View Single Post
Old 02-23-2007, 03:39 PM   #35
LOSER
Human being with feelings
 
Join Date: May 2006
Posts: 2,373
Default

Dither should always be applied when going from float to int (i.e. going from non discret, well almost , to discrete, e.g. when rendering down in REAPER, which is 64bit float, to 16bit int, for CD for instance), to mask the truncation errror with noise, in order to prevent harmonic distortion. It's also recommended to dither when going from some higher integer value to a lower e.g. from 24bit to 16bit int.

Regarding makeing presets, I will as soon I find some time. (Though I hate presets cause they are always just so rough starting points)

Quote:
Originally Posted by RedStone View Post
anyway, sorry for the thread hijack -
Loser, are any of your dither options similar, or the same as POW-R 3?
No, the are all flat (or simple high passed when HP enabled, or the noise is shifted up to the 20kHz area when noiseshapiong is enabled). AFAIK does pow-r cut the noise in the 3kHz area (cause human ear is most sensitive there) and shifting as much noise up in the 20kHz area (where the human ear can't hear at all) so considering this, I'd say HP and noiseshping enabled is the best (worse) approximation you can get from my dither...
LOSER is offline   Reply With Quote