Don't dilly dallee or dither !

wiz :
1. If you don't export a mix in the DAWs native resolution i.e. the float format it works in, but instead in e.g. 24/96, it is highly recommended to dither with TPDF, because you will re-quantize the audio and have quantization distortion if you don't at the time you write the 24/96 format file. The internal engine of the DAW is not 24 bit the format, no matter, if this is the projct setting. If i understood everything correctly.

2. Same, because again, the audio in the DAW native float format needs to be re-quantized/changes bit depth for the DACs. I did not read anything about this in my DAWs manuals, so i do not know, if it applies the preset dither on the outs to my console during mix time.

2a. If the DAW doesn't do it, theoretically, yes, dither should be applied to the console feed with DACs and a plugin could to the job. There are several mastering plugins that have a dither option. Bob O. mentioned his preferred one a while ago but i don't remember fully (Sony Oxford?)...
I will try to find out, how my DAW handles it, and in case it's needed, will have to look out for a plugin option as well... e.g. the post no.1 in this thread. There might be even free alternatives for me, triangular dither is not a proprietary thing...
It might be not too hard to find out.
Two loop recordings from "audio out" to "audio in" with the DAW and a program like "Virtual Audio Cable" that only streams the driver pins data. One with the internal dithering option set to "none", one set to the dither type you normally use, for me Triangular/TPDF. If the phase aligned recordings null with phase reverse, the dither has not been applied to the supposedly analog out by the DAW. It should be applied, though, as a quantization error distortion would happen at the DAC. If i understood everything correctly.

3. Uhm, yes. Doesn't sound very optimistic, right?

4. See 1. so i guess "yes" is the correct answer. From all i read, if you write to file and dither, TPDF is preferable to all other dithers, if there is any chance that another dither will be applied successively, which will definitely happen in the mastering house. At least this is covered in my DAWs manual...

Let's see, if i did have a final "Aha"-effect about dither. I should by now know good enough. But i seem to forget regularly about it. But as Bob stated - the software developers are those to blame. Dither should not be of any importance to us, if the DSP engineers do their jobs appropriately.
I guess this is also the reason, why the DAW developers and manuals cover this topic so sparingly.

There has been an interesting question brought up about "how does quantization distortion sound?".
This is interesting. Simply deactivate dither in a typical project export and compare to the dithered export. There is a difference on good monitoring between them. And the undithered export looses if i ABX them. Always. I tried this. Not tooo obvious, but it does...

Dither should only be applied when lowering the bit depth — so nothing is needed for 24 -> 32 bits. Yes, for 32 -> 24 though.

It's AU, though, right? I guess I could port the AU plug in through Bluecat Patchwork for PT10.

if its aax 64 it can run in pt 11 +.

According to Jim Johnston d*ther was an integral part of Bell Labs' digital telephone technology from the 1950s that all digital audio is derived from. The math doesn't support the notion of d*ther ever being optional which brings into question how much a lot of programmers actually know about audio DSP math. Apparently the appearance of AES papers about the importance of d*ther was quite a face palm moment for the actual experts as was Sony's "perfect sound forever" ad campaign.

Dave Collins started it as a protest that we shouldn't even need to talk about it. Dave and I were both very influenced by Armin Steiner and Doug Sax. We can often finish each others sentences.

After Armin's incredible Sherwood Sax equipped studio went bust, he moved to recording film scores and also joined in a partnership, I believe with Bruce Botnick, that rented early digital gear to all of the mastering studios in Los Angeles. Dave ran the gear for all of them and ended up replacing Bernie Grundman at A&M when he left to start his own facility. Dave was among the handful of us learning about digital audio from each other on the internet during the late 1980s. He also has a tremendous sense of humor so I follow him around the net.

You need to d*ther down from the internal DAWs float format to any of the consumer formats files, no matter if 24/96, 24/88.2 or 16/44.1 or whatever if you did the slightest change (like the mentioned "0.1dB volume change"). All of these formats mean a change to a lower bit depth that the consumer DACs will swallow, the DAWs internal bit depth is always higher.

You don't need to castrate the audio for any technical reason. I slipped a client some 96x24 AAC files once and he was blown away by the quality even though I had never told him they weren't 44x16.

I've been been experimenting with dither.
If an internal bounce happens itb in a 32 float environment, some might argue whether dither is needed. 
If you go from 32 float > 24 for mix to be sent somewhere, of course dither.
(though I'd tend to just give the Mastering folks a 32 file and let them do their thing)

As well, Sonnox dynamics dither on all channels going "out to the console via D/A's" lately as well (for mixing-HW & re-capture into DAW, or to tape) 
That said, It's hard to imagine all pros, let alone hobbyists, are really applying dither to every single "out" to their desk.

Interesting considerations. 

 This is very confusing topic ,  but from what I understand quantize noise shows up "harmonically" in a very un harmonicly pleasing way going upwards into the frequency spectrum,  The noise it makes is in response to the signal it's quantizing, it's not random, and it's ugly, without dither, that noise buildup track after track, can become very very ugly and give you the hard to quantify "swirling space monkey digititus",  I have some pretty good reading material on it I will post later, my keyboard on my computer took a Poo,  i'll have to put it up later .

Ok, i picked up another keyboard, so heres some good reading on digital filters, and here's a thread started by one our favorite RGO guys, with some serious..., and I mean SERIOUS heavy weights chiming in on the subject(Sir Bob O included)

repforums.prosoundweb.com/index.php/topic,5739.0.html (apparently you have to copy this entire link and paste it in your address bar, for some reason when you click it, it leaves off the tail end?)
www.users.qwest.net/~volt42/cadenzarecording/Filters.pdf


here's slightly "smug" Monty doing a great job of explaining things until he says "no one ever ruined a great recording by not employing dither", he fails to realize that while mixing, some people use upwards of 100 tracks with a zillion plugs that have quantization noise all over them, the signal dependent, nasty upward harmonic noise build up, and subsequent cascaded passing over one another build up to VERY audible artifacting levels.. aka "swirling space monkey digititus", I believe this is what bothers me so about digital recordings. 

Or, I'm totally wrong, and it's something else entirely haha