TO DITHER OR NOT TO DITHER
that is the question.
edit : let me add to this actually. .
aside from the question of to dither or not in general,
more specifically, if I have a project where all the samples in it and vocals are 44khz 16bit anyway why would I dither?. .there would be no downsampling right?. but the program I am mixing in mixes in 32 bit I think so each sample gets resampled to the mixing frequency. although with each sample it gives me the option if i want the sample to be resampled to that frequency. .I make beats in FLstudio so thats the program I'm talking about if that helps, which all samples are 16 bit anyway, but the vocals I put in are 24 bit and 88.2. . soo I dunno whats the best way i should be doing things. arrrg, please help
Ok I'm not sure that I completely understand. If you are doing a project in 44.1k then the individual tracks will also be at 44.1k. It won't differentiate. The program that you are mixing in(you didn't specify which one) should be able to change the bitrate at which you mix. So that being said if it is set at 32 bit then you need to dither. If you don't want to dither then set it to 16 bit. I hope that helps.
you want to dither. you are confusing sample rate conversion with bit depth. I can't think of any programs that don't change the bit depth when you alter the audio. if you have a 16 bit track and you change the volume, it will become a 24 bit or 32 bit or whatever bit that the program operates in. If you print that track that you changed the volume on without dithering, then you trunkate it to 16 bit. Even if your tracks are recorded at 16 bit and you then "mix" them, you are generating a larger bit depth and you should then dither the mix to the desired end bit depth.
Michael Fossenkemper wrote: I can't think of any programs that don't change the bit depth when you alter the audio.
Just thought I'd be a smart ass...
Sonoma and Genex both keep DSD at 1 bit...