i can definitely tell the difference between the 24bit source and all of the 16bit versions.
Agreed. My listening has included comparing the 16b versions against each other to determine which sound I prefer, but also against the 24b file to determine which sounds most like the master. And they're not necessarily the same.
i found myself going for the UV22HR source, but i can't tell if i like that source better, or if its just because i've heard people talk about it before and i know that its apogee.
I *almost* made the comp blind comp for this very reason. But I decided against it - it seems people are less willing to provide feedback if they think they might be "wrong" (never mind there is no "wrong" answer). In the hopes of eliciting more responses, I left it open.
Edit to add: In the hopes of convincing people to listen double-blind, a new note I added to the original post:
For those wishing to listen critically and compare these samples, please consider an ABX environment - it's really the only way to do it properly. There are several options for ABX plugins or apps if your listening's computer-based (as mine is). If your listening's CD/DVD-based, recommend having someone rename all the samples (e.g. A, B, C, etc.) so you don't know which is which, listen and take notes, and only after you're done listening review which is which.i used to use CEP, then SF...but i'm sold on the uv22hr formula in wavlab.
I agree, UV22HR sounds better to my ears than the CEP native TPDF dithering. I'll post more details from my listening notes a bit later.
Nick / Ed - did you check out the 24-96 dither shootout? I like the fact that it enables one to easily hear the noise signature generated by each of the dithering algorithms. IMO, the differences between some of the dithers are downright stunning. However, hearing the noise signature on the samples they provided doesn't necessarily translate into "better sounding" for our purposes, hence the comp. Anyway...