96 versus 88.2 khz

[Sambuca]

I've read different opinions on this but still do not have a clear answer. As I will eventually be bouncing down to CD (44.1), which sampling rate will yield the best results.

[Kurtstudio]

Digi has spent countless hours doing great conversions down from 96 to 44.1 if you are concerned 88.2 would be the prefect conversion rate, but not very practical for pop music, though I have done session at that rate for clients. Just make sure you dont' bounce to disk for the conversion step, let mastering do it or use the Step of, selcect your audio region "Shift , option K "and do the conversion there. Better use of 32 bit processors and quality.

[Sambuca]

It's a Classical/crossover project so I need to capture as much detail as possible. I've read 96 is the way to go as long as you are using top notch sample rate convertors to get to 44.1 and that the 88.2 to 44.1 'cutting-in-half' thing mathematically really isn't valid.

Thanks.

[Studio Crash]

I would like to know the reason behind the 88.2 "cutting in half" is not valid statement. It seems simple enough to me. If you mathematically remove every second sample, wouldn't that be exactly the same as if it were originally recorded at 44.1? What's the difference?

Maybe some info could be lost by the phase shift any mix eq might do after the original recording moving transients to in-between individual sample points? Even so, the actual resolution difference between 88 and 96 seems like it really wouldn't make a noticeable difference after converting to 44.1. Am I wrong?

[Kurtstudio]

Since it is a classical pop record, a directions I have had lots of experence in, I'd keep it 96 k all the way to mastering. They will have the state of the art converters thus your end product will be at it's maximun performance and quality. Have Bob Ludwig master it, he's the best. Good luck!

[Kurtstudio]

Since it is a classical pop record, a directions I have had lots of experence in, I'd keep it 96 k all the way to mastering. They will have the state of the art converters thus your end product will be at it's maximun performance and quality. Have Bob Ludwig master it, he's the best. Good luck!

[rnt_u_listnng]

I would like to know the reason behind the 88.2 "cutting in half" is not valid statement. It seems simple enough to me. If you mathematically remove every second sample, wouldn't that be exactly the same as if it were originally recorded at 44.1? What's the difference?

Hey Crash, this seems to be the day for rehashing old technical issues. On another thread, we covered recording levels...again. The answer to your question isn't quite as obvious as it seems. Converters upsample the data to the least common multiple before downsampling, so the answer is not simply "take every other sample". Do a search on the DUC, this one has been discussed quite thoroughly. The quality of the converter is more important than the particular sample rate.

Hope this helps.

John