Levels in Logic

[Ashermusic]

Thre was a lot of discussion some time ago on whether or not it is importat to try adn keep channel levels from clipping (with Pre_fader metering) in a 32 bit float app likeLogic as long as it is not clipping the output. I took the postition that it still is and some disagreed.

 

The following is my exchanege with Paul Frindle, who designed the Sony Oxford series of plug-ins and is an acknowledged expert in digital audio.

 

i apologize for the length but I think it is important.

 

PAUL WROTE:

There are 2 reasons to record, process and master at less than flat out dBFS:

 

1. To avoid hidden overs not shown on metering due to the reconstruction of the signal. This occurs mostly in D/As (variably depending on how they are designed) and to some extent within some plug-ins and digital processes. Error from this range from limiting all the way to loud 'splats' as values may fold over completely.

 

A safe margin for normal programme to avoid most of this is -3dBFS. Although some artificially maximised programme can create more than this and some test material can create 6dB of over - so real safety is only gained at around -6dBFS.

 

If you have a reconstruction meter you can monitor this effect yourself (for the mix output) and compensate manually, or if you have a suitably equipped limiting app (I.e. Oxford Limiter) you can correct these errors automatically. This does not help much with stuff within the channels of the mix itself - so prudence us still advisable.

 

2. To create headroom, using lower levels within your mix allows you to avoid clipping signals every time you do anything - it frees you up to concentrate on sound rather than red lights and radically eases the mixing process. Some plugs may actually sound better because internal overs may be avoided.

 

To do this you need to reduce levels to something sensible first thing in the playback channel - process at lower levels - end up with a mix at less than flat out - then make up the level at the very end of the mix. It sounds like you are doing this already But please note this is NOT to avoid math overs in the PT summing buss - as these are catered for already in the PT mixer

 

Ok - you talk of making mixes that are suitably modified by the output limiter? Yes, this is common practice and in fact mixing with the limiter in place is a really good idea as you instinctively adjust the mix for the best final sound. But these days we have to watch it as the industry is obsessed with loudness at the expense of absolutely everything else - we produce 2 dimensional programme that has no dynamic range. Therefore it isn't possible for you to create real dynamics in this current environment (if you want to stay in business) - instead you are limited to trying to create the impression of dynamics from the extra artefacts and distortions the limiter generates.

 

Is was with this in mind that I designed the Oxford Limiter - basically to create the impression of dynamic range when in fact there was none - and do it in a way that sounded as natural as possible. You can use this effect either to produce stuff that is loud as ever but sounds less artificial - or you can use it to produce stuff that is as bad as before but is even louder

 

I hope this is helpful

 

I THE ASKED: ul would you say this advice holds true for a 32 bit float app like Logic as well as fixed point apps like PT?

 

PAUL RESPONDED:

Yes I would.

 

The intersample peaking reconstruction problem is the same at the output of the mix, as it must be represented in a fixed point output format anyway (i.e. CD or DVD).

Whilst with float it's possible to accomodate internally numbers bigger than flat out, any process that has need to refer to actual real values might be at risk of overload (or unspecified behaviour). Why take the risk?

 

From the point of the headroom issue, things might be different in that an entirely float system from start to finish might handle overs properly - however the meters will be calibrated to a fixed point reference (and will come on willy nilly, whether the signal is clipped or not). Some systems using expansion DSP pass and process signals in fixed point (PowerCore being one example) and may not have any of the float headroom and may mess up with overs.

Again, with 140dB or so dynamic real range at your disposal, why bother to risk it?

 

The most important thing to remember is that recording and processing at lower levels DOES NOT waste 'bits'. It doesn't work like that - all your 'bits' are there all the time at all levels

[dkgross]

The most important thing to remember is that recording and processing at lower levels DOES NOT waste 'bits'. It doesn't work like that - all your 'bits' are there all the time at all levels

 

great info. I agree with you. I track at moderatly 'hot' levels', but ALWAYS leave some headroom.

 

I also use some metering from RogerNicholsDigital to keep an eye on everything.

 

http://www.rogernicholsdigital.com/products/

[David Nahmani]

I almost agree with Paul:

 

1) If your outputs peak above -6dBFS, you risk overloading at playback, depending on the D/A on the playback device (obviously consumer CD Players are more vulnerable). I have discussed this at length in the following thread: http://logicprohelp.com/viewtopic.php?t=5489.

 

2) Within Logic, meaning on the track or instrument channel strip, you can go over 0dBFS and the sound quality will be bit identical on the output as long as the output does not clip.

 

3) Why risk it? If you can avoid it, obviously do so. But sometimes I see people take hours of precious time trimming the automation on all their tracks so their track/instrument channels don't go over 0dBFS, while they could simply have turned down the master output channel. Again, the results are bit identical. My answer to Paul would be: Why not take advantage of it? Oh... and: what risk?.

[Ashermusic]

I almost agree with Paul:

 

1) If your outputs peak above -6dBFS, you risk overloading at playback, depending on the D/A on the playback device (obviously consumer CD Players are more vulnerable). I have discussed this at length in the following thread: http://logicprohelp.com/viewtopic.php?t=5489.

 

2) Within Logic, meaning on the track or instrument channel strip, you can go over 0dBFS and the sound quality will be bit identical on the output as long as the output does not clip.

 

3) Why risk it? If you can avoid it, obviously do so. But sometimes I see people take hours of precious time trimming the automation on all their tracks so their track/instrument channels don't go over 0dBFS, while they could simply have turned down the master output channel. Again, the results are bit identical. My answer to Paul would be: Why not take advantage of it? Oh... and: what risk?.

 

Re: #2 That is only true for sure if you are only using Logic plug-ins, whch few pros do.

 

David, I have great respect and affection for you but::

 

On the one hand I have your opinion. On the other I have Paul Frindle as well as Nika Aldich, Bob Katz and Terry Manning.

 

Respectfully again, I simply do not view you as their peer.

 

And it is quite easy to avoid the problem if you simply turn on pre-fader metering and are cognizant of it from the begining.

 

It has made a huge difference in the sound of my mixes.

[Ashermusic]

BTW, I feel i should clarify my last statement lest it appear I was taking shots at David. I have great respect for him. He is one of the most knowledgable people about Logic and computing that I have encountered.

 

It occurs to me that some of you may not know who thse guys are .

 

Nika Aldrich wrote the booh "Digital Audio Explained" and Bob Katz is the author of "Mastering Audio', two of the bibles of digital audio. Bob also invented the K-System metering.

 

Paul Frindle designed the Sony Oxford plug-ins, probably the most respected in the business.

 

Terry Manning has engineered and mixed some of the best and biggest records of the last quarter of a century.

 

This is an elite group, no? So when I say to David that I do not view him as their peer it is like saying to me, "Jay you are a good composer but I do not view you as John William's or Jerry Goldsmith's peer."

 

I would not be offended and I hope David is not.

[David Nahmani]

But Jay, like I said, we don't disagree, we're just not talking about the same thing. I totally agree with you and Paul that it is good practice and something to recommend to keep your levels low. It has many advantages as far as the workflow is concerned (if nothing else, because if all your tracks are above +6dBFS, well the meters won't give you much information).

 

As for the distortion introduced by clipping the channels before the output channel, there is none. This, I have proved with my test that is in a thread somewhere in that forum, and I encourage you to prove it to yourself by doing your own test.

 

Now you mention distortion in 3rd party plug-in, well all I can answer is that in my experience, all 3rd party plug-ins in the AU standard are 32 bit floating point and follow that rule. Again, feel free to do your own tests (if you deem it worthy) if you want to verify this with your 3rd party plug-ins.

 

I think it's important to know where the distortion comes from, and if you're using Logic, important to understand that There is no distortion introduced by Logic at the output of a channel (other than the output channels), even if the meters go way above +0dBFS.

 

Like I already told you and others, if you believe that you run the risk of creating distortion in a 3rd party plug-in, then the channel meter is certainly not a good tool to measure that distortion: for the sake of discussion, let's assume that there is out there a 3rd party plug-in that works in 24 bits. It would be easy to have an EQ first, that plug-in second, and maybe a compressor third on your channel strip. You boost the EQ, totally distort your plug-in, and compress the distorted signal, lowering its level. Your channel meters will NOT indicate that distortion.

 

Conclusion:

• The Channel meters show you the level at the Output of a Channel Strip (or at the output of the last plug-in if you choose Pre-Fader Monitoring).

• Since Logic is using a 32 bit floating point engine, it does not introduce distortion at the output of a Channel Strip, even if your signal goes way above 0dBFS (I tested this at about +96dBFS).

• The Channel meters do NOT show you the level at the input of a plug-in, and therefore are not a tool that can help measuring the distortion introduced by a plug-in, if any.

[toysun]

Thanks Jay, for posting this. Very important topic, and something that we all need to understand better. I have the same metering set as Dave does, I'm using them as PPM meters and using lower and lower levels, as I read (and understood most of ), Bob Katz' book.

 

It is hard for an old school 41 year old to learn to embrace lower levels, as I came up on tape, but I'm learning!

 

 

John

[dkgross]

you'll have to let me know what part of Bob's book you DO understand..cuz, it's WAYYY over my old head

[toysun]

The part about leaving mastering to the pros!

[dkgross]

well.. duh.

 

 

some of us even like leaving the MIXING to other. It's weird, but I've just gotten to where I lose all objectivity by the time I'm done writing something. I'm VERY fortunate to have two amazing engineers at my disposal to do final mixes for me. I'll send stuff to them with rough relative levels and pans, and any special plugins that I really like..but other than that, I let them do what they want.

 

It's pretty cool. Sometimes they hear something totally different than I did...a different hero moment/instrument...different eq on a guitar...etc..

 

just my .02...I think everyone should try it for a song or two. Ya never know what kind of new relationship might develop.

[Ashermusic]

Conclusion:

• The Channel meters show you the level at the Output of a Channel Strip (or at the output of the last plug-in if you choose Pre-Fader Monitoring).

• Since Logic is using a 32 bit floating point engine, it does not introduce distortion at the output of a Channel Strip, even if your signal goes way above 0dBFS (I tested this at about +96dBFS).

• The Channel meters do NOT show you the level at the input of a plug-in, and therefore are not a tool that can help measuring the distortion introduced by a plug-in, if any.

 

David the last time around you proved to me that these are true but when all or most of the channels are in the red even if the Output is not clipping there is something going on that in the sum result that was leading to a distorted mix that goes away when you lower the channel levels. It may be that certain plug-ins like Powercore and UAD-1 are not always float, it may be that when too much signal is thrown at the mix output it reacts badly when bouncing, I don't know but from my own experience it made a huge difference. I wish I still had those Logic files.

 

 

BTW, on the Apple Pro Forum BeeJay is trying those tests and he has so far not been successful in getting the two mixes to null but by his own admission perhaps he is doing something wrong.

 

Also when Paul Frindle wrote these two things they jumped out at me:

"Whilst with float it's possible to accomodate internally numbers bigger than flat out, any process that has need to refer to actual real values might be at risk of overload (or unspecified behaviour)."

 

"From the point of the headroom issue, things might be different in that an entirely float system from start to finish might handle overs properly - however the meters will be calibrated to a fixed point reference (and will come on willy nilly, whether the signal is clipped or not)"

 

Bottom line folks for me is that if you ignore the channel levels going into the red totally and depend on just lowering the output as a methodology for mixing you are at best observing very poor mix practices that at the end of the day will lead you to an inferior sounding mix. And I was getting distortion on my bounces somehow that went away when I turned on Pre-Fader metering and lowered the outputs of my audio instruments and/or 3rd party fx so that my channels were not going into the red.

 

I can't explain it scientifically but I heard it and my client heard it. And now I no longer ever have that issue.

 

As Paul Frindle says: " with 140dB or so dynamic real range at your disposal, why bother to risk it?

 

The most important thing to remember is that recording and processing at lower levels DOES NOT waste 'bits'. It doesn't work like that - all your 'bits' are there all the time at all levels :-)"

 

And with that I am done. Thank you David for your patience and courtesy with me as always. You are a class act.

[toysun]

As Paul Frindle says: " with 140dB or so dynamic real range at your disposal, why bother to risk it?

.

 

I agree, why risk it? Folks... hot levels these days (at least until mastering) are psychological, no?

 

John

[zaque]

Wow, what a great thread! Very welcome reading for the morning coffee

 

Sounds to me like you're both (Jay and David) giving good advice here.

 

Here is what I'm getting from this:

 

Jay is saying that it is worthwhile to pay attention to your channel levels because you may or may not introduce problems in your mix if your levels happen to pass the flat out dBFS. This is sound advice, especially when using third-party pluggins. Based on some professional consultation with some pretty impressive industry peoples, this advice is backed up.

 

David is saying that while it is probably a good idea to keep levels in mind don't break your back trying to "fix" all of your levels because you have a few channels in the red. Because Logic channel output will not introduce distortion, in many cases you may be safe to just roll off the master output. This also makes sense.

 

If you combine these two suggestions together you have a very practical approach to working. I think we can all safely agree that pushing everything into the red can introduce some unseen/heard problems which could trouble you "later". Pay attention to your levels but don't hurt yourself or waste tons of time (and client money) trying to get everything just under 0dBFS because it may not matter.

 

Moral of the story?

 

Follow good mixing practices as much as possible and make your best guess when it's safe to stray from that in an effort to balance efficiency with quality.

[Ashermusic]

Wow, what a great thread! Very welcome reading for the morning coffee

 

Sounds to me like you're both (Jay and David) giving good advice here.

 

Here is what I'm getting from this:

 

Jay is saying that it is worthwhile to pay attention to your channel levels because you may or may not introduce problems in your mix if your levels happen to pass the flat out dBFS. This is sound advice, especially when using third-party pluggins. Based on some professional consultation with some pretty impressive industry peoples, this advice is backed up.

 

David is saying that while it is probably a good idea to keep levels in mind don't break your back trying to "fix" all of your levels because you have a few channels in the red. Because Logic channel output will not introduce distortion, in many cases you may be safe to just roll off the master output. This also makes sense.

 

If you combine these two suggestions together you have a very practical approach to working. I think we can all safely agree that pushing everything into the red can introduce some unseen/heard problems which could trouble you "later". Pay attention to your levels but don't hurt yourself or waste tons of time (and client money) trying to get everything just under 0dBFS because it may not matter.

 

Moral of the story?

 

Follow good mixing practices as much as possible and make your best guess when it's safe to stray from that in an effort to balance efficiency with quality.

 

Man you did that well!

 

Can we send you to the Middle East to meet with Israel and Hamas?

[zaque]

Yeah, Israel and Hamas would do well to follow better mixing practices.

[shandy]

Well if your recording in 24bit you have high headroom and a low noise floor so you can even record with your peaks reaching -10 its not a problem. Why compromise your recording for a couple of dBs..

[Ashermusic]

BTW, I also posted this same stuff on the Apple Logic discussions forum and several folks there have already reported back that they tried this and noticed a big difference in the quality of their mixes. Once called it an epiphany.

 

It is very gratifying.

[malgfunk]

I agree with Jay. I've been mixing alot of Hip-Hop lately in Logic and being used to mixing on SSL's with 2" tape I tend to be more agressive by nature in that I really push the envelope and I find that my mixes have some audiable distortion, even without clipping the output. One day I mixed a project channel levels backed off a bit for more margin and my resulting mix was much cleaner. just my 02..

 

Later,

 

Malgfunk

[Ashermusic]

I just came across this post on the PSW forum that sums it up really well IMHO:

 

The relationship between recording levels and subsequent processing levels is somewhat relevant, given many here have mentioned gain staging and overhead within mix engines as the dominant reason for keeping record levels lower.

 

It is less important in floating point systems, but (as Nika implies) is is still important. This is because calculation accuracy diminishes at levels where signals attempt to exceed full scale in float processing. This loss of accuracy is a good trade off when compared to the complete barfing which happens when fixed point systems attempt to exceed full scale, but it still gives inferior results compared to "always staying below full scale". The reason Nika didnt go into detail is the argument is mostly about "how much accuracy is lost?", especially when comparing very high resolution systems using 64 bit float or double precision fixed point. Depending on who is doing the arguing (and I have read many posts in various forums over the years from Nika and others on this topic) the loss of accuracy is considered a "good trade-off" or a "significant trade-off"....suffice to say, floating point processing can be a lifesaver in many situations.

 

In other words, floating point systems do remove the worry of clipping within the processing chain, but you will still achieve best results if you never approach levels which would clip in a fixed point system.

 

Of course, not all plugins within floating point hosts are also float based. Most DSP card solutions use fixed point - the only exception I know of is the UAD1 DSP card, which can use floating point because the processor is a GPU rather than a dumb DSP chip. So all of the UAD1 plugins use floating point. On the other hand, some native plugins use fixed point even when invoked in a floating point host - a good example is the Waves plugins, many of which are fixed point. Waves do this to maintain identical processing across their entire range of platforms (or, maybe, becuase they dont see any benefit in reconfiguring their fixed point algorithms for floating point). So Waves users need to carefully watch their "gain staging", even in floating point systems.

 

The converse needs to be considered here: is there any loss in fidelity if moderately lower signals are used in floating point processing? The answer is "no".

 

Therefore, the same concept applies to floating point systems as it applies to fixed point systems: lower levels are best in modern DAWs.

 

Sean

Sean Diggins

The Tone Room

[David Robinson]

G.day, Jay, Dave and all

firstly, a little anecdote: did a tv sting a coupla year back with a large orch mock up. normally in logic my recording levels had been conservative up until this, but i said to myself, why not bump everything up and render all trks peaking at full scale..... so i did.

NEVER AGAIN. even though each audio clip sounded fine played by itself, when summed there was distortion and harshness i'd NEVER experienced in logic before, and this was with all chl faders way down, and same for sends! somewhere between -24 and -30db were the trk levels, and it STILL sounded way bad. (the master bus is always @ 0db, the level there showed normally ie, ~ -6db on highest peak.) i tried that mix thru six different converter brands and the better the conversion the uglier this got. even friends with NO AUDIO TRAINING at all could hear the compromised quality.

 

never again, never ever again.

[velvet]

I've read somewhere on the net an interview with Bruce Swedien. He told something like he is tired of mixing in digital as who the hell still has to think about not getting into zero / thinking that it is necessary to stay at lower levels but the ZERO level somehow still acts as a power we want to get close to, right?, and he told that the next "DAW" he is designing for himself will be based on analog (again)... Ofcourse he did not tell "DAW" - but I do not know the right word for it - analogue workstation maybe ) - sorry Mr. Bruce for non-accurate reproduction

 

BTW, other thing I do along with checking pre-fader metering is, that I automatically turn off the AutoGain on compressor plugins.

Most of the times.

...sounds funny: " Automatic AutoGain turn-off"

[David Nahmani]

Well Jay, it seems like you don't understand the issue (by your own words), so you'd rather trust a source that has some "clout", like the famous producers you quote. Problem is, either you misquote them, or they are wrong on certain points.

 

I cringe when I read some of the things you affirm in this thread.

 

I stand by my reasoning, which is to me as obvious as 0 = 0 and 1 = 1. You see that to me, this is not a matter of opinion.

 

I would give a detailed answer but we've already discussed this ad nauseum a few months ago and even the Logic developers said I was right. I'm not sure why you still feel the need to put that topic back on the table. It seems to me that you are not approaching this in a scientific way, since you already have an agenda before you even start experimenting. I have done the experiment, numbers don't lie.

 

Myth #1: When the channel meters hit the red, the signal is distorted. WRONG.

Myth #2: If the channel meters don't hit the red, the signal is not distorted. WRONG.

 

If you only use your channel meters as a tool to indicate distortion, you're in for a surprise. I can show you mixes that distort and never go in the red, and mixes that don't and that are all the way in the red.

 

By the way the thread you mention on the apple forums has since been updated by Justin, the developer of Accelerando, #1 Apple forum poster and highly respected developer and audio professional. And he agrees with me.

 

The most important thing to remember is that recording and processing at lower levels DOES NOT waste 'bits'.

Wow. This shows a SERIOUS missunderstanding of digital recording. I hope you didn't write that Jay? Whoever wrote that was either half asleep, or has never really understood the basic principles of digital audio recording. Sorry!

[David Robinson]

hello david, with only 7 seven years of recording in logic under my belt i'm probably just a "newbie" but, you can easily overload audio in logic in so many ways, does the tech side really matter that much to the user?

they just want their music to sound "bitchin'", and how to fix the problem.

so do i.(or i did, until i simply lowered the levels and turned the monitors up)

in my very humble opinion, logic is fanastic, no matter how you slice it, distortion and all.

these posts seem rather useless to the average user.

but David i love visiting here, so keep up the honest good work.

David Robinson.

[David Nahmani]

I believe it does matter to know and understand the tools you use to produce your music. How deep you need to know them ... depends on the person. You can certainly produce great music without understanding the technology behind. Just use your ears. I have heard some fantastic productions from people who wouldn't know if their DAW is set to 16 bit, or 24 bit, or 44.1KHz or whatnot.

 

But what pains me is when I see people use the wrong tool for the wrong job. Maybe it's a teacher's deformation? I just can't help but point to them that they are not doing it right. That does certainly not mean their music suffers from it, it's just.. well you know, when you watch a mixing engineer who keeps adjusting a fader's position while mixing, and you know that particular fader isn't routed to anything? Doesn't that make you want to tell him? Meanwhile he might have made a great mix without you telling him...

 

But either way you better trust your ears, as that's the final metering tool, the only one that will never lie.

[David Robinson]

Dave, been known to do a bit of teaching as well, still do, and i certainly prefer a student who asks plenty of probing questions about things.

keeps me on my toes and certainly keeps things heading in a positive direction.

cheers, will post more when i feel i can contribute something worthwhile.

thanks, David R.

[dkgross]

some interesting reading

 

http://akmedia.digidesign.com/support/docs/Mixing_in_the_Box_26689.pdf

 

http://akmedia.digidesign.com/support/docs/48_Bit_Mixer_26688.pdf

[Ashermusic]

The most important thing to remember is that recording and processing at lower levels DOES NOT waste 'bits'.

Wow. This shows a SERIOUS missunderstanding of digital recording. I hope you didn't write that Jay? Whoever wrote that was either half asleep, or has never really understood the basic principles of digital audio recording. Sorry!

 

David I already conceded that your points about the 2 myths are correct. But apparently there is more going on. The reason I brought it up again is that the original post I reported here explainded things in a succint way I did not think had been made clear before. I have no emotional investment in proving myself "right".

 

And I readily concede that I do not understand the math the way you do and if I am drawing improper conclusions I apologize in advance.

 

Here is my exhange with Paul Frindle:

Quote:

Originally Posted by Ashermusic

Paul would you say this advice holds true for a 32 bit float app like Logic as well as fixed point apps like PT?

 

Yes I would.

 

The intersample peaking reconstruction problem is the same at the output of the mix, as it must be represented in a fixed point output format anyway (i.e. CD or DVD).

Whilst with float it's possible to accomodate internally numbers bigger than flat out, any process that has need to refer to actual real values might be at risk of overload (or unspecified behaviour). Why take the risk?

 

From the point of the headroom issue, things might be different in that an entirely float system from start to finish might handle overs properly - however the meters will be calibrated to a fixed point reference (and will come on willy nilly, whether the signal is clipped or not). Some systems using expansion DSP pass and process signals in fixed point (PowerCore being one example) and may not have any of the float headroom and may mess up with overs.

Again, with 140dB or so dynamic real range at your disposal, why bother to risk it?

 

The most important thing to remember is that recording and processing at lower levels DOES NOT waste 'bits'. It doesn't work like that - all your 'bits' are there all the time at all levels

 

Here is who Paul Frindle is (although he no longer works for Sony)

 

Plug-In Design & Quality Evaluation

35 years of pro audio experience. Mixing and technical engineer in studios in France , London and Virgin Mobiles. Designer at Solid State Logic for 8 years. Founder member of the Sony Oxford Group, responsible for ADC and DAC designs, and all DSP modules such as equalisers and dynamics. Currently role is designing/innovation for all Oxford Plug-In products, and responsible for the operation and audio quality of all plug-ins.

 

So when David says, "Whoever wrote that was either half asleep, or has never really understood the basic principles of digital audio recording." I would say that is a little arrogant on his part. On the points they agree. fine. On the points where they disagree I am then forced to decide who has more credibility. And David simply has not (yet) earned the right to be considered an equal in technical knowledge to Paul Frindle. And that has nothing to do with "clout". It has everything to do with their relative esperience and proven record of important accomplishment. I would feel exactly the same if I were hearing a disagreemnt between a USC PH d in Composition and John Williams.

 

I consider David my friend. I take none of this personally and hopefully neither does he when I say that while David has a degree in computer science and is very knowledgeable when his opinion collides with Paul Frindle's I will accept Paul's, not to mention Nika and Bob Katz.

 

Here is another LONG thread but I recommend reading it. It addresses 32 bit float in the middle. Note the exchange espoecially between renowned engineer and Nika Aldrich who wrote "Digital Audio Explained".

 

And for WHATEVER reasons ALL of them recommend NOT clipping your channels more than necessary. Which I believe David has said agrees with although his heart does not seem to be in it. Since he told me he never uses Pre-Fader metering when he mixes I cannot conclude he takes it very seriously.

 

And with this, I am done.

[David Nahmani]

Well Jay, then I will call it a brain fart. With all of that on his resume, I hope all it was is a brain fart. they happen to the best of us.

 

But don't tell me that when you do a fixed 24 bit recording, levels don't matter because even when you record at a low level you don't waste bits! 24 bit recordings offers about 144dB of dynamic range between the quantization noise threshold and 0dBFS. The lower the level of your recording, the closer you are to the noise threshold. Obviously, with 144dB of dynamic, you have quite enough headroom to play with to not worry about it. But you definitely DO waste bits as you choose a lower level. I don't care if your name is Quincy Jones, George Martin or Eddy Kramer, you still waste bits.

 

When processing, you don't waste bits when processing at lower levels. This is because you are working at 32 bit floating points, and, hold on to your seats, THIS IS THE SAME REASON WHY YOU DON'T WASTE BITS AT HIGHER LEVELS TOO!!!.

 

So in essence, half of that quote is wrong (the half about recording), and the other half indirectly proves my point (the half about processing).

 

The delicate thing is I am not against their recommendation of mixing at lower levels, I just point out that they are recommending it for the wrong reasons.

 

And Jay, please answer this one by yourself. I'm sure you don't need Paul's help on that: how in the world does keeping your signal below the red on the channel meter helps ensure that the signal is not distorting at the input of a plug-in that was placed BEFORE the channel strip!!?? Even if you turn on pre-fader monitoring, all you're measuring is the level at the output of the last plug-in on the chain.

 

So if the problem is "some DSP like PowerCore", then I don't see what it has to do with our discussion on channel meters? Of course, if you mix a low levels, you have a better CHANCE at not distorting somewhere else, but you're still leaving it all to chance.

 

This is kinda like looking at your gas gauge to know if your tires are inflated. OK so when you last took gas, you checked your tire pressure. So chances are, if you still have gas, your tires are still inflated. But then again, maybe the temperature has changed, and your tire pressure with it! Not a very scientific method.

 

Same thing here. Take an Audio track, process it through an EQ, then a compressor. If you boost some frequency bands in the EQ, it is really easy to go over 0dBFS at the input of the compressor, and NOT SEE IT ON THE CHANNEL METERS! That's because the compressor brought back the level to something reasonable.

 

Your Audio signal goes through many different plug-ins before the channel meter, and many after. If we accept that some plug-ins distort the signal when it is above 0dBFS, then you should meter your signal at the input of EACH plug-in. That's how we used to do it in the analog days: meter the audio at the input of each external processing rack.

 

And if Logic's Audio Engine DID introduce artifacts when you hit 0dBFS, then you would have to meter your signal at the INPUT of the master output channel, not on each individual track.

 

On the following picture, all I did was grab 5 Apple Loops and place them on 5 Audio Tracks. Then I placed an AdLimiter on the master output, as many of us do. If you look at the meters, all is well in wonderland. Insert a level meter in the first plug-in slot of your master output. THAT shows you the level on your summing mix. It hits at almost +3dBFS.

[David Nahmani]

Same session, this time a couple of plug-ins on one of the channels (an EQ, and a Compressor). If you look at the meters, all is well in wonderland again, right? Look at the level that hits your compressor, measured by the level meter plug-in: you're hitting +1.7dBFS!

 

This is why I find the recommendation of keeping your meters below the red to avoid distortion when processing your signal with 24 bit fixed point PLUG-INs to be wrong, and dangerous, hence my measuring your oil level with your gas gauge comparison!

[Ashermusic]

1. Well Jay, then I will call it a brain fart. With all of that on his resume, I hope all it was is a brain fart. they happen to the best of us.

 

2. Obviously, with 144dB of dynamic, you have quite enough headroom to play with to not worry about it. But you definitely DO waste bits as you choose a lower level. e bits.

 

3. When processing, you don't waste bits when processing at lower levels. This is because you are working at 32 bit floating points, and, hold on to your seats, THIS IS THE SAME REASON WHY YOU DON'T WASTE BITS AT HIGHER LEVELS TOO!!!.

 

So in essence, half of that quote is wrong (the half about recording), and the other half indirectly proves my point (the half about processing).

 

The delicate thing is I am not against their recommendation of mixing at lower levels, I just point out that they are recommending it for the wrong reasons.

 

4. And Jay, please answer this one by yourself. I'm sure you don't need Paul's help on that: how in the world does keeping your signal below the red on the channel meter helps ensure that the signal is not distorting at the input of a plug-in that was placed BEFORE the channel strip!!?? Even if you turn on pre-fader monitoring, all you're measuring is the level at the output of the last plug-in on the chain.

 

So if the problem is "some DSP like PowerCore", then I don't see what it has to do with our discussion on channel meters? Of course, if you mix a low levels, you have a better CHANCE at not distorting somewhere else, but you're still leaving it all to chance.

 

5. This is kinda like looking at your gas gauge to know if your tires are inflated. OK so when you last took gas, you checked your tire pressure. So chances are, if you still have gas, your tires are still inflated. But then again, maybe the temperature has changed, and your tire pressure with it! Not a very scientific method.

 

Same thing here. Take an Audio track, process it through an EQ, then a compressor. If you boost some frequency bands in the EQ, it is really easy to go over 0dBFS at the input of the compressor, and NOT SEE IT ON THE CHANNEL METERS! That's because the compressor brought back the level to something reasonable.

 

Your Audio signal goes through many different plug-ins before the channel meter, and many after. If we accept that some plug-ins distort the signal when it is above 0dBFS, then you should meter your signal at the input of EACH plug-in. That's how we used to do it in the analog days: meter the audio at the input of each external processing rack.

 

And if Logic's Audio Engine DID introduce artifacts when you hit 0dBFS, then you would have to meter your signal at the INPUT of the master output channel, not on each individual track.

 

On the following picture, all I did was grab 5 Apple Loops and place them on 5 Audio Tracks. Then I placed an AdLimiter on the master output, as many of us do. If you look at the meters, all is well in wonderland. Insert a level meter in the first plug-in slot of your master output. THAT shows you the level on your summing mix. It hits at almost +3dBFS.

 

1. Arrogance on your part IMHO.

 

2. David if you have enough bits you have enough bits so any you lose are not wasted. Maybe this is a French to English issue. Unused and wasted are not synonyms. The dictionary defines it as "to use carelessly, extravagantly, or for no purpose." That is what Paul and Nika Aldrich mean.

 

3. Agreed. So in conclusion lowering the level has no MEANINGFUL effect on the bit vitality. We are mixing in a real world, not a theoretical one.

 

4. Regarding all the rest, I am guessing you have not read through the whole PSW thread. You should. Is it perhaps possible you could learn something?

 

No one, including me, is saying that avoiding channel clipping solves every problem and that turning on pre-fader metering allows you to catch every potential problem. I am saying that based on my experience and what these experienced, accomplished, and technically remowned guys say it helps a lot and that there IS NO DOWNSIDE to it.

 

I cannot tell you technically in detail how the internal combustion engine works. I cannot tell you why the Theory of Relativity is provable and certain aspects of Quantum Physics are equally provable and yet they contradict each other apparently.

 

I don't care. For whatever reasons you choose people, Paul Frindle and Nika Aldrich's, mine, or David's more limited, if you turn on pre-fader metering and keep your channel levels reasonable based on what you see there you will most likely end up with a better mix. And in the real world, that is what matters.

 

Andnow, I REALLY am done with this.