88.2 benefits??

[nst7]

That's even more interesting because he mixes analog (last I checked). It makes me curious why he does it. I mean, if it's going to be converted to analog for mixing, why bother?

[PTUser NYC]

Originally Posted by loosestones

Just spent 2 weeks mixing with TLA, he dithers everything down to 44.1 before he starts mixing... interesting.

He may well sample convert down to 44.1, but he doesn't dither down. Dither is error correction for word length truncation i.e. change in bits, not rate.

[loosestones]

Originally Posted by PTUser NYC

He may well sample convert down to 44.1, but he doesn't dither down. Dither is error correction for word length truncation i.e. change in bits, not rate.

yeah, my bad... that's what I meant to say. His reasoning behind converting down to 44.1 was that it sounded more "rock" and harder in the top end than 48... and yes he's mixing from his tools rig (version 5.x) into his SSL / tape. Whatever sounds good right?

[WernerF]

You can't argue with TLA's success story but good god, even the staunchest proponents of higher sample rate benefits admit that there's no sonic difference between 44 and 48 as it's just a couple of bits. Obviously he can believe whatever he wants to believe but we all might benefit by a quick google search of "power of suggestion". And maybe add "smoke and mirrors".

[Will Russell]

FWIW- I only record at 96k. I believe 96k has been established as the dvd audio standard. (BTW- the theory that 88.2 is better because the math is easier when converting to 44.1 is fallacious- search for discussions by Nika) I record at 96k because with my 192 I/O there is a clear and obvious improvement in quality. I hear it in transients, depth of field, high frequency detail. To me it is worth the extra cpu load and disk space.

It is my understanding that it is easier and cheaper to make great sounding converters that work at higher sample rates.

Test it your self. If you hear a difference at higher sample rates, then go for it! If you don't hear a difference, then don't worry about it!

[JFreak]

Originally Posted by Will Russell

It is my understanding that it is easier and cheaper to make great sounding converters that work at higher sample rates.

Yes, you're right, the problem frequencies (18-24kHz) are audible on 48k but inaudible (24-48kHz) on 96k which makes it a whole lot easier to design stuff that should be "perfect".

Perfection as it is means "nobody can hear the difference" and therefore the optimal sampling rate should be 64kHz which would be just as good as 96k but consume 50% less extra CPU cycles and disk space compared to the 48k -> 96k upgrade.

Close to nobody needs 192k, it's all just marketing hype. And not all plugins work with 192k sessions to begin with, and most that do are RTAS only.