With all the advances in DSP power and software improvements, if one decides to use HD (Accel) with PT TDM 6.X and mix ITB, should one use the (stereo/surround) DitheredMixer, or the "vanillas" that are the *default* upon fresh install?
Others may have an opinion as to why one should sound better than the other. I've just always used the regular non-dithered mixer. I think it sounds great + I see no reason to switch. I have not done the test you mentioned. Has anyone else. Lynn? I'd be curious to hear the results.
It's a fairly detailed article detailing how the mixer works in PT, however, it leaves a number of questions unanswered.
I'm not sure how things have changed since the last time I looked into it, but when the dithered mixer was first introduced, it did a little more than just dither to 24 bits at the output of the mixer, it also dithered to 24 bits any signal going to an internal buss, the 24 bit output of a bounce, and was designed to offer 36dB of overhead versus the normal mixer's 30dB. The dithered mixer also only functioned in stereo...the surround mixer was completely different and didn't dither at all. But like I said, I haven't looked into it in a LONG time...I wouldn't doubt it if some of these things have changed since the introduction of HD and the *new* dithered mixer plugin.
I think the most important thing to note is that the theoretical noise floor of 24 bits, dithered or not, is close to 140 dB...where the dynamic range of most super high end converters is still only around 120 dB (not sure about the HD 192, pull out those spec sheets to find out for me)...so the noise floor due to the converters going in can't be lower than 120. When truncating or dithering to 24 bits, chances are that the real noise floor (probably around the 20th or 21st bit) is sitting far above the theoretical 24 bit lowest noise floor rendering differentiating b/w the two mixer plugins probably impossible (at least as far as the dithering is concerned).
If a difference is noticed b/w the two stereo mixer plugins in PT, it probably has more to do with the differences having nothing to do with the dithering, like the added overhead...but like I said, I don't even know if that's how it works anymore.
I know this doesn't really answer your question, but at least I hope it may give some insight into why a difference is noticed if noticed at all. I'm interested in someone pulling off a true blind test to find out too.