Ableton Forum

Hi,

I noticed that Live 7 will include more advanced techniques for when performing Downsampling of audio, such as when rendering audio to disk. Good stuff! I'm just curious as to what mechanism was used, anyone knows? I'll be the first to admit I'm a bit of a geek when it comes to things like that.

Was maybe an anti-alias filter applied before the downsampling was performed?

Cheers

I don't think they did anything in that area, where did you read that? the tweak to the audio engine has to do with 64 bit summing, which won't affect downsampling. there is a new render mode (I think, right?).

using anti-aliasing to improve rendering would be more of a cover up than a fix, ya know?

"Aaahh Geek out!
Le Geek, C'est Chic"

Render has a lot of new options now (sadly still no MP3 support?!??! WTF?!?!?) These include Pow-r 1, 2, 3 dithering and a bunch of other options that are pretty cool.

Here's a cool article talking about POW-R dithering options:

http://proaudioreview.com/october99/Weiss-web.shtml and http://gearslutz.com/board/so-much-gear ... -22-a.html

Seems the consensus is to use Pow-R 3.

holy @#$% nice articles!!

Noise-shaping curves are tailored to the limits of human hearing. For 16-bit output, the object is to keep the total power of the curve at 16 bits, but to lower the noise in the areas where the human ear is most sensitive. In critical bands it is possible to lower the noise to 20-bit levels.

pure poetry.

Tone Deft wrote:I don't think they did anything in that area, where did you read that? the tweak to the audio engine has to do with 64 bit summing, which won't affect downsampling. there is a new render mode (I think, right?).

using anti-aliasing to improve rendering would be more of a cover up than a fix, ya know?

"Aaahh Geek out!
Le Geek, C'est Chic"

I think I read it in a Newsletter, it mentioned 64 Bit summing, Pow-R dithering and I think optimized sample rate converison, which I assumed (possibly naively) referred mainly due to downsampling and the aliasing it can sometimes cause. I'll have to take a look and see if I still have that in my inbox. Yes, possibly 'anti-aliasing' was the wrong term, thats why I was curious to know what they have done in that area. The geek in me was alive and well this morning!

nebulae wrote:Here's a cool article talking about POW-R dithering options:

http://proaudioreview.com/october99/Weiss-web.shtml and http://gearslutz.com/board/so-much-gear ... -22-a.html

Seems the consensus is to use Pow-R 3.

Interesting article. That guy Bob Katz is a bit of a guru afaik.

hey man, you're right, nice catch! from the 10/7/07 newsletter (October 7th for the Eurofolks and their weird shorthand).

ABLETON LIVE 7
==============

Version 7 renews the core of Live.

* The enhanced audio engine improves fidelity with 64-bit mix summing,
POW-r dithering, optimized sample-rate conversion and other
advances.

gaaah!! I really want to see their white paper on the audio engine.

sample rate conversion ALWAYS produces artifacts, some of which are due to aliasing. looking back at Gerhard's speach in that video and his emphasis on the team taking a hard look at the audio engine my guess is that they improved the SRC (sample rate conversion) code and I think it's a safe bet that all SRC code has anti-aliasing filters.