Quantcast
Stacking Theory - Gearspace.com
The No.1 Website for Pro Audio
Stacking Theory
Old 16th October 2008
  #1
Gear Guru
 
Ethan Winer's Avatar
 
🎧 15 years
Stacking Theory

In THIS thread people are talking about "stacking preamps" in the sense that using the same preamp for multiple tracks affects a total mix more than using one preamp for one track. I've heard the same arguments for A/D/A conversion. I think this needs a discussion all its own, so I'll go first. heh

In this sense, "stacking" means the preamps are used in parallel. Any coloration present in the preamp will be repeated for all of the tracks, so when all of the tracks are mixed together they'll all contain that coloration. So far so good - if a preamp has a 4 dB boost around 1 kHz, that's the same as using flat preamps and adding an EQ with 4 dB boost at 1 kHz on the mix bus.

However, no competent preamp has a response nearly as skewed as that. Even modest gear is flat with 1 dB from 20 Hz to 20 kHz. But if a preamp does have a frequency response coloration - whether good or bad - it can be compensated for with mix bus EQ as explained above. It's not like mixing 20 tracks needs 20 times as much cut to compensate. So then we have to consider distortion and noise - the other two audio parameters that affect the sound of a preamp (or A/D/A converter etc).

First, understand that artifacts and other coloration from gear used in parallel like this does not add the same as when the devices are connected in series. I've seen (and performed) A/D/A tests where a track or mix is sent through 10 or more A/D/A conversions in series, then you listen to the 10th or 20th generation to more easily hear the degradation. But this is not the same as stacking in parallel. In series is far more demanding.

This brings us to coherence. Noise and distortion on separate tracks do not add coherently. If you record the same mono guitar part on two analog tape tracks at once, when played back the signals combine to give 6 dB more output. But the tape noise is different on each track and so rises only 3 dB. This is the same as using a tape track that's twice as wide, or the difference between 8 tracks on 1/2 inch tape versus 8 tracks on 1 inch tape.

Likewise for distortion. The distortion added by a preamp or A/D/A on a bass track has different content than the distortion added to a vocal track. So when you combine them in a mix the relative distortion is the same as for each track. Thus, there is no "stacking" accumulation. There might be more total distortion when mixed, but that is a function of the mixing process itself, not the qualities of each track.

In that other thread one person said the damage from stacking has never been proven to his satisfaction. Someone else told him to record a session with better preamps and then he'll agree. The problem with this is it's not a proper controlled test. The only way to test this is to change only the preamp and nothing else. If you record a rock band one day with preamps "A," and a jazz trio the next day with preamps "B," it's impossible to assess anything meaningful about the preamps! The only viable way to test this requires splitting each microphone into different preamp pairs, and recording all the pairs to separate tracks. But that isn't really needed once you understand that parallel stacking does not combine coherently.

--Ethan
Old 16th October 2008
  #2
Gear Addict
 
Nishmaster's Avatar
 
🎧 15 years
Ethan, you know that you and I agree on much, but I'm going to have to disagree slightly. While you are of course entirely correct that the distortion and noise from two preamps "stacked" (in parallel) does not correlate, that does in fact constitute a definite measurable effect.

I'll take two opamps as an example. A TL072 will have THD rise by as much as 10% past 10k or so. A 5532 won't. So in a mix full of this theoretical TL072 device, there will be a measurable increase in high frequency THD, whereas it will be absent in the 5532 device. This is, of course, an oversimplification, but does go toward my larger point, which is that these effects will most certainly be audible.

Now, I will concede that most devices these days have such low distortion and noise that I'm not sure how this actually affects things at the end of the chain, but without proof either way I'm holding off on any final judgment.
Old 16th October 2008 | Show parent
  #3
Lives for gear
 
BradM's Avatar
 
🎧 15 years
Let me see if I'm understanding what Ethan is saying. So if I have a preamp with a 0.5dB bump at 200Hz that makes my track sound a little muddy and I record 16 tracks with this preamps, the total bump at 200Hz will be 0.5dB and not 8dB? This seems to give greater justification (in my mind) for using one EQ across the 2bus to adjust the overal frequency content of my mix. If I wanted to eliminate that 0.5dB bump from my mix it seems I could use one EQ with a 0.5dB dip at 200Hz, or I could use 16 EQ's on each individual track set for the same thing, which would probably introduce a lot more phase shift across the mix. Or do phase shifts not correlate either?

Ethan--thanks for starting a discussion about scientific principles that more of us should understand better. The pseudo-science that runs rampart on the internet is a disservice to us all.

Brad
Old 16th October 2008 | Show parent
  #4
Lives for gear
 
Mike Brown's Avatar
 
🎧 10 years
I think people mistake using different preamps that compliment each source with the fact that the "stacking effect" of using the same preamp on everything makes things sound "worse".

A lot of awesome records were tracked & mixed through one console, using all the same PRES, EQ's, and compressors.

I am not sure why people fool themselves into hearing tangible differences, especially when most music gets put on ipods and played through FM transmitters and earbuds anyway.
Old 16th October 2008 | Show parent
  #5
Lives for gear
 
allencollins's Avatar
 
🎧 15 years
It must be the crappy sound of those traps in your room
Old 17th October 2008 | Show parent
  #6
Lives for gear
 
peeder's Avatar
 
🎧 10 years
I am 100% convinced there is more to the sound of a preamp than can be plotted with frequency response (via 20Hz-20KHz sinewave sweep), distortion (THD and IMD), and noise.

I think the circuits rise ring and settle differently, they respond differently to different levels and dynamic complexities (e.g. memory effects, capacitor draining, sag, etc.), and they have uneven impedance loading and will thus match up with individual microphones differently.

I don't have any means of illustrating this at hand as I do for frequency response distortion and noise. But I am quite certain I hear consistent differences that are not explained by those charts. I think charts that would display the effects I'm describing could be developed and may already have been...

So an argument based on oversimplified premises will lead to simplistic (and likely wrong) conclusions.
Old 17th October 2008 | Show parent
  #7
Lives for gear
 
scott petito's Avatar
 
🎧 15 years
track 10 things with a GML then repeat with a neve you will hear the difference



Cheers
SP
Old 17th October 2008 | Show parent
  #8
Lives for gear
 
chrispick's Avatar
 
🎧 15 years
I'm confused.

Are we saying that the unique sonic qualities of a preamp -- their characteristic frequency responses or contributed harmonic overtones -- are not compounded additively over a full mix?

Ethan says "...there might be more total distortion when mixed, but that is a function of the mixing process itself, not the qualities of each track," but how do we divorce the mix from the equation? Isn't "the mix" integral to the whole "stacking" argument?
Old 17th October 2008 | Show parent
  #9
Gear Guru
 
2 Reviews written
🎧 10 years
Quote:
track 10 things with a GML then repeat with a neve you will hear the difference
He wasn't saying you wouldn't hear a difference in the frequency response, but that that can be corrected with EQ, though I guess that assumes that the frequency response deviation is not dynamic in some way.
Old 17th October 2008 | Show parent
  #10
Lives for gear
 
peeder's Avatar
 
🎧 10 years
Thesis: there cannot be an EQ, nor distortion/noise generator, that will make the results of a GML 8302 preamp into those of a Neve 1073, nor vice-versa.

Yes you can't "undistort" a sound but there is more to it than what is accounted for in your philosophy Horatio.
Old 17th October 2008 | Show parent
  #11
Gear Guru
 
2 Reviews written
🎧 10 years
I was speaking only to frequency response. If the frequency deviation of a device is non-dynamic, that should be undoable via EQ. To the degree that it's dynamic in some way, it would be harder and harder to do. But if a pre-amp has a known deviation from flat and it's a fixed deviation, then that's really no different than putting an EQ there, in terms of frequency response. I'm not saying that anything specifically has a fixed deviation, just saying if it was.
Old 17th October 2008 | Show parent
  #12
Gear Maniac
 
stuntbutt's Avatar
 
🎧 15 years
Quote:
Originally Posted by scott petito ➑️
track 10 things with a GML then repeat with a neve you will hear the difference
I think Ethan is saying that the sonic differences in stacking 10 tracks from each preamp would be the same as comparing a single track from each preamp.
Old 17th October 2008 | Show parent
  #13
Jai guru deva om
 
warhead's Avatar
 
🎧 15 years
The whole "stacking" thing, well I don't know...all records were made with the same console preamps for decades.

Microphone selection and placement, EQ'ing while building a song like a puzzle one piece at a time, making sure they all fit as you move forward and listen in context, a .5dB "build up" at 200hZ won't really matter in the end and the thought never even crosses my mind along the way if I'm using one pre or 6.

War
Old 17th October 2008 | Show parent
  #14
Lives for gear
 
Sigma's Avatar
 
🎧 15 years
Quote:
Originally Posted by warhead ➑️
The whole "stacking" thing, well I don't know...all records were made with the same console preamps for decades.

Microphone selection and placement, EQ'ing while building a song like a puzzle one piece at a time, making sure they all fit as you move forward and listen in context, a .5dB "build up" at 200hZ won't really matter in the end and the thought never even crosses my mind along the way if I'm using one pre or 6.

War
yeah i don't get it either..it sounds good or it doesn't ..electrodynes had a pre , 2 band eq's and a cue send..sure were alot of hits on them..stand alone pro pre's weren't really used much till the 90's ..at least from what i am aware of.

i am kinda confused as to the whole point of the thread..could be just me though

i think people over anal ize stuff.. listen and adjust the tools in the chain..you can turn a mic a 1/4 inch and the sound is different.. it ain't rocket science
Old 17th October 2008 | Show parent
  #15
Lives for gear
 
Saudade's Avatar
 
🎧 15 years
interesting....always wondered about stacking too, but never thought of it as "series vs parallel".

I think nobody would dare (heh) to conduct a fair test in the method Ethan suggested, using maybe 24 channels of Behringer preamps vs same channels of X brand (insert your favourite boutique priced brand) direct to same high-end A/D converters, summed in the same DAW. Cos either way the test turns out, it's gonna hurt to know the truth for a lot of people
Old 17th October 2008 | Show parent
  #16
Moderator
 
TonyBelmont's Avatar
 
🎧 15 years
Old 17th October 2008 | Show parent
  #17
Gear Guru
 
2 Reviews written
🎧 10 years
I would worry about stacking too many pre-amps together, since you might bump into them and knock them over.

I can see how this or that pre-amp might have characteristics that would make it aesthetically undesirable to run a lot of tracks through them, if they have radical amounts of character, maybe a Germ or something like that.

I do though think that the 'use it on a bunch of tracks' thing is commonly used as an escape clause by golden ears when they are presented with evidence that doesn't support their beliefs. And I think that's true even if there might be some truth to it. It's just commonly trotted out at any point where there's no way to 'splain away what they are hearing.

I've said many times that it would be great if someone who had the equipment and high quality room to do it would track a reasonably sized piece where every mic was split into two different pre-amps of significantly different character but otherwise going through the indentical, high quality, input chain. So every track would be there twice, just through different pre-amps. Then the difference within the context of a number of tracks could be examined.

That's the only way to get a result that no one can dismiss. Well, some way to do it will be found if it contradicts existing beliefs, but at least it would tell an unambiguous story about the results of layering pre-amps of the same type since the resulting mixdown will indicate what the difference is.
Old 17th October 2008 | Show parent
  #18
Lives for gear
 
vincentvangogo's Avatar
 
🎧 10 years
I am not technical so please forgive me if this sounds stupid, but isn't it about other factors than just distortion? eg subtle harmonics emphasized by the pres. So a pre that is more 'musical' in the harmonics it emphasizes will remain/become more musical as they are stacked, while ones that aren't as 'musical' won't.
Old 17th October 2008 | Show parent
  #19
Lives for gear
 
gainstages's Avatar
 
5 Reviews written
🎧 10 years
Interesting thread. \

While I respect Ethan's knowledge I'm more in Peeders court on this one. In a very unscientific way that I can't begin to prove, I do feel that some commonality in preamps can help the "cohesiveness" of a project as a whole. I believe that this probably is an issue more for those who are less experienced... a great pro could record a project without using the same preamp on more than one track, and by matching the right preamp to the task and also to the project as a whole, come out with a very cohesive and organic sounding project. A less experienced person could take the same collection of preamps and struggle to obtain a cohesive sound, yet feasibly do somewhat better with a single set of preamps.

Over the years as I have interacted with hundreds (if not thousands) of engineers and recording musicians, I feel i have seen this evidenced in the work of many of them. there are a number of other factors to consider here, so "proving" anything would be difficult, and frankly it would be pointless other than for the sake of bragging rights as to who is right and who is wrong.

Rule #1 - if it sounds good, it is good

I personally do prefer to use the same, or very complimentary preamps, for much of the project, and use different preamps on key tracks to help them find the right "pocket" in the mix... but everyone has their own way. I may be right, i may be wrong, but it isn't worth time spent testing it for me.

now if someone wants to do the test and share the results, of course I'm interested!
Old 17th October 2008 | Show parent
  #20
Lives for gear
 
chrispick's Avatar
 
🎧 15 years
Quote:
Originally Posted by Saudade ➑️
interesting....always wondered about stacking too, but never thought of it as "series vs parallel"
I have to say I don't even understand this distinction in this context. What's meant by "parallel stacking?" That sounds oxymoronic to me. Like "jumbo shrimp" or "sanitary sewer."
Old 17th October 2008 | Show parent
  #21
Gear Guru
 
2 Reviews written
🎧 10 years
Parallel is what we are all doing. Series would be the output of one into the input of another. No one would probably ever actually do series stacking in reality, I wouldn't think. It was just being put forward as the scenario where effects would add up in a way that would be much more obvious. I.e. if you lined up 10 Neves in a row and ran the same signal through them all, vs recording the same thing 10 times to 10 tracks each via one Neve and then sum them to a mix.
Old 17th October 2008 | Show parent
  #22
Lives for gear
 
gainstages's Avatar
 
5 Reviews written
🎧 10 years
Quote:
Originally Posted by Dean Roddey ➑️
I would worry about stacking too many pre-amps together, since you might bump into them and knock them over.

I can see how this or that pre-amp might have characteristics that would make it aesthetically undesirable to run a lot of tracks through them, if they have radical amounts of character, maybe a Germ or something like that.

I do though think that the 'use it on a bunch of tracks' thing is commonly used as an escape clause by golden ears when they are presented with evidence that doesn't support their beliefs. And I think that's true even if there might be some truth to it. It's just commonly trotted out at any point where there's no way to 'splain away what they are hearing.

I've said many times that it would be great if someone who had the equipment and high quality room to do it would track a reasonably sized piece where every mic was split into two different pre-amps of significantly different character but otherwise going through the indentical, high quality, input chain. So every track would be there twice, just through different pre-amps. Then the difference within the context of a number of tracks could be examined.

That's the only way to get a result that no one can dismiss. Well, some way to do it will be found if it contradicts existing beliefs, but at least it would tell an unambiguous story about the results of layering pre-amps of the same type since the resulting mixdown will indicate what the difference is.
I don't think this test would completely settle the argument. who is going to do the test? an "average joe" or a "golden ears"? there would have to be some sort of a rough mix done in order to compare this, and who is going to make those decisions? also -if you use identical settings on both projects, that still wouldn't be a "fair" comparison. I drive my rear wheel drive truck slightly differently than my front wheel drive van, even though they have virtually identical controls - a "fair" comparison wouldn't subject the results of one to be improsed upon the other. at the same time, if you allow slight variations in the settings, who is mixing them? if it's two different people, that is hard to call scientifically "fair". If it is the same person who does both, that is also hard to call fair... if the first one sounds better is it because his ears were fresher? if the second one sounds better is it because he had already mixed the song once and was able to do a better job having had a "warm up"?
what about the preamps themselves? how did we choose the 24 different premamps for version one and the 24 same preamps for version two? surely someone will claim that a different set of preamps would have a different result.

See the problem? someone will always have a fairly legitimate arguement that the test is either incomplete or not fair.

If the differences in a test are HUGE, then even a slightly incomplete or slightly unfair test could be a valuable educational tool. however, if the differences are more subtle - which is pretty realistic - than all the variables come into question and everyone will pick apart the test trying to support their own beliefs.

I've already stated my beliefs. Yes, I do believe their is some value to stacking tracks. No, I don't believe its the holy grail of recording and I don't even follow my own rules quite a bit of the time. I think the difference is less and less the more experience you have.

wow, I've wasted too much time on this topic already...
Old 17th October 2008 | Show parent
  #23
Lives for gear
 
chrispick's Avatar
 
🎧 15 years
Quote:
Originally Posted by Dean Roddey ➑️
Parallel is what we are all doing. Series would be the output of one into the input of another. No one would probably ever actually do series stacking in reality, I wouldn't think. It was just being put forward as the scenario where effects would add up in a way that would be much more obvious. I.e. if you lined up 10 Neves in a row and ran the same signal through them all, vs recording the same thing 10 times to 10 tracks each via one Neve and then sum them to a mix.
Oh, I know what "parallel" and "series" mean. I was just poking fun at the term "parallel stacking."
Old 17th October 2008 | Show parent
  #24
Gear Guru
 
2 Reviews written
🎧 10 years
They could just make the raw tracks available for anyone to examine. And, in terms of what's being discussed here, it doesn't need to be mixed. We only care about the changes in the recorded signal, then summed. So a summed version of each could be made available for everyone as a starting point in the analysis.

If the differences in the resulting summed tracks are tiny, then end of story. The mix is pretty much irrelevant since the differences in the starting points are so small.

If the differnces are large, then make the raw tracks available, and anyone can take them, apply some EQ or whatever they feel is required and see how closely they can match the two. If they can be matched with EQ, or EQ plus a little harmonic enhancement or something like that, then that's an answer as well, because it shows what the real differences are.

If no one can really match them well, then that's also an answer as well.
Old 17th October 2008 | Show parent
  #25
Gear Guru
 
2 Reviews written
🎧 10 years
Quote:
Oh, I know what "parallel" and "series" mean. I was just poking fun of the term "parallel stacking."
It works if you lay on your side.
Old 17th October 2008 | Show parent
  #26
Lives for gear
 
WunderBro Flo's Avatar
 
2 Reviews written
🎧 15 years
Quote:
Originally Posted by Dean Roddey ➑️
If the differences in the resulting summed tracks are tiny, then end of story. The mix is pretty much irrelevant since the differences in the starting points are so small.

If no one can really match them well, then that's also an answer as well.
Such a recording HAS been done by Lynn Fuston on his excellent "Preamp Paradise" disc. One song recorded with different pres over and over, with a test setup that minimized human error as much as possible (VERY important point). I mixed the tracks together to get one Neve-only-mix, API-only-mix, DW-Feran-only mix etc. Exact same mix, just the source tracks recorded over and over again with different pres.

The result so far has been that...
1) the differences between whole mixes were tiny, just as with the single tracks
2) noone could match them (not "just" me, a lot of trained & untrained ears have tried to) or point even one out as being tube, clean or transformer colored.
3) everybody who listened was pretty much floored because everyone was expecting at least clearly audible to huge differences.

From this test, the conclusion can only be that when using the preamps the way Lynn used them (mics-micpre-AD with calibrated levels but most probably not driving the pres into saturation), there is no "culmulative effect".
It would be great to pass those mixes out to everyone, but that is not possible because I respect LynnΒ΄s request NOT to spread them around. You gotta buy the disc and set the mixes up yourself to experience it.

Rock on!
Pat

ps - the only thing of value to these preamp&stacking questions are actually recordings done at the same human-error-minimizing level. Opinions (and explanations of such opinions without actual recorded facts to back them up) do not really matter much because everyone has one and they heavily differ on these things.

So if anyone did a test at a similar minimal-human-error level that yields different results, up with it! We want to know/hear about it!
Old 17th October 2008 | Show parent
  #27
Lives for gear
 
🎧 15 years
Noise would be additive atleast, no?
Old 17th October 2008 | Show parent
  #28
Lives for gear
 
scott petito's Avatar
 
🎧 15 years
Quote:
Originally Posted by stuntbutt ➑️
I think Ethan is saying that the sonic differences in stacking 10 tracks from each preamp would be the same as comparing a single track from each preamp.
yes I understand what he meant and I was also very specific with the Preamps I chose....


cheers
SP
Old 17th October 2008 | Show parent
  #29
Kush Audio
 
u b k's Avatar
 
🎧 15 years
vibe accumulates, no way around that.

10 tracks of cheap focusrite hardness sounds way harder than 10 tracks of v72 roundness. summing is very much a serial phenemenon to my ears; a pre with peaky mids at 1-2k will impart that peakiness onto nearly every element, and the sound is *not* fixable via an eq cut on the mix.

if it were, every ass-slathered set of tracks i've ever been given to mix could've been transformed into api goodness with a few well placed tweaks, but they weren't.

almost every guy here who mixes for hire knows exactly what i'm talking about. you cannot process out graininess, smallness, blur, hardness, or stridence. there are things we can do to help, but there are definite limits to what can be done to alter or rescue the tone of a recorded source, and (more to the point) bad tone piles up quick when the faders go up.


gregoire
del
ubk
.
Old 17th October 2008 | Show parent
  #30
Lives for gear
 
WunderBro Flo's Avatar
 
2 Reviews written
🎧 15 years
Quote:
Originally Posted by u b k ➑️
10 tracks of cheap focusrite hardness sounds way harder than 10 tracks of v72 roundness. summing is very much a serial phenemenon to my ears; a pre with peaky mids at 1-2k will impart that peakiness onto nearly every element, and the sound is *not* fixable via an eq cut on the mix.
ubk, just curious,
how do you explain then that noone can tell a focusrite isa428 mix from the same mix just tracked with DW Fearn VT2 or UA M610 preamps (without saturating the pres) ? Should not at least be there some noticable hardness or roundness? Do you think it is all dependent on if the pres get saturated (which was not done in LynnΒ΄s recordings, at least not on purpose)?

Besides, I do not see any reason for blaming the preamps everytime I get bad sounding tracks form someone. If something sounds bad, there is no way to find out afterwards if it was bad preamp sound, mic technique or any other piece of gear in the signal chain. I have heard so many tracks that sounded really, really bad, and the guys recording it used really expensive highend preamps. Nobody blamed the preamps in those cases. But as soon as someone uses non-highend or the "wrong" highend preamps, everybody goes "If you had recorded this with xxx/yyy instead, this sh*t would sound beautiful."

I am still looking for the recording that demonstrates the "culmulative effect" on a serious basis (minimized human error). If it is possible to record/mix a full band without it (must be a coincidence then, right) then there must be a way to record/mix with it. I wanna hear it! Who got some?

Rock!
Pat
πŸ“ Reply

Similar Threads

Thread / Thread Starter Replies / Views Last Post
replies: 60 views: 15701
Avatar for lucey
lucey 23rd May 2004
replies: 9 views: 2565
Avatar for RightOnRome
RightOnRome 7th July 2020
replies: 80 views: 19412
Avatar for Fishy1500
Fishy1500 2nd January 2009
replies: 67 views: 22840
Avatar for MarkyGoldstein
MarkyGoldstein 29th August 2013
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