A rolling slew rate graph analyzer plugin for Windows (64 bit VST2)



Download link: https://jelstudio.dk/VSTplugin_SlewRateGraph/index.html

The plugin has no user-controls, so you just load it onto the track where you need to monitor the slew-rate.

It takes a stereo-input (In Reaper it also accepts a mono-input) and will show left-channel in white color and right channel in red color (When the signal is mono the 2 graphs overlap and become a light shade of red)

It has a 5 second window.

It has 2 static mustard-yellow lines on the screen.
One is showing the typical ceiling for music mastered in the 70s/80s and the other music mastered in the 90s (They just show a general ball-park of where music from the particular decade tends to max out, so it's not something that's written in stone)

Music mastered after year 2000 will go off-scale-high (You don't need this analyzer if you intend to have slew-rates that are that fast)

It is inspired by Airwindows' "METER" plugin (The slew-rate lane on that plugin, which I personally find very useful) but I defined the 2 typical ceilings using a range of CDs from my own collection (No vinyl, but CDs with music from the 70s, 80s, 90s and beyond year 2000. It's actually kind of crazy how it keeps going up, decade after decade, and isn't really helping the music sound any better in my subjective opinion. So my recommendation is that you at least keep your slew-rates inside the screen of this plugin, and even better if you keep your fastests moments around the center, but it's obviously up to you )

It doesn't affect the audio at all. It's just a visualizer.

And, if you want to hear an example of what modern music sounds like with a slew-rate limited to the 70s/80s ceiling (Which I used my other plugin Stylus to do. It uses the default-setting of Stylus), then listen to this music-video:

Thankyou for this - I look forward to staring at it :D

You're welcome

If you get dizzy from staring at it... that's how you know your slew-rates are too fast

On a more serious note; I suspect part of the reason the slew-rates appear faster on post-2000 music, is in part because it's so much louder than music was pre-digital era. It's probably not only because digital allows things to be brighter, though that may also play its part.

And, from my own experience since Airwindows' METER got me focused on slew-rates, there are 2 main ways to deal with it (IF one wants to deal with it of course) : some type of EQ (Sounds cleaner, but leaves the music over-all darker in tone) or some type of dedicated slew-rate limiter/clipper (Which will always cause more aliasing and distortion, but leave the music sounding over-all brighter in tone)

If Chris hadn't made METER, I probably would not have ever spent so much time looking into this phenomenon. But it really does have a greater impact on the sound-quality than it appears to have at first sight.

"Faster isn't always better"... she said

Great music you've done there. Sounds (and looks) wonderful.

and thanks for the plugin.

Yes, thank you muchly, that is a very detailed slew analyzer.

Thanks

I'm going to look into if there is any sense in trying to get the slew-rate to show up for various frequencies (Like the style of a spectrum-analyzer, but for slew-rates instead)

I don't know if having that info for music would be useful though, but I'm thinking it's worth an experiment anyway

Mostly because I'm wondering if there perhaps might be some kind of ideal slew-rate per frequency that it would be useful to target when mastering music.

This current plugin only shows the total slew-rate, which is also what Airwindows METER shows in its middle-lane, and maybe that's enough info, but I'm really fascinated with this metric the more I learn about it (Mostly because since I began controlling it my masters sound so much better than what I was able to achieve before)

Rabbit hole alert!

I entirely get your point though.

Thanks for that warning, and yes I did fall into it but luckily it turned out not to be very deep in this case (Only half a day deep, so that was relatively quick to map out )

I did a test where I split the incoming audio into 5 bands (Where I could change the cross-over frequencies on the fly)

Then I measured the slew-rate independently for each band.

It turns out that when you look at slew-rate this way, you get a very similar appearance to what you can get with a normal spectrum-analyzer set to show peaks at very quick refresh-rates.

So a slew-rate meter divided into frequencies appears not to be particularly useful as a plugin on its own (I mean, it can be*, but Voxengo's "Span" set to show peaks (RT Max, as its called in that plugin) with averaging-time set to its minimum (So the display updates as often as it can) showed pretty much the same thing, so using that instead would be just as useful for an indication of how the slew-rates are at each frequency)

*It more or less showed how much dynamic range there was at each frequency.

To add to the list of ways to deal with slew, the ADPTR Sculpt 'Transient Sculpt' module (reinstalled, loaded in 'Edges-Less' preset, sat back) does the job and seems quite tonally unobtrusive. So ye olde free Flux BitterSweet might also. I guess it makes sense that overenthusiastic transients will send the skew through the roof both in analog and digital worlds.

Good find
I have "ADPTR Sculpt" myself and it's a great plugin.

I use it mostly for expander-stuff, since I find it has a very soft and relaxed sound for such things
(It's been quite a while since I used it last though, but it actually beat Acustica's "Ocean" for me in 2023 on an electronic album I did. "Ocean" sounded very frantic and jittery to me, while Sculpt was somehow less sharp-sounding but still had more punch, being smoother and more pleasant (For that particular album at least, AND with whatever my user-skills were able to drag out of the 2 plugins of course. Maybe other people would have had different outcomes, as GUIs obviously also play a part in what one gets out of a plugin))

But the way you can set-up the way it triggers its 4 modules (And how you can ratio exactly how much it works on the mid vs side-channel at the same time) is just very quick and useful when you need to target specific parts of a sound. Easy to work with.

I don't know how well it can do slew-control though, as I never tried that (Because I never even considered slew-control last time I used it), but it does have that transient-module (With its many options, including one for soft and hard transients, if I remember correctly) so definitely worth checking out

Lol, what I liked about the Edges-Less preset was that it locked me out of the up/down expansion modules because it clearly knows it I'm not safe around things like that.

If you don't have it and is free atm from summat like 'production expert' forum the Klevgrand Richter compressor has a big 'transients' knob which it turns out notably soothes the skew. So today it went on the mix buss and the Sculpt went off to tidy up the bass line.

Back where it started with airwindows, his new dub plate also does a bit of skew deharshing and rather blissfully doesn't have any user controls so I used different mix amounts to one on the bass buss and one on the drums to combine mix balancing with taming the nasties.

The other slew lesson I learned today from having been a bit heavy handed with the Triaz snare random slop sample start etc features is that as a metric for setting the volume level of a snare hit using your plug-ins readout made me make balance adjustments I wouldn't have thought of making.

So then I thought, how do I get my lufs-i mix buss out be as high as possible whilst keeping the slew rate as low as can be?
Slew rate limiters are basically the magic erasers so I set Sloth to stun on the channels and Platinum Slew on the mix buss and then worked my way seeing how much slew I could get away with using and still maintain basic audibility. The kick drum's Sloth started off at 60/52, got moved in to 10/2, ended back out where it started.

Oops this is going on a bit but the other thing I noticed which now I hear in your track is how (I guess because it's so non-slewy) you can use the absence of slew to reverb with confidence.

I use this plugin for the final output slew-rate control

https://jelstudio.dk/VSTplugin_Turnt...Sim/index.html

Regarding LUFS... I've more or less given up focusing on those.

I know Chris (Airwindows) is gonna slap me silly for saying this, but I actually focus on average-RMS on a spectrum-meter when I level the music. I aim for -36 dB FS on a 4.5 dB-per-octave slope (I use Voxengo Span for that, and included a picture of the spectrum for the song I attached in the first post. The upper graph is the maximum and the lower graph is the average, and it's the average graph I put most focus on myself, the maximum I only turn ON occasionally to check if there is something completely crazy going on)



And when I have those levels set, then I just catch stray peaks with a limiter or clipper (Or some combo of that)

When I compare my tracks to actual hit-tracks (You know, stuff like Miley Cyrus and Selena Gomez) I'm always way lower in LUFS than they are (Even a few years back before I came across Chris' METER and WAS trying to hit as loud as possible. I was just never up in the same territory. The earlier stuff I have on streaming is a sad testament to the fact that fighting the loudness-war is a losing game for me So naturally I've given it up)

So I've shifted my paradigm away from LUFS and loudness. In part because METER was a huge eye-opener for me, but also in part because, and I agree completely with Chris on this; modern music sounds way too congested and cramped and almost claustrophobic (There's zero room to breathe in modern super-loud music, and that may work for a 2½-minute single, but once you listen to a radio-station that plays a lot of those it quickly gets too much. Maybe I'm just too old now, but I never listen to music-radio anymore. I think the last time I really listened to FM-radio was back in the 90s)

But anyway, back to LUFS; I think, if you want to 'game' that metric, that you would be best off having zero reverb on any transients and only keep it to sounds that are already naturally non-transient (Transients hardly register in LUFS-meters at all, but reverb-tails do. Also, as you say, reverb on transients are generally much less transparent-sounding than on flowing tones, which obviously makes perfect sense given transients leave no sound for the reverb to hug on to and piggy-back on or hide behind, so to speak. But obviously reverb should be used wherever it makes sense for the music and personal taste)

I love dark and narrow reverbs myself and almost always keep them down deep. I find that helps too, but that's probably just more a matter of my personal taste.

Oh yeah man totally true!!
I actually went on a bit of a spending spree to find different hardware that lets me slow down the slew rate in mastering and its amazing how true this is. Slow slew rate is so much more pleasant to listen to. And its possible to have all the transient detail you want without the fatigue.

I do indeed have your other plugin, but I didn't really mean gaming the lufs metric as such, more when I take a look at the reading every once in a while once a tranche of mixing has been done to see it higher rather back down at -24 lol

Sorry I mis-read you (English isn't my native language, so mistakes do happen )

Yes, the slew-rate and final volume of the song does fight against each other by the very nature of how it all works.

Raising the volume automatically also raises the slew-rate, since obviously higher volume requires higher slew-rates.

So balancing those 2 is indeed, pun intended; a balancing-act

And even worse; when an end-listener turns up the volume on their system, they also could run into slew-rates that are too fast for their system to handle without distorting.

I don't know how quick modern ear-plugs, such as air-pods, are when it comes to how fast they can slew without distorting, but since I'm sure a lot of people do listen to music on such devices (And probably often on very high volumes too) I do think it is a potential weak-spot that it's easy to over-look when trying to balance ones music as efficient as possible.

I just recently shifted from dynamic head-phones to planar, and one of the most immediate things I noticed was how much better the planar ones are at transients (And thus faster slew-rates)

On the dynamic ones it always sounded kind of soft (But distorted, as I can hear now that I have the new head-set to compare on), and is very possibly one reason I never really heard my own music as being too bright.

On the planar ones I can much more clearly hear the 'ticks' of fast slews. They don't distort, as the dynamics do, but instead play the actual ticks which come across as increased sharpness and brightness.

And if air-pods are anything like dynamic head-phones (Which I'm guessing they are, but I don't know, I'm just speculating about them here as I've never tried air-pods myself) then they may mask faster slew-rates as brightness-distortion.

And that's perhaps an entirely different can of worms; the stuff people listen to music on, which may be significantly worse than what we use to create music on

I saw a youtube-video, not long ago, where this person was saying that his recommended way to master music was to master it to the lowest common denominator. So basically to master it so it sounds good on a poor system, because if a poor system can handle it without distorting, then a good system can too.

I don't know if that's really the best advice (Personally I always just try to make it sounds as good as I can on my own equipment, and my own ears, not really worrying about what better or worse equipment other people might hear it on, or how much better or worse their hearing might be. Mine tops out at around 13 kHz, so anything above that is something I can't really control reliably anyway, even if I used the best available technical equipment, and therefore prefer to tug away by low-passing it slightly to make sure it doesn't do anything too obscure that I'm not aware of), but I can certainly see a kind of logic in such an advice; if it works on a low-end system, then certainly a high-end system will also be capable of handling it.

But it's a double-edged sword really, and I'm not sure there's an easy fits-all solution (Now, since you mentioned rabbit-holes earlier, THAT's definitely also a rabbit-hole one could easily disappear in for a very long time )

It definitely makes sense to check your mixes on as many systems as you can. I've heard a mix of mine that I did on a home midi-fi ghetto blaster played out in a biggish auditorium and yeah, no, lesson learned. I personally need decent monitors in my life.

The stepdaughter let me have a listen on her planar headphones a couple of years ago and I was like wow on the transients, so you have tipped me over the edge on getting a pair lol

Agree on the monitors!

And I don't think you'll regret getting a pair of planars.
I was a bit worried about weight before I decided to finally get a pair (I've been, should I? Shouldn't I? for a while, but recently finally decided I wanted to give them a chance), as people seem to say they're very heavy, but to me they don't really feel heavier

Yeah I got the Sundara closed backs (£119 for next day delivery, AliExpress even cheaper) and the solid chunk of birch wood makes them look heftier than they actually are. Very decisive soundstage was my first impression.

Congratulations with those

One note I read online about closed-back planars; be careful not to create an excessive vacuum or over-pressure inside the cups (Either by pressing the cups quickly against each other or dragging them quickly from each other if they are closed up to begin with)

Apparently such flow of air in or out of the cups can be bad for the drivers if the pressure/suction is strong enough.

I don't know if it's a thing, and people write all kinds of nonsense on the internet, but better safe than sorry I guess

2 yr old daughter once stuck a slice of cake in a vhs player so I can see that conceptually happening though perhaps because they come from the hifi rather than the studio world the Sundara earpads are quite plump cushiony things that it seems hard to imagine ever approaching vacuum status.

Anyway, back to slew. Here's what happened in course of five days when using your plugin to rabbit hole on the causes of slew and how to ameliorate them working with prerecorded material and in-daw stuff. Some instrumental reworking also occurred, but not of a hugely drastic variety.

My music is kinda leftfield, gaming the slew rate seems like a shortcut to sweetening the bitter pill lol

v1.



v2.



tl:dr — this plugin rocks

I got more experimental in v2 with the Triaz chance but used its handy randomize sample feature to go for drums that sounded impactful but were low on slew and from the v1 Vital stab sound I learned that transposing samples upwards increases their slew, and vice-versa. And it all seems to operate on a complex logarithmic rather than linear basis. And then there's envelopes. And filters.

The slew increases you get from global volume gains should ideally be 'gain on a wire' predictable, so my feels is that if you aim to curtail/compress it to as low a level as possible early doors that gives you more wiggle-room at the finalizing stage.

There is something about the slot-opening in VHS machines that just invites such behavior from curious minds

It was this reddit-post where I read it: https://www.reddit.com/r/headphones/...ic_headphones/

I'm pretty sure cake inside the cups wouldn't be too much fun either though



I over-whelmingly prefer V2 to V1

V1 sounded very harsh/sharp on my monitors, so whatever you did to arrive at V2 I definitely think those were the better decisions.

I found the drums were also much more pronounced in V2, the kick had more weight, and even the snare too, where they were close to invisible, or masked too much behind everything else, in V1.

To my ears V2 sounds balanced and with good contrast, where V1 is much too piercing and is more like a wall of sound (A thing I've been struggling with myself in a lot of my music. Especially before I came across Airwindows' METER)

I also liked the video I must say. I don't smoke myself, but I do eat cake and very much prefer people who live and let live

Very interesting observations in this thread.

The hardware Elysia MusEQ - the warm button does this - it slows down the slew rate. I never really messed with it but now I'm going to out of curiosity for the results.

Also I have two hardware MusEQs. I'm going to try stacking them and turning the warm on both and see how much more that affects the slew rate.

Rabbit hole alert, indeed.

Thanks for the plug, JEL.

Oh man wow.
You must let me know how that goes!
Send me some audio when you have experimented if you can!

One of the other more obscure ones I found, but can't afford is there is a slew knob on the Dave Hill preamp thingy. That looks cool.

I forgot to mention this in the previous post; when you raise the pitch 1 octave, you double the frequency AND the slew-rate.

An example of the correlation:

If we begin with a tone at 100 Hz with a slew-rate of 0.05, then the following will be true:

200 Hz would have a slew-rate of 0.1
400 Hz would be at 0.2
800 Hz would be at 0.4

So slew-rate follows the same logarithmic doubling-behavior as pitch (Every time you transpose up an octave, you get a slew-rate that is twice as fast)

When it comes to volume things are a bit more murky, since the term "double" can mean several things. But the most relevant, when working with digital audio-files, are the terms sound-pressure and voltage where the slew-rate follows volume 1 to 1, meaning the slew-rate doubles every time you add 6 dB gain (A single bit-shift) to a sound.



You're welcome

My initial expectation would be that stacking should not make a huge difference, as I would expect it to be just like enforcing the same speed-limit twice, but knowing analog I would not be willing to bet money on that expectation

So I would really like to hear what you discover from that experiment

The Newfangled Audio Punctuate transient processor works quite well at 5 bands (the default 26 is silly, lol) for de-skewing stuff like snares and hi-hats.

I don't know that one. Can you tune the frequencies the individual bands cover? (For example, set all 5 bands to each cover a narrow area in the upper spectrum)

The test I did just had the entire frequency spectrum divided into 5 broad-bands, normalized for slew-rate within each single band, and that didn't lead to anything that was really useful over a single-band slew-rate limiter.

But I didn't try bunching up a range of narrow-bands close to each other at the upper frequency-spectrum, where I suppose it could make some sense to have even more granular control.

Interesting idea, the more I think about it (I just wonder if it would end up being more destructive to the sound over-all, since you'd need a different approach when creating narrow-band filters than when having broad-band filters. I have never really done anything that involved setting up multiple narrow band-pass filters though, because of their more aggressive shape, so it's not something I know very much about. But theoretically I suppose it's possible to hone in on slew-rates that are the result of specific instrument-attacks. I suppose we're slowly venturing into the area of super-fine/super-accurate dynamic-EQs at this point, or something like a multi-band D-esser perhaps)

Just had a look and it seems like you can tune the frequencies in Punctuate as you wish in its filter bank window. Though yeah, there may be other consequences to so doing.

So, I was messing with a patch in the TAL J-8 and ended up with this sorta berimbau being slowly tuned up sound. Looked up, saw I had gone from low key slew to through the roof stuff. Which made me wonder what would happen slew-wise on an actual Jupiter 8 with as close as settings?! Conceptually you could use your analyzer as a metric for correlating software skeuomorphs to their hardware bosses.

Now I've got the hardware reconfigured it's also very useful for making sound design decisions. The drum machine goes into a TC Nova od/d pedal which gives you all the options of in series or parallel. No longer do I have to rely on my tinnitussy floppy things—which may be vibing with the oomph of parallel, but if slew rate analyzer strongly suggests series, who am I to disagree? Which also has the happy consequence that I can mess guilt-free with the very digital drum machine.

Keep meaning to ask, whereabouts in Denmark are you? Me and the gf were in Copenhagen in October to see A Place To Bury Strangers at Loppen but we drove there and back so saw various bits of elsewhere, a very long sunset driving up from Hamburg.

I haven't attempted to measure any hardware yet, but it's an interesting thought. I have a ton of old tape-based multitrack-recordings that I think it would be fun to check like that (To see what their slew-rate limits are)

Of course the AD-converter used is going to play a part too, in such measurements. As well as the sample-rate used when digitizing the analog signal.

I currently run my DAW (All my digital audio actually) at 96 kHz sample-rate, but all the old tapes were digitized at 48 kHz, so that of course means some type of interpolation of the sample-rate will happen that may also affect the resulting slew-rates.

You're from Hamburg? I kept wanting you to be from the UK (I think it's because whenever I see your profile-avatar I think about an Oxford University professor in a tweed-jacket )

Hamburg is, or at least was when I was there, a beautiful city

I was there not too long after the 911-attacks, and there was a huge police-event going on at the time (I think it was that one of the highjackers had some connection to Hamburg or some relation to it or something, I don't remember the details exactly, just that there were police everywhere at the time), but walking through the city along the canals after dark was spectacular. And in the sun-light of day all the red-brick buildings with green copper-roofs is very special

It's somewhat similar to Copenhagen in that way, which is also a very beautiful city in my opinion (Or again; was, as each year brings more and more 'modern' stuff that doesn't always bring the same charm along with it)

I'm currently located in a smaller/medium town named Næstved, which is about an hour's drive (Wenn der keiner stau auf dem autobahn ist... if my broken German here isn't complete nonsense ) south of Copenhagen (It's Denmark's equivalent to San Diego because it, like San Diego, has the most stabilized climate of the entire country )

Yes, one of the things the north has going for it is the very prolonged sunsets because of how the sun's curving along the horizon as it descends, which is definitely amazing to experience

The south is warmer for sure, which I personally like a lot, but the transition from day to night is very abrupt there.

It's really fascinating how different things like this can be from place to place

Driving from Hamburg to Copenhagen you probably came quite close to Næstved, as it is, more or less, in between the two.

Small world

No lol, I am in the UK, Hamburg just seemed the simplest place to start the driving story from and the dog is/was indeed British.

Yeah looking at maps we'd have gone quite close to yours on the drive back via the ferry, it seemed like a nice bit of coastline after the could have been somewhere in the UK stretch down from Copenhagen.

Due to the accident that befell the UCX I had to dig out the 48khz MR-18 and its doubtless bog-standard AD. But then again, its Midas pre-amps are just velvetier than the RME ones, so . . . rabbit hole alert