The person who sent me the Audist HUD MX-1 warned me against it. I've already measured the jitter and I am afraid! Very afraid. Jitter measurements do have some correlation to good sound, at least to me. Eventually I will listen to it and report back.
It's interesting. I've grown (or re-grown) to prefer DACs with a more natural sound. By this I mean mostly the R2R chip based DACs of yesteryear; although I really do appreciate the hyper-detail some of the modern sigma-delta DACs provide. The really funny thing is that we've all gotten use to this sigma-delta sound, and forgotten how these older chips used to sound. Most of you younger guys have no idea.
The other day, someone had asked me if there were any other measurements I would be interested in, and I replied "the ultrasonic band past the range of human hearing." I know this sounds like hocus-pocus, and I can see a lot of self-appointed "objectivists" huffing and puffing in outrage (to the point of registering for the sole purpose of telling me I am full of ****) about what I am going to say. The ultrasonic characteristics of gear do matter. The audio engineers I've spoken to, ones who have done this for a while, the ones who I think make good stuff, do look at this range.
Well, I found out fairly recently, that two of my favored DACs, the Gungnir (AKM4399) and AGD M7 (PCM1704), which I feel have the best vocals, best bass texture, and most natural sound (#2 and #1 respectively in these sonic categories) without any treble weirdness, actually had similar performance characteristics in the ultrasonic range. That is a low amount of crud, and with most of the crud way off past the spectrum of human hearing. I thought this was an interesting coincidence - but still hardly enough data to make any generalizations or correlations to the subjective experience.
Now I have yet to see the ultrasonic spectrums of the other sigma-delta DACs, especially the ones which I felt sounded kind of nasty, but it would be interesting to identify their chips and see the pre-filtered ultrasonic spectrum. My guess is that something like SABRE would have a lot of noise and other garbage up there. I might look up some of the spec-sheets of some of the sigma-delta DACs to see if there's any information on this. I do find that it's interesting that SABRE supposedly makes everyone who gets their hands on one sign an NDA and that their spec-sheets are super secret to the public.
Alas the ladder DAC ships are no longer being made in volume. Something like the PCM1704 last time I checked was $75 per piece, and you need one piece per channel. The industry went to sigma-delta because of all the hoopla on having 20 bits, 24 bits, 32 bits. Even though Redbook (CD) is 16 bit and mp3 is, well mp3. Bottom line was well, the bottom line. The sigma-delta chips being cheaper to make. Something about trimming the resistor ladders to be super precise supposedly makes the R2R chips more expensive. I'm don't know much about that, but it would make sense. The level of accuracy required for the resistors would be paramount - the differences in level between LSB and MSB of 20 bits is HUGE - 1:2^20. You do the math. I can only fathom the accuracy and precision required.
One of these days, I'm going to open this stuff up and measure the ultrasonics.