Sample rate: Down converting...is pitch shifting avoidable?

[kindesign]

Hello, longtime reader first time poster....

I've been recording on PCs for a few years now and I still manage to make mistakes. For example, tonight I recorded a live band and accidentally recorded two of the tracksat 96 khz. I am using Nuendo ( a brilliant program IMHO) and a Laya 24/96.

For some reason Nuendo is acting very strange with the 96 khz audio files. It plays and edits fine, but the time meter/cursor runs about twice as fast as the audio plays making it difficult to edit, punch or even know what the heck is going on. Also, some of my plug-ins aren't available in 96 and I am a bit annoyed at myself for making this error.

I recorded the subsequent projects at 48 khz and everything works fine and as expected.

I would like to know if anyone A) knows how to get Nuendo to work properly with the 96 khz sampling rate and B) how to batch convert 96 to 48 without making the Chipmunks latest indie album. (down converting pitch shifts the file...a good trick in the right circumstances, very annoying in the wrong ones) Any help would be greatly appreciated.

Kindest regards,

s.e.t.h.

Kindesign Sound Labratory
Chicago, Il.

[mjau]

I honestly have no idea if this will help (or is even the question you are asking), but I've had to convert 96 and 48khz to 41 for cd's, and I use Voxengo's RateBrain (freeware) to do it. Seems to be a good converter.

[drliebs]

Here I go singing the praises of Cool Edit Pro again...even though it is not my main app. It has batch file options that are easy to use, including sample rate conversions. I don't think converting the sample rate properly will effect the pitch, it might effect the sound though. I am not sure what process does effect the pitch. You could just patch them out and re-record them at the desired frequency.

[cassembler]

Chopping a sample rate in half is significantly "better" for your audio than, say, going from 48 to 44.1. The reasons are simple: if you're cutting the rate in half, you (theoretically) remove every other sample and make a small change to the header of the file. If you have to remove every [(48-44.1)/44.1]/1 samples, the math gets tricky and then you have to "adjust" some of the sample around those that get taken out, and audio is unquestionably affected for the worst.

The effects are most apparent in transient sources. If one of your 96 tracks are drums, you're SOL my friend. It's going to fuck your transients like a bitch in heat.

My advice: use a good two-track editor (wavelab, sound forge)/ downsample the files ONCE and ONLY ONCE and just live with the results.

[cassembler]

PS: if you're going to CD only, track everything at 44.1. If you're going to eventually do a DVD thing, 48 or 96 is ok. But if you're ONLY going to CD, 44.1 from start to finish is HIGHLY advised for the above reasons.

[drliebs]

That explains what happened to me then...I tracked at 48K on a Tascam HD24. The project switched gears on me and ended up getting finished on a PC. So then I loaded them into Pro Tools, down sampled to 44.1 wav files for compatibility with the PC I was using and everything sound less crisp, soft and mushy. What was described above as effecting the transients was exactly what happened and now I understand why. Had I known how it was going to turn out I could have planned for it, but it is just as easy to make a mistake.

If the downsampling is going to effect your audio the way it did mine I would really suggest patching it out and re-recording it. Converting to analog and back to digital with the right sampling rate seems like the least of all evils to me.

Thanks for the lesson cassembler.

[cassembler]

All of this depends directly with the quality of the resampling algorithm; I suspect 90+% of the people on this board (myself included) could not detect the difference between downsampling 96 and 88.2 to 44.1 ASSUMING THE CONVERSION WAS OF VERY HIGH QUALITY.

Hmmm.... perhaps I must qualify my stance on the subject to include "it's better to track at 44.1 the whole way, unless you're positive that your sownsampling is of very high quality."