaces, easy to understand. Distortion; add hard clip and distortion to an 808 to bring forward on small speakers. Chorus to add body to BGV, Haas for spreading a mono track, all sound good.
iZotope RX
[top]Regarding your RX7 use - I find I expect to pass at least the lead vocals through this to clean it up on every mix. Is that something you find yourself doing, or is it later in the process, only fixing the stuff that stands out? - psycho_monkey
Some producers will have taken care of enough of this before they send that I don't need to do this extensively on every song that comes in.
Here is how I clean vocals: First I want a good rough mix going; I'll get the vocals at least with basic settings on compression, EQ, and de-esser. I'll get the volumes and balances close. Much of the time this is done for me as the producer has sent the ProTools session of his rough mix. I don't run a complete top to bottom pass of RX-7. I'll open up a few Audio Suite plugins (turn target off so you can open multiple plugin windows).
- RX-7 Mouth DeClick - default to start
- RX-7 DeClick - multiband algorithm, default to start
- Gain - -6 to -9dB
- EQ3 1-band - High Pass, maybe 150Hz to 200Hz.
For mouth noises; I’ll select maybe one second of the audio around the click I hear and zap it with Mouth De-Click. I’ll listen back to what it has done, and adjust if necessary.
Some clicks are better resolved with standard De-Click. Plosives I’ll hit with the EQ3 filter, big breaths and some excessive sibilance I’ll hit with the Gain. I’ll also take this time to cut (region separate and mute) and fade the beginning and end of phrases as needed, and cut any noise where there are no vocals.
I have to be careful here of room noise appearing and disappearing as the vocals come in and out, so sometimes it might need to be left in or faded out slowly. Other times if it is really annoying I’ll go into full RX-7 Connect / RX-7 Monitor mode and deal with room noise, hums, or other, deeper issues.
I’ll do the same to each BGV part, but here I’ll often completely mute breaths and make a quick fade in on the vocal. Also sibilance on BGV’s can be treated much more harshly.
If voices in the BGV’s stack are out of time with each other in the attack or release, I’ll fade back the ones that are early or held long. Often this is to make sure that the Left and Right panned vocals are beginning and ending at the same time, and timing with the Lead vocal is also suitable. I’ll do this for each stack of Background Vocal parts, and any other vocals (vocoder parts might need a lot of breath reduction, gang vocals might need a lot of dead space mutes).Again, as I’m doing this, I’m listening back and adjusting everything I’m doing as I go.
So it can be a very tedious process in a session with tons of stacked BGV’s, or with bands with a lot of lead vocalists (boy band, K-Pop).
No shortcuts. Especially on vocals, because we are naturally so attuned towards vocal comprehension, each fix needs to be consciously made.
[top]Are there situations where the client actually WANTS clicks and bad edits in? Is this a conversation that you'll have before diving into tedious vocal clean up mode? - rhythmic5
Yes, this sometimes happens; usually towards the end of the mix they start to regret that as those bad edits or clicks suddenly become much more prominent because everything else in the mix is nice and clear and clean and present.
There are some producers that we work with that I know don't want things fixed or changed much, so I'll take a lighter hand with those.
Also, if the producer puts a vinyl crackle track on the song, I'm much less inclined to go in and spend a lot of time fixing vocal clicks.
[top]What are some tips and techniques you have for getting very wide and focused synth/bass pads? - JanetB
What you do is hire Max Martin and his WolfCousins to do production! Seriously, this is not something that is created in the mix; it is the producer's choices and skills.
See also new mix by Serban for Tones & I "You're So ****ing Cool". Beautiful Steve Mac production and wide!
Don't be afraid to pick a pad that is a bit out of phase for this.
[top]What techniques do you use to make your mixes hold up in mono and stereo? Eg uptown funk and 24 carrot magic. - Scubaman 6000
To be honest, mono compatibility isn't really a top concern. I don't listen down to the mix in mono.I do pay attention to it as I'm working. I am listening for phase and watching the phase meter on the DK Audio meter that sits directly in my line of sight.If I hear or see something going out of phase, I'll find which track is doing it an see if flipping the phase on one side of the stereo channel makes it better, or if it changes the sound too much. I won't sacrifice the sound for mono compatibility. Some choices by a producer to have a super wide sound in a mix might make it mono incompatible, but I don't think I've ever heard anyone really care about it these days.
[top]Track Levels - Do you "normalize" the rough tracks before they go through your chains or do you adjust your chains to accommodate different signals and levels? - mix_dome
I don't normalize rough tracks. If the recorded signal is so faint that I can't even see the waveform in the standard view, I might AudioSuite Gain the tracks up to that I can at least see waveforms.
Other than that it is just using regular fader volume or gain settings on any compressors to bring the level up. Also the Master Fader volume can be brought up to get the gain structure set correctly before it hits the master bus plugins.
Achieving "perfection" with quick turnarounds and tight deadlines
[top]Could you share some workflow concepts, like how a top mixing firm actually "DOES" the job so quickly so to speak, that would be superbly appreciated. - CanadaSC1
Several levels going on here. First, do we 'create' chart topping mixes? Or do the songs top the charts and drag the mix along? Would a 'bad' mix of a Taylor Swift or Bruno Mars single not top the charts? It is kind of a chicken and egg problem isn't it.
Obviously when topping the charts we are blessed to work with really good artists, usually very well done recordings and productions, great songwriting. I think a great mix makes the artist, producers, and A&R more comfortable that they have a great product to release.
Dealing with the demands of clients, deadlines, and timelines is a universal issue. First, deadlines are mostly meant to be broken. How many times we do a mix under a deadline, submit it, it is approved, even mastered. Then two weeks later; “we’re going to send you a new bass-line, and the singer has some new ad libs to drop in.”
So you do what you can by the deadline, and leave it up to the people who enforce the deadlines to say if the song is done or not. Some deadlines are hard and real. In those cases it might not be the best mix that can be, but obviously in something so rushed, the perfect mix is not the goal of the project.
Mixing songs for TV releases is usually the hardest deadline. In the mixes for Songland that Serban mixed; Jonas Brothers “Greenlight”, Julia Michaels “Give It To You” the files arrived about 3 days ahead and had to be ready for TV broadcast with an absolute hard deadline. For Ariana Grande’s “Thank You, Next” single files arrived in the morning and had to be out to mastering by that evening.
In these cases you have to fall back to basics. Make the mix sound good overall, don’t nitpick fine details. Fix things that stick out. Don’t get panicked and spend every hour working on the mix; get a good mix going, then walk away. Work on something else for a while; come back with a fresh perspective. Do the mixing in many shorter sessions rather than one long one; it is totally counter-intuitive when a deadline is crushing you; but your first 30 minutes of mixing might be your most productive; so create those first mix revelations over and over again. Don't concentrate too hard, let you subconscious hear things that aren't hitting right. Vary the volume, listen loud for a bit, listen soft for a bit, don't burn out your hearing trying to get the chest-thump on the kick. Of course being really efficient and organized in general helps in these situations.
Here is how our workflow here operates. When a song comes in, I deal with the files. My job at that point is to make sure that technically it is ready to mix. First, do the files sent match the rough mix? Same production? These files are saved on a drive called “Original Files” Next, “Save Copy In” and copy it all to you “Prepped Mixes” drive. NEVER work on the original files. You will, at some point, want to refer back to the original session or files untouched.
Then I will get the session organized in the way we like it here; drums at the top, percussion, bass, guitars, keys, synths, BGV, Adlibs, Lead Vocals. Everything put in order, labelled, and groups made to easily identify parts. I’ll also do any cleaning of clicks, pops, bad edits, noise.
I’ll simplify the session if necessary and if I am able to. Clean up the I/O Setup and get rid of all unused busses, inputs, outputs. Rename and move busses that interfere with our “effects template”.
Drums going to an aux bus with no plugins? I can copy the volume moves to the tracks and send it all out the master bus. Lead Vocals and BGV’s feeding the same Vocal bus? I will duplicate and create an aux for just the Leads and one for just the BGV’s. the entire goal here is to get something into Serban’s hands that is organized, cleaned, and easy to understand. If he hears a tamborine, he knows where in the session it can be found, it is labelled, and grouped with the Percussion.
He should then only need to worry about the artistic decisions of mixing and not any technical details that take him out of the mixing headspace.
If you don’t have someone to do this work for you; as I don’t when I’m mixing my own projects, you can approach it in the same way. Do the work in two steps. First act as the technical engineer and do all of the organizing and cleaning tasks. You might identify things to hit later in the mix, but don’t spend time mixing at this point. When you have the mix organized and technically ready, then copy the files to you “Mix Sessions” drive. Again, don’t work on your cleaned up session or files; you will want to refer back to them at some point.
Also back up everything regularly. Have multiple copies and backups. Stay organized that you are working on the current file, but have backups ready. Nothing kills momentum faster than “oops my drive crashed and I lost everything”.
[top]Do you have any delivery requirements for the producer, or is it usually just "send the session as is?" - Mattjhuber
We work with so many different producers, who are working in so many different ways, that we just need to take whatever they send. Sometimes it is a full ProTools session, complete as they left it. Sometimes people will commit or remove some or all plugins (removing just makes it harder for us). Maybe half the time it is not a ProTools session at all, but a "pile of files" multitrack. This can range from a producer's "stems" (we always ask for at least dry vocals), to some producers who send a couple of hundred files with every instrument and vocal track separate as well as printed effects for each vocal track (hopefully separated by folders such as BGV Wet, BGV Dry, Lead Wet, Lead Dry, BGV printed effects, Lead printed effects). That's when I put on my white lab coat and goggles and get to work!
[top]You mention prepping the files yourselves. Do you have assistants at MixStar, and if so what roles do they play - is this the sort of thing you'd delegate, or do you like to be hands-on from the start? - psycho_monkey
We don't have any assistants here; it is a two-man operation. We have had assistants in the past, but as we don't do any recording and are mixing in the box, there was little for them to do. Ultimately they got frustrated, we understood and helped them to move forward elsewhere. There was just not much for them to do.
If we set up another room, spent another, let's say $75k for another full ProTools setup, all the plugins, etc; maybe they would be able to start to prep mixes and print stems. As businessmen, we have to weigh the pro's and con's of that and the cost/benefit. So far it would not be a good financial move to spend a bunch of money on equipment, set up payroll, etc. so that we can get stems out a bit faster. We can't mix any more quantity than we are now, so any additional prepping help would not be useful.
[top]I’d love to know if you’re mixing ITB 100% or do you still use a couple of hardware units in your system? I saw a 33609 in the The Weekend’s Dolby Atmos Mix article. If you do use it (maybe it’s just sitting in the rack), how do you use it? Mix bus? Drum bus? Something else? - Oroz
Yes! 100% in the box mixing. The 'bounce to disk' is what is sent to clients and to mastering. Never touches analog in the mix, no external summing, no console. Everything is routed out the A 1-2 Master Fader.
Now, on occasion, and this is getting very rare, there might be an insert on the master fader. That is what the Neve 33609/J has been used for. But before anyone goes out and buys one because we have it; I think that last time we used it on the insert was maybe 4 years ago!
We are also using the Avid HD I/O, and internal clock. Because there are no D/A and A/D conversions happening, aside from going to the monitors and external meter, we don't need any fancier digital converters or clocking.
