Why Can't I (#2) ???...[with the right gear]

  • Thread starter Thread starter mark4man
  • Start date Start date
mark4man

mark4man

MoonMix Studios
Sorry to expand on chazba's thread, but...

My ME uses a newish method for mastering that gets things louder without compression. He outputs the stereo files or stems via one world class D/A...pushing the output 'till it's at the brink of distortion...then backs it off a bit...& captures w/ another world class A/D...clipping the input just a tad.

So...mastering engineers:

A) Have I described this correctly...is this the new method?

B) What...save for experience...would prohibit a decent mix man from accomplishing the same thing with his own tracks?

I've been looking at the LynxStudio Aurora 8...which was heralded by Martin Walker in June's S.O.S. magazine as having some of the most pin-point accurate imaging he's ever heard. Can what I've described (if accurate) be accomplished by one box, like the Aurora...in an output to input loop? Or does one need separate boxes?

Thanks,

mark4man
 
mark4man said:
What...save for experience...would prohibit a decent mix man from accomplishing the same thing with his own tracks?
Click to expand...
Obviously if you have the "right gear" and duplicate the procedure exactly, there's no reason why that can't be done in a diving bell as well as in a mastering studio.

But that's not mastering.

Mastering is about setting the correct volume, not the loudest. And even then, that's only a relatively small part of the mastering job. One still has to polish the sound of the mix, which that procedure does nothing to address. And then there's the PQ editing, the assembly of the tracks into a coherent album, etc.

The idea about having a different set of ears and/or having them in a different environment than the mixing is to optimize the ability to polish the track in terms of response, dynamics, gain, and emotion by performing the second half of post production in a fresh environment where any coloration that may be inadvertantly introduced by the particular character of the mixing environment and bias or fatigue by the ears of the mixing engineer are eliminated.

G.
 
Sorry but that procedure isn't close to what mastering is about, it's just a technique (and a bad one at that IMHO) for limiting. Clipping converters can introduce alias frequencies that can make music sound harsh. That said, many do it. The first I've heard of it was in the 90s (from Greg Calbi I believe), so I wouldn't call it a new technique.

Mastering is about listening to music with an objective and experienced ear to help determine how the music can sound its best for the final medium it is going to be released on.

That doesn't come in a box.
 
I'm not sure you could do that with the lynx. I think most guys that do this use 2 computers. Because undoubtedly you would want your mix at a high resolution, whearas you want the captured master to be at 16/44.1. I don't know of a device that can do two formats duplex but then again I haven't really researched it either. I have tried this myself and not had good results.
 
SouthSIDE...

You're absolutely right; (& that's the most wonderful explanation of true mastering I've ever read...I'm completely serious !)

But what about the guy who comes to you & is 100% sure that the tracks have been mixed to sound just perfect the way they presently sound...the separation, EQ & level of all instruments & vox are just the way they should be...& he doesn't want to mess with that sonic signature by brick walling or compressing...he just wants it a little louder (like maybe 3 > 4.5dB)?

See…we’re all spoiled (down here at the project studio level) by floating-point DAW’s. It gives us a false sense of burgeoning RMS we can’t really achieve in the real world (the world of fixed-integer 0dB ceilings.) We hear these nice, loud, clear F-P mixes…knowing that’s the way it should sound; & wish the mixdowns could sound identical. The only way to accomplish that is either to just draw back the master faders to legal level & then tell your fans to turn up the CD…or use the new method (which I’m sure has it’s own inherent limitations…spurious artifacts at the clip points & that sort of thing.) Then you hear guys that rave about certain pieces of gear that do this phenomenally…like that hdwe. piece is the Holy Grail.

mark4man
 
mark4man said:
SouthSIDE...But what about the guy who comes to you & is 100% sure that the tracks have been mixed to sound just perfect the way they presently sound...the separation, EQ & level of all instruments & vox are just the way they should be...& he doesn't want to mess with that sonic signature by brick walling or compressing...he just wants it a little louder (like maybe 3 > 4.5dB)?
Click to expand...
Why you picking on me? ;) :D

J/K. I'll invite others to answer this as well, but since you asked me, here's how I'd personally answer it (YMMV).

I've not tried that procedure you've described, but just thinking about what it is and how it would work, it sounds to me like just using some gain staging trickery and minor clipping along to way to maximize volume. It's still boosting and limiting, it's just doing it via a different method.

And like Tom already said, that's not a colorless method any more than using a limiter is. He's absolutely right, clipping a converter introduces (or at least can introduce) digital aliasing and artifacting that is not very often considered "pleasing". Perhaps if done right, it may be relatively inaudible, but the exact same is true of "standard" limiting. And if the right limiter is used in the right way, the resulting "coloration" could wind up much more pleasing than the client might realize.

If a client walked up to me and set me up with the situation you describe, my response would be, "I'll be happy to do it, if it can be done." If somebody wants to be foolish enough to pay hard-earned cash for a lousy 2 or 3 decibles, that's their problem.

But I'll also advise them from the outset that there's no guarantee that their mix, as good as it may sound now, will still sound the same when we squeeze another 2 or 3 dB out of it. It's not always a limitation of the gear, it depends greatly upon the quality of the tracking and the mixing more than anything else. Some mixes may sound optimal at, say, -14dBRMS (just to pull out a number), but when you try pushing them harder than that the mix starts falling apart because the lower-level stuff which doesn't sound so good in and of itself begins to gain in prominance as volume increases and dynamic range decreases. This is true whether it's done via a limiter or a clipped converter.

G.
 
If anyone did that to one of my projects, I would politely ask for my money back. I'm sorry, but it just sounds like kind of an irresponsible method.

Maybe if you wanted your stuff to sound like crushed, distorted dog shit, ala RHCP By the Way, Californication or something, then by all means, knock yourself out.
.
 
Please do not take offense to this, but if anyone told me that they were "100% sure that the tracks have been mixed to sound just perfect the way they presently sound...the separation, EQ & level of all instruments & vox are just the way they should be" I would say that they have spent too much time on that project and more than ever need a new set of ears involved. In this situation what i would do, if it were me, is turn it over to my ME of choice. I would explain to them that at least in this instance at least that what I am really after is just a little level increase. Then I would ask them to make two files. One with just my request, and then one with what THEY thought should be done. 98 out of 100 times I bet I would like the ME's version better than my request. First off, I am a tracking and mixing engineer, not an ME. Secondly, when I am fully involved in a project like that, I, like so many others, lose touch with the simple stuff because I am so worried about the little things. A fresh set of ears will often pick those things up almost instantly. With a little knowledge, experience and equipment on their end those things are usually pretty easy to fix, but have a muach larger impact on the final product than I would have ever guessed.

The whole propcess is kind of a catch 22. Every time I think I have a song just perfect and that the ME won't be able to make it much better, they exceed my expectations. The truth of the matter is, the better I do on my end, the better the ME does. When my mixes are already that close, it seems like the ME has to do less, yet nets a much larger impact, in a positive way on my material. The most important thing in these situations though is to make sure that you have the right ME for YOU.
 
$ .02

sounds to me like just another way to mangle some audio and confuse the client into believing something was accomplished..... that said.... if you like it doit..... minor question though to other commentors.....are you sure that aliasing is in volved...though that was a harmonic artifact introduced when 1/2 sampling freq is close to an overtone and we used to use decimating filters to "eliminate" but now with dirrect conversion chips no longer an issue.... not wanting to argue just clarify if you can.... thnx
 
dementedchord said:
sounds to me like just another way to mangle some audio and confuse the client into believing something was accomplished..... that said.... if you like it doit..... minor question though to other commentors.....are you sure that aliasing is in volved...though that was a harmonic artifact introduced when 1/2 sampling freq is close to an overtone and we used to use decimating filters to "eliminate" but now with dirrect conversion chips no longer an issue.... not wanting to argue just clarify if you can.... thnx
Click to expand...

I've run a few tests by clipping internally and viewing the results in spectrafoo. Definitely were there.

Basically when you clip audio you are creating square waves which add additional harmonics that may be above Nyquist and then become aliased if not handled correctly. So you're correct as well.

Check out the Signal Overload section of the following for other info:
http://en.wikipedia.org/wiki/Anti-aliasing_filter
 
Last edited:
Assuming one is using the best converters there are and any aliasing is indeed handled properly or even avoided altogether, one still has the artifact of clipping itself. The idea that a clipped waveform is less audible or better sounding than squeezed waveform (as via a limiter) is a bit ludicrous to me.

The bottom line is there will always be those young ears that feel that paying a few percentage points of signal integrity in order to get one or two decibels of loudness is a worthy bargain. They'll just have to learn on their own.

G.
 
I disagree. To my ears, clipping will often sound better than limiting because the volume will not be pumping. but often limiters sound better, too.
 
FALKEN said:
I disagree. To my ears, clipping will often sound better than limiting because the volume will not be pumping. but often limiters sound better, too.
Click to expand...
See, to me volume pumping is nothing more than a natural sign that one is squeezing their mix too damn hard. Even a couple of dB before such a breakdown usually sounds like artificial crap processing to me, regardless of the instrument used to squeeze it.

This is exactly what I meant by how some ears are willing to make the trade and others aren't. For me, once the dynamics are flattened past a certain point, the sound quality decreases. For others, as the RMS goes up, so does the quality.

It's to these folks I say, "Pull out your Bic lighters, folks...I've got some tuned port speakers and 5-band EQs to sell you."

G.
 
FALKEN said:
I disagree. To my ears, clipping will often sound better than limiting because the volume will not be pumping. but often limiters sound better, too.
Click to expand...
Clipping and Pumping are equally annoying to me. :mad:
 
I have to tell you guys (the limiting advocates)...this seems to be a legitimate method (converter looping...or whatever it's called.)

I'll give you an example: I have a 60's British Rock style number with a really nice Tenor Sax solo at the break. The aspect that makes the sax just shine is the fact that it rides above the mix, slightly dominating the rhythm section. This allows the reverb to sound natural; & the sax really soars as a result. The way it sounds in the floating-point mix environment is perfect.

But every type of brick walling ever tried on this composition changes the character of the sound...obviously because the sax represents the loudest portion of the waveform during the solo; & it gets flattened with compression or limiting...even in the most transparent attempts, it loses body & sparkle.

I just got my tracks back two days ago; & the (posted) method my ME used, which just makes the mix louder...retained all the bite & beauty of the sax. It was clearly the superior approach for this particular song...and as such...I can see a certain validity to this method. I wouldn't be so quick to discount this process...the clipping in the stereo master's waveform is minimal; & there isn't much digital harshness to speak of. I'm not saying it's a miracle method, but to my ear, so far, it beats the pants off typical compression/limiting.

mark4man
 
mark4man said:
But every type of brick walling ever tried on this composition changes the character of the sound...obviously because the sax represents the loudest portion of the waveform during the solo; & it gets flattened with compression or limiting...even in the most transparent attempts, it loses body & sparkle.
Click to expand...
Of course it does; when pushed too hard, brick wall limiting sucks. I don't think you'll find many people who'll dispute that.

I don't believe many of us - well at least not me - are saying that the method you describe is illegitimate, just that it's just another way of skinning the same cat.

Again, I think where we may be talking past each other, is in just where we each sit when it comes to the worth and importance of extreme RMS level. Correct me if I'm wrong, but what you're basically saying is that with the method you describe, one can push the RMS a couple of dB further using that method than one can with non-extreme limiting. Put another way, if one uses a brick wall to try and push the RMS as high as this guy is by misusing his converters, the limiters would make it sound like shit.

I don't necessarily disagree with that. You're probably correct there. My point is that if you use "standard" compression methods (for lack of a better term) and *don't try to push them harder than they want to go*, that you'll get a better sound than you will with the digital clipping. Will that sound have as high of an RMS level as the clipped signal? Maybe not, maybe it will only be -13dB and not -10dB. Who cares? It sounds better! Play the two back at the actual same playback volume and the -13 one will sound better because it's not pushed harder than it wants to go, whereas the -10 one is being pushed into digital clipping.

As I said before, there are those willing to sacrifice sound quality for a couple of dB, and there are those to are unwilling to make that sacrifice. It's a matter of which side of that fence one wants to fall upon.

G.
 
All he is doing is turning the volume up.

That and you get the sound of the converters, that's about it. It's possible his AD converter has a limiter circuit in it, so there might be some limiting involved.

But really he's just turning up the volume with possibly a peak protection limiter involved.
 
I will take limiting EVERY time over pure converter clipping. First off, anything in clip on the converter is basically limiting anyhow. Then again, if you give me a choice of clipping some Weiss converters a little and using a DBX 166 for mastering my mix, I may just choose the Weiss. As far as the clipping the converters method goes, that sounds very similar to me to normalizing a file to a peak volume of -.1, and then adding 2 or 3 db of gain to the file, and then renormalizing the peak volume to -.1 again. End result? A digital emulation of a clipped converter (to some extent). Basically, some converters implement certain forms of limiting and compression in their overshoot ranges to reduce the amount of digital garbage created by positive dbfs values. This in and of itself is just a limiter or compressor, just maybe t\not the brand you are used to.
 
I'm afraid this is quite common. Have you looked at the waveform of a recent (meaning within the last few years) release? Can be quite shocking for those of us trying to avoid digital overs, pondering whether two or three consecutive full scale samples are to be considered overs. The worst examples that I have in my record collection are definitely RHCP - Californication and Mars Volta - Deloused in the Comatorium. Those are so incredible distorted, you'll find as much as 50, 70 or even more consecutive samples hitting the roof. And I mean, these recordings are done by the "pros of the pros". :eek:
 
mark4man said:
I have to tell you guys (the limiting advocates)...this seems to be a legitimate method (converter looping...or whatever it's called.)

I'll give you an example: I have a 60's British Rock style number with a really nice Tenor Sax solo at the break. The aspect that makes the sax just shine is the fact that it rides above the mix, slightly dominating the rhythm section. This allows the reverb to sound natural; & the sax really soars as a result. The way it sounds in the floating-point mix environment is perfect.

But every type of brick walling ever tried on this composition changes the character of the sound...obviously because the sax represents the loudest portion of the waveform during the solo; & it gets flattened with compression or limiting...even in the most transparent attempts, it loses body & sparkle...
Click to expand...

Every dog has it's day, just like every technique.

Distortion has it's uses, clipping is another form of distortion. I think that when you are saying things like sparkle and bite, you are hearing the result of this technique adding additional harmonics, not the fact that brickwall limiting is very different in the way it treats transients (with soft limiting/knee turned off).

Personally I think that there may be more controllable ways of achieving this type of distortion, with a possibly better result. However if you like it, and it accomplishes your sonic goal, job done.
 
You must log in or register to reply here.
Back
Top