I always though I understood the use and purpose of the POW-R dither in Logic well, but something that cropped up when I was using Live tonight made me suddenly think I'm unsure. Basically, I went to bounce a file in Live for the very first time, and realised the program has no built in dither option. I understand that a dither is required once upon each incidence of bit depth reduction in a file and why, but I suddenly thought to myself, when I use the dither in Logic to bounce a song to a 16/44 file for use on a CD - just what am I dithering? The source audio is floating around in Logic as floating point 32 bit. Do I need to dither that? Or is it only if my track contains incidences of fixed 24 bit audio? Some of my songs contain no audio files, only virtual instruments. In summary, I guess I don't understand the nature of 32 bit floating point when it comes to dithering, and seek some clarification. Thanks as ever for help! - Wade

> > On Nov 6, 2006, at 4:14 PM, Wade wrote: > > I always though I understood the use and purpose of the POW-R > dither in Logic well, but > something that cropped up when I was using Live tonight made me > suddenly think I'm > unsure. > > Basically, I went to bounce a file in Live for the very first time, > and realised the program has > no built in dither option. I understand that a dither is required > once upon each incidence of > bit depth reduction in a file and why, but I suddenly thought to > myself, when I use the dither > in Logic to bounce a song to a 16/44 file for use on a CD - just > what am I dithering? Hi Wade, since many functions manipulate levels, the 16/44 files benefit if they are mangled in a wider "enclosure". Or, "dimension", if one may say so. I remember in old days we homestudioists had those Audiomedia cards with SD II software. The hardware could only play directly when the file was 16 bit, so any sourcefile you might get was 16 bit. But what a "aah" reaction I got when I saved one of those 16 bit files (from a client) as a 24 bit one and applied same simple edits - it sounded so much better than doing same with same 16-bit material saved as 16 bit files! This story was not about dithering yet but it is very analogue to what happens with 16 bit samples when they get fx-ed in Logics 32 bit architecture, they just may get "taller" and more dynamic. You hardly will have any 32 bit sourcefiles, they all will be 16 bit or 24 bit. So you need dithering anyway. Virtual instruments may be an exception, but I don't know for sure. Of course you will turn dithering on when bouncing off Logic to some shorter words, say, 16/44. Sounds OK to me most times. By the way, the POW-R algorithms have a good reputation compared to others. As about Ableton Live - I always played my liveshows using Logic, not Ableton. Even in concert you can notice that Live sounds muddier, with less dynamics (depends on the music style as well, but it's there). I thought that it has to do with Live's unique pitch handling, but maybe the lack of built in mastering does it's part too. Stick to Logic if you want sound quality. Stick to Live, if you need easily and quickly use same material in different speeds. Can you put some waves Ultramaximizer into Live's master? best maxim