Quantcast
Mix elements half disappear in MONO. Need help troubleshooting. - Gearspace.com
The No.1 Website for Pro Audio
Mix elements half disappear in MONO. Need help troubleshooting.
Old 4th February 2017
  #1
Lives for gear
 
Ain't Nobody's Avatar
 
🎧 10 years
Mix elements half disappear in MONO. Need help troubleshooting.

I've always avoided widening plugs and typically thought that plus the occasional half-hearted check in mono was enough, but I'm coming to the painful realization that I don't know squat about mono compatibility. Time to bite the bullet and learn before I churn out another mix where elements half disappear when summed to mono.

1) VERBS - When using stereo reverbs, what exactly about them makes them "Mono friendly" or not? If I'm creating my own reverb patches (Relab 480L or similar) what should I be looking out for, avoiding, or making sure to include in each patch? I'm hearing at least a 3db drop in most of my verbs when mix summed to mono.

2) PANNED DELAYS - I would think this would be an easy translation into mono if I have, say, a ping pong where 1st 16th delay is panned hard L, and second is panned hard R. Somehow, even this, though, is noticibly lowered in volume when in mono. Feel like I'm losing my mind on this. Is this a panning law issue in the DAW? Why would hard L and R panned delays sound significantly louder in stereo, and softer in MONO?

3) SYNTHS - I generally do things like instead of using one saw wave, using two slightly different starting waveforms, then tuning one up 10 cents and panning hard L, and then tuning the other down 10 cents and panning hard R. Of course, I expect it not to sound "wide" in mono, but I would think it should sound at a similar volume. It doesn't. Sounds made this way sound much lower volume in Mono. I was always under the impression that I could do such basic detuned panning so long as I avoided any artificial means like widening plugins. Have I had this wrong? Is any sound created with detuned panning like this doomed to collapse into a tiny (and much lower volume) version of itself in mono?

More importantly, how do I fix these things? For instance, on the synths, am I missing some sort of m/s trick?

How are other producers dealing with this? What sort of techniques exist to get stereo verbs, panned delays, and thick synths without sacrificing mono compatibility.

AND/OR... is there some issue with the way I'm summing to mono. I'm just hitting the MONO button on Logic gain plug on master output (since hitting the mono switch at bottom would get rid of all of my stereo plugs on master channel.) Is there something about that that's not "true mono?" Would it make any difference if it were via summing cable or some other method?
Old 4th February 2017
  #2
Lives for gear
 
voodoo4u's Avatar
 
🎧 10 years
When things fold or disappear when subbed to mono, its strictly a phase issue. When information in the left and right channels are identical, when subbed to mono it will be exactly the same. All stereo information is to some degree slightly out of phase in one way or another and to varying degrees. We do not always think of it in those terms. Whether its a different EQ on both channels or slightly out of time (ie. two different mics on a source), or flipped polarity, they are all indications of the same thing. Out of phase information.

Sometimes it comes down to a matter of degree. Slightly out of phase information will give you a slightly wide stereo image. The closer two signals get to being 180 degrees out of phase, the less mono compatible they will become to the point where at exactly and completely 180 degrees out, they would cancel each other out.

The trick with stereo imaging and mono compatibility is to to have just enough of the right information out of phase to give you a good stereo image but not so much that mix collapses when subbed to mono. One of the characteristics of sound that that help you with this is to know that low frequencies are fairly non directional and the higher the frequency, the more directional sound becomes. I keep this in mind when I mix and I tend to leave low frequency info mono and in the center and I will try to get high mid and high frequency stuff ie cymbals, shakers, percussion to have more stereo (out of phase) content.

You mentioned MS info. The nice thing about something miced MS is that you can control the amount of center information and how much or little it will collapse when subbed to mono.

You mentioned delayed information. As long as any left-right delay is greater than 20ms or so it will not collapse in mono (Haas effect) so this would not be the source of the problem when subbing to mono.

Most of the time, companies that put out pre-created samples keep this in mind when creating them. This includes verbs and synths. I would not trust them entirely though and always test them myself. If I find that an individual sound collapses too much, I would just pull the pans in from 100 per cent to maybe 50 or 60. Whatever works.

Last edited by voodoo4u; 4th February 2017 at 06:08 AM.. Reason: TMI
Old 4th February 2017
  #3
Lives for gear
 
minor_glitch's Avatar
 
🎧 10 years
You can use a phase correlation meter to help you troubleshoot the problem. If the bar is constantly going to the left of zero, you've got problems. Something like Flux:: sound and picture development can help (look at Phase near the bottom of the plugin). Just start disabling/muting things until you find what's causing it.
Old 4th February 2017
  #4
Lives for gear
 
🎧 10 years
Quote:
Originally Posted by voodoo4u ➑️
When things fold or disappear when subbed to mono, its strictly a phase issue. When information in the left and right channels are identical, when subbed to mono it will be exactly the same. All stereo information is to some degree slightly out of phase in one way or another and to varying degrees. We do not always think of it in those terms. Whether its a different EQ on both channels or slightly out of time (ie. two different mics on a source), or flipped polarity, they are all indications of the same thing. Out of phase information.

Sometimes it comes down to a matter of degree. Slightly out of phase information will give you a slightly wide stereo image. The closer two signals get to being 90 degrees out of phase, the less mono compatible they will become to the point where at exactly and completely 90 degrees out, they would cancel each other out.

The trick with stereo imaging and mono compatibility is to to have just enough of the right information out of phase to give you a good stereo image but not so much that mix collapses when subbed to mono. One of the characteristics of sound that that help you with this is to know that low frequencies are fairly non directional and the higher the frequency, the more directional sound becomes. I keep this in mind when I mix and I tend to leave low frequency info mono and in the center and I will try to get high mid and high frequency stuff ie cymbals, shakers, percussion to have more stereo (out of phase) content.

You mentioned MS info. The nice thing about something miced MS is that you can control the amount of center information and how much or little it will collapse when subbed to mono.

You mentioned delayed information. As long as any left-right delay is greater than 20ms or so it will not collapse in mono (Haas effect) so this would not be the source of the problem when subbing to mono.

Most of the time, companies that put out pre-created samples keep this in mind when creating them. This includes verbs and synths. I would not trust them entirely though and always test them myself. If I find that an individual sound collapses too much, I would just pull the pans in from 100 per cent to maybe 50 or 60. Whatever works.
There might be some good suggestions in here, but I also saw a few things that I don't think are actually true. Unfortunately I'm on my phone right now so I can't reply in detail, but I will when I get to a computer. Basically, saying that stereo is simply about "phase" is making the situation way too simplistic. It's about difference, which may or may not include phase. Completely different signals with absolutely no phase relationship can be stereo as well.
Old 4th February 2017
  #5
Lives for gear
 
Waltz Mastering's Avatar
 
1 Review written
🎧 10 years
Quote:
Originally Posted by Ain't Nobody ➑️
I've always avoided widening plugs and typically thought that plus the occasional half-hearted check in mono was enough, but I'm coming to the painful realization that I don't know squat about mono compatibility. Time to bite the bullet and learn before I churn out another mix where elements half disappear when summed to mono.

1) VERBS - When using stereo reverbs, what exactly about them makes them "Mono friendly" or not? If I'm creating my own reverb patches (Relab 480L or similar) what should I be looking out for, avoiding, or making sure to include in each patch? I'm hearing at least a 3db drop in most of my verbs when mix summed to mono.

2) PANNED DELAYS - I would think this would be an easy translation into mono if I have, say, a ping pong where 1st 16th delay is panned hard L, and second is panned hard R. Somehow, even this, though, is noticibly lowered in volume when in mono. Feel like I'm losing my mind on this. Is this a panning law issue in the DAW? Why would hard L and R panned delays sound significantly louder in stereo, and softer in MONO?

3) SYNTHS - I generally do things like instead of using one saw wave, using two slightly different starting waveforms, then tuning one up 10 cents and panning hard L, and then tuning the other down 10 cents and panning hard R. Of course, I expect it not to sound "wide" in mono, but I would think it should sound at a similar volume. It doesn't. Sounds made this way sound much lower volume in Mono. I was always under the impression that I could do such basic detuned panning so long as I avoided any artificial means like widening plugins. Have I had this wrong? Is any sound created with detuned panning like this doomed to collapse into a tiny (and much lower volume) version of itself in mono?

More importantly, how do I fix these things? For instance, on the synths, am I missing some sort of m/s trick?

How are other producers dealing with this? What sort of techniques exist to get stereo verbs, panned delays, and thick synths without sacrificing mono compatibility.

AND/OR... is there some issue with the way I'm summing to mono. I'm just hitting the MONO button on Logic gain plug on master output (since hitting the mono switch at bottom would get rid of all of my stereo plugs on master channel.) Is there something about that that's not "true mono?" Would it make any difference if it were via summing cable or some other method?
Have you run commercial tracks through the button, and found the results aren't similar?
Old 4th February 2017
  #6
Lives for gear
 
🎧 5 years
Cool thread, very interesting!
Old 4th February 2017
  #7
Lives for gear
 
Stitch333's Avatar
 
🎧 10 years
mix and print in mono.
You'll know if it works or not right away.

Plus it's 'old school'
Old 4th February 2017 | Show parent
  #8
Lives for gear
 
voodoo4u's Avatar
 
🎧 10 years
Quote:
Originally Posted by unitymusic ➑️
There might be some good suggestions in here, but I also saw a few things that I don't think are actually true. Unfortunately I'm on my phone right now so I can't reply in detail, but I will when I get to a computer. Basically, saying that stereo is simply about "phase" is making the situation way too simplistic. It's about difference, which may or may not include phase. Completely different signals with absolutely no phase relationship can be stereo as well.
There are really only two basic elements to sound: amplitude and frequency. When two independent sources are played at the same time, there becomes a sum/difference relationship that happens. That's it. Everything is related to these two. When I talk about "phase" I'm referring to a much more complex definition than you may be thinking of. Phase, a sum/difference in time between two sources, and amplitude are all there is.

Think outside the box about phase (greater than 0-360 degrees) and you'll see what I mean. In stereo/ mono compatibility,in EQ (phase differences), harmonic series (phase differences) delay (phase differences) ADSR (once again, there will be differences in the phase relationship.

Completely different signals with absolutely no difference in phase relationship are mono. They cannot be stereo. They can come across two speakers, but they'll be mono. If you hear a difference, then it's within your speakers or room. It's after the sound leaves your DAW.
Old 4th February 2017
  #9
Lives for gear
 
Ain't Nobody's Avatar
 
🎧 10 years
Good stuff. Keep it coming.

I should have clarified a few things in original post (and asked a few more questions), but since there are already responses, I'll try here:


1) Anything panned hard to one side or the other will sum into single channel at -3db, right? I think I've been somehow assuming (without ever really thinking it through) that it's a non-issue since the other side was at 0db. My thinking was that the summed mono signal is at -3db from the hot side, BUT, it was only going out one speaker originally. Now that it's summed, it's playing on both speakers. Adding one speaker at same input level yields +3db, so it's a wash. Down 3db from summing, and back up 3db from playing out both speakers instead of only one when it was stereo. I'm thinking now that that assumption is likely incorrect and at the heart of some of these issues. If so, I guess I've been missing a fundamental concept that ANYTHING panned to the side automatically loses energy upon summing, and that there's a direct tradeoff between width of panning and mono level, so if I'm going to pan anything at all, I need to ask myself how far out I want it to be while remembering that the further out it goes, the closer it gets to it's -3db summed limit. I'm sure I must have learned all this stuff when I started out, but this is what happens when there's 25 years between when you learn something, and when you put it into practice. Damn, I'm old, but hey, it's never too late to reteach an old dog old tricks.

2) I'm already summing everything below 124Hz (just a figure I arrived at by playing around until it sounded right) to mono on master channel for other reasons (maximum impact of low freq's for dance music, mono subs in clubs, etc.)

3) I create most of my synth patches from scratch. One oscillator, LFO, delay, etc. at a time, so I'm not at the mercy of pre-packaged sounds with phase issues. Most of the pre-packaged stuff I use (like Scarbee bass), I use in mono. The only exceptions are some piano patches, etc, but they're rarely the main ingredients, and most don't have these issues. The issues I'm having are mainly with patches I create from scratch.

4) What is a sensible rule of thumb for when to use mono rather than stereo reverbs, and sensible rules about HOW to use stereo reverbs to get maximum compatibility? What are some sensible guidelines for the creation of stereo reverb patches that avoid (or at least minimize) compatibility issues? What safeguards can be taken when using premade patches like IR's? Anyone using the fusion IR's in Reverberate? I'm not, but looking into it, and wondering whether they might cause issues.

5) Is there anything that can be done to increase mono compatibility for panned delays, ping-pong delays, etc?

6) I'd really appreciate if someone could walk me through exactly what to do with the synth situation since I'm doing more remixes now, creating all my own sounds, and it's an increasingly important issue. Basically, like I said, typical situation is that I use something like Omnisphere that has two different sound sources. I choose a different version of the same sound (saw waves sampled from two different machines), pan them hard R and L, and detune so one is up 10 cents or so, and the other down by the same amount. Now, I have been under the (mistaken) impression that this was ok since there is a signal on both sides, and I'm not widening "artificially". Clearly, I've been wrong in that assumtion, though, as there is a significant drop in level when summed.

So... the question is what to do about it. Am I correct in assuming that there will be phase cancellation no matter what I do, and that the best I can do is to experiment with phase flips and delays per side until the cancellation is in a frequency range that does the least damage? Right now, I've got one patch (in a remix I was getting ready to release until I did mono check) where the fundamental nearly drops out. I'm guessing this is because the wave forms were essentially 180 degrees out of phase within the sample bank, correct?

So, should I be dialing it in as described above, or should I be using the Haas effect as prescribed earlier... essentially bypassing significant phase issues by having enough of a delay between them that all key frequencies are protected from collapse? If so, should I then shift the overall stereo image to adjust for the perceived shift due to Haas effect?

Something else? Just bring them in to 75% width or something like that to lessen the damage?

In case anyone hasn't guessed by the questions, I'm not really interested in just mixing in mono. Personally, I love all the 60's recordings where all the guitars are on one side, keys on the other, etc. I wouldn't find them nearly as engaging in mono. Mixing in mono is the easy way out. What I am interested in is developing a more nuanced "best of both worlds" approach since the vast majority of listeners are listening in stereo. We've lived through 60's lounge recordings, modern pop production, the "Wall of Sound", and many other fads. I'm looking to borrow from any and all production styles that have something to offer just like I borrow influences from other musical styles and eras in my music.

I'm not obsessed with super wide effects or anything like that, but I do like to hear some kind of movement between speakers so image isn't static. I also have no particular interest in trying to recreate a live soundstage or anything like that. I'm a studio musician, and treat studio recordings as their own entity free from the constraints of trying to recreate some artificial "reality." Basically... If it sounds good (in mono as well as stereo), do it.
Old 4th February 2017 | Show parent
  #10
Lives for gear
 
Ain't Nobody's Avatar
 
🎧 10 years
Quote:
Originally Posted by Waltz Mastering ➑️
Have you run commercial tracks through the button, and found the results aren't similar?
No. I suppose it's possible I'm being too hard on my own stuff. Still, just because someone else's mix collapses doesn't mean I'm OK with mine doing so.

Point taken, though. I don't have much perspective on what is "acceptable" and what's not. Looking to explore that here and gain some collective wisdom.

I've been reading up last day or so about ME's listening to other commercial releases with side info solo'd, etc. Sounds like valuable research, and the kind of thing I could get into and gain insight from over time, but I've got otherwise finished mixes that need attention right now, so time management dictates practical solutions first and philosophical insights second.
Old 4th February 2017 | Show parent
  #11
Lives for gear
 
Ain't Nobody's Avatar
 
🎧 10 years
Quote:
Originally Posted by minor_glitch ➑️
You can use a phase correlation meter to help you troubleshoot the problem. If the bar is constantly going to the left of zero, you've got problems. Something like Flux:: sound and picture development can help (look at Phase near the bottom of the plugin). Just start disabling/muting things until you find what's causing it.
Yeah, I've started using one. I'm switching back and forth into mono until I get this figured out that the visual representation isn't really needed at the moment, though. So far, everything it shows me is something I can hear when I switch. Still watching it though as I assume the more familiar I become, the more I might learn something new I wasn't paying attention to.
Old 4th February 2017
  #12
Registered User
 
The obvious thing here is to post the track.
Old 4th February 2017 | Show parent
  #13
Lives for gear
 
voodoo4u's Avatar
 
🎧 10 years
Quote:
Originally Posted by Ain't Nobody ➑️
Good stuff. Keep it coming.

I should have clarified a few things in original post (and asked a few more questions), but since there are already responses, I'll try here:
Wow, so many questions. I'll take a shot at answering a few to the best of my ability. Maybe somebody else can get the rest.

1) In some DAWs the -3db setting can be changed. It's not always -3. It should be remembered that if you have two (L-R) completely in phase signals panned hard left and right at 0db and you pan them to center and reduce each by -3db, the combined sum should in theory bring you back 0db. If you're finding that when you sub to mono, your overall level drops, it's probably an indication that you have more out of phase info than in phase.

5) As I mentioned earlier, as long as your delays are greater than 20ms or so, when they're summed to mono the signal should be perceived as a delay rather than a phasey, level reducing smear.

Unless you're running a pure sine wave identically into both channels, there will always be some phase difference. Even one mic on a mono source will introduce acoustical phasing as soon as sound hits a boundary, but that's a completely separate issue that doesn't apply when were addressing stereo. Where the information is in phase, the info will sum and get slightly (between 0 & +3db) louder (minus the pan rule). Where the material is more out of phase than in, the material will subtract and get slightly quieter. Of course, the signal (music) is so incredibly harmonically complex, we don't always perceive it this way. We just hear it as overall quieter or the same (not necessarily louder as our -3db pan rule comes into play).

Now the big question you really need to ask yourself is: DOES ALL THIS REALLY MATTER? I'm an old guy too and I was trained in a time when mono compatibility was a big deal. It had to do with all AM radio stations broadcasting in mono across the airwaves and mono playback devices. Anybody trying to get airplay had to be very careful to be sure their mixes were mono compatible. Now, most radio has gone digital and stereo, all downloadable music is in stereo and almost every single playback device, including our TVs are stereo, virtually eliminating the need for mono compatibility. But sometimes old habits die hard.
Old 4th February 2017 | Show parent
  #14
Lives for gear
 
Ain't Nobody's Avatar
 
🎧 10 years
Quote:
Originally Posted by voodoo4u ➑️
Wow, so many questions. I'll take a shot at answering a few to the best of my ability. Maybe somebody else can get the rest.

1) In some DAWs the -3db setting can be changed. It's not always -3. It should be remembered that if you have two (L-R) completely in phase signals panned hard left and right at 0db and you pan them to center and reduce each by -3db, the combined sum should in theory bring you back 0db. If you're finding that when you sub to mono, your overall level drops, it's probably an indication that you have more out of phase info than in phase.

5) As I mentioned earlier, as long as your delays are greater than 20ms or so, when they're summed to mono the signal should be perceived as a delay rather than a phasey, level reducing smear.

Unless you're running a pure sine wave identically into both channels, there will always be some phase difference. Even one mic on a mono source will introduce acoustical phasing as soon as sound hits a boundary, but that's a completely separate issue that doesn't apply when were addressing stereo. Where the information is in phase, the info will sum and get slightly (between 0 & +3db) louder (minus the pan rule). Where the material is more out of phase than in, the material will subtract and get slightly quieter. Of course, the signal (music) is so incredibly harmonically complex, we don't always perceive it this way. We just hear it as overall quieter or the same (not necessarily louder as our -3db pan rule comes into play).

Now the big question you really need to ask yourself is: DOES ALL THIS REALLY MATTER? I'm an old guy too and I was trained in a time when mono compatibility was a big deal. It had to do with all AM radio stations broadcasting in mono across the airwaves and mono playback devices. Anybody trying to get airplay had to be very careful to be sure their mixes were mono compatible. Now, most radio has gone digital and stereo, all downloadable music is in stereo and almost every single playback device, including our TVs are stereo, virtually eliminating the need for mono compatibility. But sometimes old habits die hard.

As to whether it matters, I've been in clubs where the PA was in mono, been to countless gatherings where music was plugged into one of those all in one single speaker PA setups they sell in every Sam's Club (and Wal-mart?), been in stores or other places where the system was either in mono, or the speakers were placed so far apart that you could only hear one at a time, heard people play music on laptops that were in mono (all of the student laptops at my school are mono), etc. There are also millions of times a day that audio is played on devices that may technically be stereo, but the speakers are right next to each other and essentially summed by the time they reach the listener who is thousands of times further away from either speaker than the speakers are from each other. Millions more times daily, audio is summed to a single mono speaker via bluetooth. Actually, hardly a day goes by at my school that I don't witness at least one student or staff member effectively listening in mono whether they're aware of it or not.

I personally care about the real world situations more than the theoretical situation of an audiophile listening on a rarified system in a rarified environment. Bottom line is that it does matter in countless daily situations, but dialing in where to draw the lines on compromises will, of course, be a matter of taste.

As for panning law, that's where I'm digging in right now to make sure I'm not causing a problem with my summing method. It's ultimately what happens in the real world that matters, not what happens in my daw, but if my own panning law is causing it to sum DIFFERENTLY than it would in the real world, then that would be a monitoring error. Currently, I have it set to -3db compensated. Burning multiple output files right now and checking daw sums to mono vs stereo burns summed outside daw to see if I'm causing issue via panning law.
Old 4th February 2017
  #15
Gear Guru
 
drBill's Avatar
 
🎧 15 years
One thing to (possibly) consider -

Mono is the new stereo. What I mean is, instead of all these "stereo" patches with reverb, chorusing, delays, widening tricks, etc., switch out to mono patches. Prophet5 - mono. Jupiter 6 - mono. Super Jup - Mono. Mini Moog - mono. All the classics - mono. Ditch the big "stereo-zing" wide effects, and start printing mono tracks with no effects. Pan them wide of center. Keep effects out of your patches until mix time where you can control them better. Watch how wide you mix starts sounding - and it's almost a guarantee it will translate to mono perfectly. (Verb being the possible exception, but I kinda expect that from verb anyway)

A bunch of stereo patches = a mono sounding mix to me. A bunch of mono patches panned wide = a very stereo vibey mix for me.
Old 4th February 2017 | Show parent
  #16
Lives for gear
 
voodoo4u's Avatar
 
🎧 10 years
Quote:
Originally Posted by Ain't Nobody ➑️
As to whether it matters, I've been in clubs where the PA was in mono, been to countless gatherings where music was plugged into one of those all in one single speaker PA setups they sell in every Sam's Club (and Wal-mart?), been in stores or other places where the system was either in mono, or the speakers were placed so far apart that you could only hear one at a time, heard people play music on laptops that were in mono (all of the student laptops at my school are mono), etc.
Ok, so stepping out of the world of theory and into practical application. For me, it comes down to tracking and mixing techniques and habits. Of course, like any engineer who's been at it a while, I've developed my own "what works for me" rules of mixing.

1) Low frequency instruments are almost always mono. Almost without exception for me, the kik is dead center mono, the bass is dead center mono the snare is dead center mono (kik and snare leakage into the overheads will contain stereo info, but I'm very careful about making sure my drums are acoustically in phase. Hats get panned slightly, cymbals are always stereo mic'd.

2) Piano is almost always stereo (checked for phase coherency)

3) Main vocal is always dead center and I never put an effect on the vocal that would put it out of phase or cause it to dip when subbed to mono.

4) More and more often these days, when I'm putting instruments into a dense band mix, I'm recording other instruments in mono ie. acoustic guitars, dobro, violins and panning them gently to widen my mix and give the illusion of a live off the floor mix. This way, it maintains mono compatability. If the mix is less dense, I'd probably record some of these instruments in stereo.

5) If I really feel the need to get a wide stereo image on things such as BG vocals or crunchy electric guitar, I'll often double track to avoid complete mono incompatibility. I'll most often pan in 3 O'clock/ 9 O'clock or so, so that the mix won't collapse.

6) If something really needs to be stereo (if it's going to be prominent), I'll often record it MS so I can have control over how it sits in the mix.

There are dozens of little things I think of when tracking and long before I get to the final mix that will ensure mono compatibility. Some might think my mixes are kind of boring, but I'm OK with that. They certainly end up being punchier for it.

Last edited by voodoo4u; 4th February 2017 at 09:53 PM.. Reason: more info
Old 4th February 2017 | Show parent
  #17
Gear Addict
 
🎧 5 years
Quote:
Originally Posted by Ain't Nobody ➑️
1) Anything panned hard to one side or the other will sum into single channel at -3db, right? I think I've been somehow assuming (without ever really thinking it through) that it's a non-issue since the other side was at 0db. My thinking was that the summed mono signal is at -3db from the hot side, BUT, it was only going out one speaker originally. Now that it's summed, it's playing on both speakers. Adding one speaker at same input level yields +3db, so it's a wash. Down 3db from summing, and back up 3db from playing out both speakers instead of only one when it was stereo. I'm thinking now that that assumption is likely incorrect and at the heart of some of these issues. If so, I guess I've been missing a fundamental concept that ANYTHING panned to the side automatically loses energy upon summing, and that there's a direct tradeoff between width of panning and mono level, so if I'm going to pan anything at all, I need to ask myself how far out I want it to be while remembering that the further out it goes, the closer it gets to it's -3db summed limit. I'm sure I must have learned all this stuff when I started out, but this is what happens when there's 25 years between when you learn something, and when you put it into practice. Damn, I'm old, but hey, it's never too late to reteach an old dog old tricks.
I'm pretty sure your problem has nothing to do with pan laws or the basic summing fundamentals above. The way pan laws are set, and all the mathematics you've described regarding stereo vs. mono monitoring directly impact your workflow, but not the sound.

Try starting with two simple, mono sources and no effects. Pan one source dead center and the other hard left. Adjust levels until they sound like the same volume.

Now start panning the second source from hard left slowly through center and on towards the right. Depending on your pan law settings, the relative volumes of the sources may be held constant, or may change. If your pan laws haven't maintained the volume relationship for you, you'll turn the levels up or down as needed to get the right relationship between them. So the pan laws effect your workflow, but not the ultimate sound, because you'll adjust the mix until it sounds the way you want. When summing to mono with simple, mono, uncorrelated sources like this, the relationship will stay the same.

If summing to mono changes the relationship, something is wrong. I'd guess in your case that it's heavy phase cancellation on correlated stereo sources, which most of your remaining questions delve into. Creating these "artificial" stereo sounds isn't my strong suit, so I'll leave that subject alone.

One last thing: I can imagine one scenario where your pan laws could reek havoc on your mono checks. That's if the DAW software is doing some trickery to go mono instead of literally summing the two channels as it should. For example, maybe it's internally switching all of your pans to center, effectively making everything mono, so that it can save that one additional summing step. Programmers love efficiency! If that were the case, your pan law settings could have a major impact on the result. It seems unlikely to me that a DAW would be designed this way, but you never know...

I was shocked, really and truly shocked, that Studio One (and apparently most DAWs) doesn't have the equivalent of a PFL solo button. When I hit solo, it mutes everything but my solo selection and it's output bus. In most cases this is only a minor inconvenience, but occasionally it's a nightmare. If I'm recording someone doing overdubs and I solo their mic to listen more closely, that mutes all of the other channels, which shuts down the performer's monitor mix so he can't hear the tracks he's playing along with. This is preposterous. If I have a pair of mics in mid/side configuration routed through an M/S decoder in the mix, I can't quickly solo either individual mic to hear the raw sound because they're only available through the decoder and all other inserts in their output chain. I have to jump through lots of hoops and undo half my mix just to hear an original track raw. It's insane. Anyway, these tangential rants are just to illustrate that DAW functions don't always work the way you'd expect.

If it is anything to do with deceptive panning behaviour from your DAW, then your experiment with rendering stereo tracks and then summing them to mono outside of the DAW should reveal the truth. Good luck!
Old 4th February 2017 | Show parent
  #18
Lives for gear
 
Ain't Nobody's Avatar
 
🎧 10 years
Quote:
Originally Posted by drBill ➑️
One thing to (possibly) consider -

Mono is the new stereo. What I mean is, instead of all these "stereo" patches with reverb, chorusing, delays, widening tricks, etc., switch out to mono patches. Prophet5 - mono. Jupiter 6 - mono. Super Jup - Mono. Mini Moog - mono. All the classics - mono. Ditch the big "stereo-zing" wide effects, and start printing mono tracks with no effects. Pan them wide of center. Keep effects out of your patches until mix time where you can control them better. Watch how wide you mix starts sounding - and it's almost a guarantee it will translate to mono perfectly. (Verb being the possible exception, but I kinda expect that from verb anyway)

A bunch of stereo patches = a mono sounding mix to me. A bunch of mono patches panned wide = a very stereo vibey mix for me.
I'm not using any widening on any patches other than as described panning of detuned synths. Synth patches created from scratch, and most others mono. I don't have fx until mixdown. I'm not using factory presets with built in fx.
Old 4th February 2017 | Show parent
  #19
Lives for gear
 
Ain't Nobody's Avatar
 
🎧 10 years
Quote:
Originally Posted by voodoo4u ➑️

1) Low frequency instruments are almost always mono. Almost without exception for me, the kik is dead center, the bass is dead center the snare is dead center (kik and snare leakage into the overheads will contain stereo info, but I'm very careful about making sure my drums are acoustically in phase. Hats get panned slightly, cymbals are always stereo mic'd.
Doing that... except generally using synthetic snares, so no issues. Is sometimes use AD or somilar, but no real phase issues with drums.

Quote:
Originally Posted by voodoo4u ➑️

2) Piano is almost always stereo (checked for phase coherency)
Using piano patches. Haven't checked them specifically, but I think they're ok.

Quote:
Originally Posted by voodoo4u ➑️

3) Main vocal is always dead center and I never put an effect on the vocal that would put it out of phase or cause it to dip when subbed to mono.
Hmmm... usually doing this in verses, but one of my absolute favorite things to do is to put single voice in center for verse, then have dramatic shift by having wide tuned doubles (real doubles, not software shifted) in the chorus. Having some issues with this too, but it's definitely one of the techniques I want to keep if I can tame it a bit.

Quote:
Originally Posted by voodoo4u ➑️

4) More and more often these days, when I'm putting instruments into a dense band mix, I'm recording other instruments in mono ie. acoustic guitars, dobro, violins and panning them gently to widen my mix and give the illusion of a live off the floor mix. This way, it maintains mono compatability. If the mix is less dense, I'd probably record some of these instruments in stereo.
More or less doing this. Most stuff is from vst's, but I do record some percussion, guitars, etc. It's always mono, and generally some subtle panning to get more of a "conversation" going where one instrument answers another from slightly different positions.

Quote:
Originally Posted by voodoo4u ➑️

5) If I really feel the need to get a wide stereo image on things such as BG vocals or crunchy electric guitar, I'll often double track to avoid complete mono incompatibility. I'll most often pan in 3 O'clock/ 9 O'clock or so, so that the mix won't collapse.
Doing this, but probably pushing it too far. May be tailoring vocals too much (increasing phase issues by making them too similar on each side). I also tend to pan them wide. I've also got to start digging into any chorus or other fx I'm using that may be exacerbating issues here.

Quote:
Originally Posted by voodoo4u ➑️

6) If something really needs to be stereo (if it's going to be prominent), I'll often record it MS so I can have control over how it sits in the mix.
I'm not actually recording more than 1 mono channel at a time. Any stereo voices are coming from vst's (like piano or organ patches, etc.)
Old 4th February 2017 | Show parent
  #20
Lives for gear
 
Ain't Nobody's Avatar
 
🎧 10 years
Quote:
Originally Posted by Ebeowulf17 ➑️
I'm pretty sure your problem has nothing to do with pan laws or the basic summing fundamentals above. The way pan laws are set, and all the mathematics you've described regarding stereo vs. mono monitoring directly impact your workflow, but not the sound...

One last thing: I can imagine one scenario where your pan laws could reek havoc on your mono checks. That's if the DAW software is doing some trickery to go mono instead of literally summing the two channels as it should.
I've got it at -3db compensated for workflow purposes so I don't have to keep adjusting levels when I pan something to get equal loudness, but questioning whether hitting mono button within master channel is really doing the same thing as, say, using a summing cable, or summing within interface.

In other words, not entirely sure if the "sum to mono" function when pressing the button is exactly the same process regardless of panning law, or if panning law affects how it interprets that command.
Old 4th February 2017 | Show parent
  #21
Gear Addict
 
🎧 5 years
Quote:
Originally Posted by Ain't Nobody ➑️
In other words, not entirely sure if the "sum to mono" function when pressing the button is exactly the same process regardless of panning law, or if panning law affects how it interprets that command.
Yeah, it's a fair question. I'd certainly hope it's exactly the same process, but you can't take anything for granted.
Old 4th February 2017 | Show parent
  #22
Lives for gear
 
Ain't Nobody's Avatar
 
🎧 10 years
Quote:
Originally Posted by 300lbsofjoy ➑️
The obvious thing here is to post the track.
It's actually a bunch of things noticed on a number of tracks, but here's something that is clear and simple illustration of the synths I was discussing. Synth is two similar but not identical wafeforms with one panned hard L and detuned by 10 cents, and the other panned hard R and detuned by the same amount in the opposite direction.

Delay is ping pong panned hard R and L.

Verb is stereo 480L patch (Relab).

Unless I screwed something up, the "Mono" one should be just single mono file summed by selecting mono output for Logic.

"Mono button" is dual channel output, but with "mono" button engaged on gain plug on output channel.

"Stereo" is the normal stereo output.

"Summed outside Logic" was output in stereo from Logic, then imported into QT and summed to mono there (to test for panning law or other influences upon summing within Logic)


Haven't done full analysis yet, but quick listen certainly seems to indicate that pressing the mono button (outputting mono signal on both tracks) is lower in level than burning a mono file and then playing over both speakers.

Also of note is the loss of many high frequencies in any mono summed situation. I guess this is since the fundamental waveforms are probably nearly exactly alligned, but the waveforms are not quite identical and slightly detuned, so it would be only some of the overtones that truly cancel.

Also noticing a dip in ping pong delay upon summing, but since it's mainly the high frequencies I hear defining the delay, it's hard to tell how much of that is a difference in delay level, and how much is a dip in the synth sound overall in the higher frequencies.

edit - OK, now I'm really confused. Mono button is clearly NOT doing the same thing as summing externally, but to make it even more confusing, the synths mono and synths summed outside SHOULD be identical... but they're not. WTF is going on? Summing in QT rather than Logic loses less of the high frequencies, and something else is different (loses less verb maybe) Can't account for that. Anyone got an explanation? (and a CORRECT way to monitor how mix will sound when my stereo mix is summed out there in the real world).
Attached Files

synths mono button.wav (2.52 MB, 14596 views)

synths stereo.wav (2.52 MB, 14510 views)

synths mono.wav (1.26 MB, 14328 views)

Synths Summed outside Logic.wav (1.26 MB, 10032 views)

Old 4th February 2017 | Show parent
  #23
Lives for gear
 
voodoo4u's Avatar
 
🎧 10 years
Quote:
Originally Posted by Ain't Nobody ➑️
It's actually a bunch of things noticed on a number of tracks, but here's something that is clear and simple illustration of the synths I was discussing. Synth is two similar but not identical waveforms with one panned hard L and detuned by 10 cents, and the other panned hard R and detuned by the same amount in the opposite direction.
Yeah, there's something about this sample (synths stereo) that's causing it to be almost 90 degrees out of phase pretty much all through the frequency spectrum but particularly the low end. It's not the delay and, although the rev is wide stereo, it's not the rev. The way I can tell is by summing the track to mono and flipping the polarity on one channel back and forth. It's in the programming. It could be in the detuning.

Listen to Dr Bill's advice and go back to a mono source, add rev and delay and watch how much better it'll sit in a mix.
Old 4th February 2017 | Show parent
  #24
Gear Guru
 
drBill's Avatar
 
🎧 15 years
Quote:
Originally Posted by voodoo4u ➑️
Listen to Dr Bill's advice and go back to a mono source, add rev and delay and watch how much better it'll sit in a mix.
This ^^^^
Old 4th February 2017 | Show parent
  #25
Lives for gear
 
Ain't Nobody's Avatar
 
🎧 10 years
Quote:
Originally Posted by voodoo4u ➑️
Yeah, there's something about this sample (synths stereo) that's causing it to be almost 90 degrees out of phase pretty much all through the frequency spectrum but particularly the low end. It's not the delay and, although the rev is wide stereo, it's not the rev. The way I can tell is by summing the track to mono and flipping the polarity on one channel back and forth. It's in the programming. It could be in the detuning.

Listen to Dr Bill's advice and go back to a mono source, add rev and delay and watch how much better it'll sit in a mix.
Oh, I don't doubt in the least that many of these issues can be overcome by using mono sound sources, but that's defeating at least some of the purpose. I also get that this a bass sound, and as such, it wouldn't be a bad idea to pull the panning into center and let any phase cancellation happen within a mono patch, but for other synth patches I create that are not basses, the concept remains that the goal is to have some width in stereo and also have it not collapse in mono.

Ultimately, the question is how much of the stereo mojo can be salvaged, and how much is a fool's errand if mono compatibility is considered? Do I use wide panning and detuning on every sound source? Of course not, but I would at the very least like to dial in exactly what I can and can't get away with when designing synth patches

I've clearly underestimated how problematic some basic things like wide panning can be in getting consistency from mono to stereo, but I'm trying to suss some sort of rule of thumb about what's problematic. For instance, if it's the detuning causing issues, can some delay on one side help?


and... what the heck is going on with every file coming out different? Using mono button seems to be dropping level considerably, but setting output to mono isn't, and outputting stereo input to mono output is somehow working differently inside Logic than QT. Is it a panning law thing? Just that the mono button is automatically dropping the level by 3db to avoid overs while the other methods are not and creating clipping? Something else? Until that bit is figured out, I can't trust anything I hear when summing to check compatibility.
Old 4th February 2017
  #26
Lives for gear
 
voodoo4u's Avatar
 
🎧 10 years
I get that having a big, fat, stereo synth sound can be addictive. There's no doubt that it sounds great by itself. But when it comes to mono and mono compatibility in a mix, it just doesn't work. What you've essentially created in your patch is the Haas effect and the delay and differences are way under 20ms (I think the Haas effect becomes a perceptible delay over 15ms actually, but I'd have to read up on it again) and clearly in the territory of phase cancellation in mono. With mono compatibility it's a balancing act. You can't have both too much stereo information (phase/time) and perfect mono compatibility.

I had a feeling you weren't pickin' up what I was layin' down go back and read my posts again and let it sink in deeply.

Last edited by voodoo4u; 4th February 2017 at 11:29 PM.. Reason: more info
Old 4th February 2017 | Show parent
  #27
Lives for gear
 
Ain't Nobody's Avatar
 
🎧 10 years
Quote:
Originally Posted by voodoo4u ➑️
Yeah, there's something about this sample (synths stereo) that's causing it to be almost 90 degrees out of phase pretty much all through the frequency spectrum but particularly the low end. It's not the delay and, although the rev is wide stereo, it's not the rev. The way I can tell is by summing the track to mono and flipping the polarity on one channel back and forth. It's in the programming. It could be in the detuning.
Experimenting with it right now. I think I get now the correlation with phase and meter. +1 is 0 degrees out of phase, -1 is 180 degrees out, and 0 is 90 degrees out, right?

Looks like it was the actual waveforms in this case. Using the exact same waveform on both sides and turning off the detuning pegs the meter at 1, but even turning the detuning back on keeps it fairly close to 1 at all times.

Will retry experiment using waveforms that are checked to not have inherent phase issues with each other. Looks like good news in that the detuning and panning itself doesn't seem to be a total kiss of death.

And yes, I understand and appreciate that it's a "you can't have your cake and eat it too" situation... just coming a little late to the party trying to figure out where in the spectrum I'm comfortable staking a claim. If anything, I think I'll want to err a bit on the side of being compatible, but I'll need to understand the nuances more fully before I can walk the line with skill and purpose.
Old 5th February 2017
  #28
Lives for gear
 
stinkyfingers's Avatar
 
🎧 10 years
turn pan law off (set to 0).
Old 5th February 2017 | Show parent
  #29
Lives for gear
 
Ain't Nobody's Avatar
 
🎧 10 years
Quote:
Originally Posted by stinkyfingers ➑️
turn pan law off (set to 0).
I will as an experiment to see how it's affecting summing within Logic, but this is an otherwise finished mix with 50+ elements in place, so that's going to screw up all kinds of elements unfortunately.

Still... even if that solves the mystery of the drop when summing with button, it doesn't explain why I get different results summing upon bounce in Logic vs summing in QT after bouncing stereo in Logic. Still have no clue what's going on there (or more importantly what that means about how to monitor mono properly while working in logic). Not sure if my duet has a mono button. Will check.
Old 5th February 2017
  #30
Lives for gear
 
stinkyfingers's Avatar
 
🎧 10 years
fwiw your "mono" track contains only the left side...so it did not collapse L & R to mono, but only output the left channel.
hence half of it literally does disappear...
could be a setting or routing option ?

Last edited by stinkyfingers; 5th February 2017 at 06:10 PM..
πŸ“ Reply

Similar Threads

Thread / Thread Starter Replies / Views Last Post
replies: 125 views: 19047
Avatar for Deleted 0ea68bb35a726d0
Deleted 0ea68bb35a726d0 2nd February 2021
replies: 209 views: 47207
Avatar for ALESIS-ION
ALESIS-ION 6th December 2016
replies: 604 views: 71846
Avatar for t_tangent
t_tangent 5 days ago
replies: 9051 views: 735371
Avatar for TrenchRun
TrenchRun 8 hours ago
Topic:
Post Reply

Welcome to the Gearspace Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…

Forum Jump
Forum Jump