Track Lag (solved)

momo wrote on 5/22/2002, 7:27 AM
Hello All,

Fingers crossed, touch wood, etc., I think my track lag problem is solved. For anyone interested in the conversation, see post "Track Lag (revisited)"

SonicPCH had suggested that I use a 48 KHz sample rate since my SB Live sound card does not support 44.1 KHz. I made this change, and was able to record almost 20 tracks last night - every note stayed right where I put it, and the only reason I stopped was because it was 3:00 in the morning. Prior to this, using a 44.1 KHz sample rate, I’d be lucky to get even half that many tracks down before samples would begin dropping and lag became apparent.

The topic of track lag seems to come up in here from time to time, so I post this as a new topic in the hopes that it may help other SB users.

Thanks again, Sonic.

momo

Comments

pwppch wrote on 5/24/2002, 1:36 AM
Cool and glad to hear it solved your problem.

Sucks that you have to do this, but that is the nature of the card you are using. Working at 48 kHz vs 44.1 is really only a harddrive space problem.

Peter
TeeCee wrote on 5/24/2002, 9:12 AM
Harddrive space problem, plus if (when) the audio makes it to a CD, it will have to be 44.1 kHz. So, he will have to convert it sooner or later and potentially deal with conversion artifacts in his audio. Or pick up a card that does 44.1 kHz right.

TeeCee
momo wrote on 6/2/2002, 8:37 AM
When I (eventually) get a better card, it will do way more than 44.1 KHz 'right.'
;^)

As far as HD space goes, I just installed another 80GB HD which will be used solely for audio, so I'm not concerned about space for a little while at least.

Stepping down from 48 to 44.1 for CD audio (ie: rendering mix to wav) hasn't introduced any obvious artifacts as yet. But increasing the number of tracks from < 8 to now well over 20 has been, literally, an epochal event - I've not needed to use any more than 25 tracks over the past few weeks, so to be honest, I don't know how many I can pump into it now. It's a whole new world! Thanks again to everyone who helped.
TeeCee wrote on 6/5/2002, 12:51 AM
"Stepping down from 48 to 44.1 for CD audio (ie: rendering mix to wav) hasn't introduced any obvious artifacts as yet."

To you and maybe with the material you are working with. I don't advise doing this any more than you have to.

TeeCee
momo wrote on 6/5/2002, 10:21 AM
What SHOULD be the most obvious artifacts?
TeeCee wrote on 6/6/2002, 1:39 PM
You will obviously lose some upper range as you drop to a lower sampling rate. That part is not guaranteed to be heard because it is so high. Depending on the anti-aliasing filters used (or not used) you could lose some audible high end that would have been recorded from scratch at 44.1kHz. In the case of no anti-aliasing filters (which should only be if you checked a box to tell the software not to use them for effect or because you didn't understand them), you should get frequencies over 22.1kHz folded down in an inharmonic way (depending on the algorithm used, I don't know what's popular or works good). You will probably have math round off damage done to the lowest bit that did not need to occur.

What this will potentially result in is a sound file that just does not sound as good as it did prior to conversion. It's unnecessary work done to a sound file. Digital audio transfer should be perfect. Digital audio computations are not. It's finite math.

TeeCee
momo wrote on 6/6/2002, 4:05 PM
Okay TeeCee, thanks for that. So, what happens when one records using hardware capable of 96 kHz resolution – wouldn’t the final rendering to 44.1 kHz CD audio exhibit similar issues, and if so, wouldn’t a transfer from a 96 kHz source be considered imperfect? I understand that higher sample rates produce better sound resolution - I ask only because I’m presently looking at the Aardvark Q10, which (presently) can only sample at 44.1, although drivers supporting 96 kHz are supposedly in the works.

momo
TeeCee wrote on 6/11/2002, 11:35 AM
The answer is in your question. The hardware is capable of 96kHz sampling. It's not locked a 96kHz. Some of the Creative Labs products are locked at 48kHz and resample in software (your sync issues, etc.). If I were to sample at a higher rate than 44.1kHz, I would sample at 88.2kHz. Downsampling should be a much simpler operation (technically, you could throw out every other sample). I have heard that sampling at 96kHz and resampling at 44.1kHz is worth the artifacts, but I've also heard that it isn't. My hardware is capable of recording at 96kHz. I record at 44.1kHz, 24 bit exclusively.

TeeCee