Why does consolidation &or bouncing tracks=loss of sound quality

[John Kurzweg]

I've held off on this topic for a while because when I'm working I dont have time to A-B stuff as much as I'd like-much less have the time to do any research-But here it is......
I've noticed for a while now(starting with my old Mix System and now with HD)That every time I consolidate my files to get rid of edits & crossfades that my new files dont sound quite as good as they did before the duplication-Most noticeable on low end material and exteme highs.This effect is subtle but its there-Its even more noticeable when you bounce or buss a track with or without a plug-in,-(try bncing or bussing a track and then compare the original with the new bounced track)-And yes all my settings are at 24 bit etc.. I've tried this with and without dithering, noise shaping etc(Im currently doing everything at 96k/24bit with no dither)I am not the only one who has heard this effect-several producers have asked me if I noticed this slight loss of depth and clarity.SO.......Do I need to use a high quality external clock to remedy this?-or is this a clocking issue at all? Shouldn't a consolidation or duplication sound EXACTLY like my files did with edits and crossfades pre-duplication?

[Wolfgang Eller]

Your first post? [img]images/icons/rolleyes.gif[/img]

Have fun! [img]images/icons/smile.gif[/img] [img]images/icons/smile.gif[/img] [img]images/icons/smile.gif[/img]

Cheers Wolfgang

[shoo]

thats like saying after I defrag my drive the tune doesn't sound the same. Get some analysis of some files and check what changes take place. I would have thought you'd only really get slight phase changes instead of high frequency loss.
I think more producers complain about this more than engineers - it gives them something to winge about

[zboy2854]

Well, there's a really easy way to check if it's a real phenomenon or just your ears playing tricks on you. Keep the original track with all the edits, etc., then duplicate that track, and on the duplicate, consolidate it first, then in Audiosuite, Invert the consolidated audio, and play it back at the exact same levels and panning along with the original. If when playing both tracks together you hear perfect silence, even when turning up your speakers to the max, the consolidated track IS identical, and it's just you. If you still hear something, well then that's a new can of worms...

[jaysun]

a very trusted engineer friend of mine, who was practically my mentor, asked me this: "have you ever noticed when you are tracking and then you hit stop and listen back that it sounds a little different?" he says this is clock flutter...so he just purchased a high end master clock generator....

as far as consilidating ive never done the a-b thing but that's worth checking out...

last but not least, after bouncing to disk is where ive noticed the biggest differences...but my gut tells me that topic has been gone over before on this forum...

[sphereman]

I agree with John on this one guys. Sometimes its just not a thing you can measure with phase. Ive refrained from consolidating as much as possible in the past 12 years. Also it seems to be a cumilitive affect over a bunch of tracks and not really just one. Sometimes my mix just starts to suck after ive done a bunch of consolidating, maybe i suck but whatever the reason i just avoid it.

Mills Logan [img]images/icons/smile.gif[/img]

[zboy2854]

But Mills, it HAS to be something you can check with phase, it's inherent in the physics of sound. If a sine wave is duplicated and put 180 degrees out of phase with itself, they cancel each other out completely, and there is no difference, and the flaw is in the mind of the listener. Note, however, that I said completely, which is why I said when doing the test to turn your monitors up as loud as they can go, because if the difference is extremely subtle, you may hear what you think is silence at normal listening levels, but at loud listening levels you may hear what is left over after cancellation, and if you did hear something at that stage, then that is the case for something being wrong with consolidation.

[andrew haller]

I've noticed s slight change in sound as well.
I end up putting alittle eq on it.
Hey John, great work on the Creed stuff.

[John Kurzweg]

I suggest any one who is skeptical of this effect try printing a mix of 32 tracks or more of a song that has never been consolidated and then consolidate the tracks and print the exact same mix-my assistant protools engineer used to think I was crazy until we started comparing mixes-I've also done the phase reversal thing many times-the only thing I haven't done yet is print the stereo bounce of the phase test-that might tell me something-or maybe not-but that will have to wait until me and my P.T.rig are in the same town again.I suspect something may be going on that doesn't show up in the graphics-how accurate is the graphical display in any DAW anyway? Are there some that are superior?I doubted myself for a while back in 1986 when I told a room full of music lovers&stereo junkies that the (at that time 44. 16bt)CD version of Tom Pettys Damn the torpedoes sounded like something was wrong compared to the vinyl and every one in the room thought "my ears were playing tricks on me".By 1992 that didn't sound so crazy to those same people who were beginning to understand the limits not only of digital audio but of the test equipment we were using.Anyway-Pro tools is getting better and better all the time-I just want to know whats causing this and more importantly-How To Solve IT-Digi didnt really have an answer and they suggested I post a topic.

[Rich Breen]

Originally posted by John Kurzweg:
I suggest any one who is skeptical of this effect try printing a mix of 32 tracks or more of a song that has never been consolidated and then consolidate the tracks and print...

<font size="2" face="Verdana, Arial">A more scientific way to do this test is to take a single track with edits, duplicate the track, consolidate the duplicated track. Instantiate a "Time Adjuster" or "Trim" plug on both tracks, and flip the phase on one of them. Then look at the result on a bit scope (Spectrafoo is very effective for this) and observe any bit activity. BTW, you will also need to turn dither off (in other words use the non-dithered stereo mixer instead of the dithered one) or you'll see bit activity in the LSB no matter what.

I have done this test *many* times and in my system, there is no activity down to the 24th bit level, meaning the files are identical. If consolidation does not result in identical files on your system, then you should investigate the possibility of SCSI read errors, or some other system problem.

Best,
rich