Quote:
Why doesn't Schiit release jitter numbers for any of their DACs?
Because it's a huge conspiracy! Because it's 120,000 ps! Because we're incompetent engineers! Because we're under contract from the US government to maximize the amount of jitter in our systems to mask the subliminal messages broadcast by Cthulu and the Elder Gods thanks to their awakening in 1945 by the first atomic tests!
Or, er, well, none of the above. The reality is a lot more complex. Three reasons, really:
1. If you're familiar with the CS8416 datasheet, you won't be too surprised by Bifrost's jitter numbers.
2. Jitter isn't necessarily the be-all end-all one true measurement for digital audio. And it's easy to misinterpret. For example, the RMS jitter number for our USB input is lower than SPDIF (as you'd expect), but the spectrum is entirely different. What's most important?
3. Jitter is hard to measure accurately at low levels. Most people would have you think it's as simple as hooking it up to the AP or DScope and getting a chart out. But hey, let's look at the AP and Dscope's residual jitter numbers: 600ps for the AP 27XX and 1000ps for the AP 525x and Dscope. Not exactly 1ps territory, hmm? That's why we have not one, but two Stanford Research SR1 audio analyzers, one kitted out with Option 3 and 4 (1ppm internal reference and ultra-low jitter analyzer with 8ps residual jitter.) The only better option for measuring jitter than the full-kit Stanford is a $30,000 piece of gear that only measures jitter, which will get you down to 2ps or so.
Bottom line, yeah, we'll eventually have some properly-caveated numbers for Gungnir, but we don't think people should be buying gear based on one single measurement. And Anaxilus, if you'd like to do some measurements after release, feel free! It'll be interesting to see how they correlate with our own, and it'll give me more confidence if they're in concurrence.
PS: And, no discussion of jitter would be complete without noting that Mike Moffat was one of the first people to theorize jitter was a cause of digital audio degradation, was one of the first to measure it back in the 1980s, and was the first to use VCO PLLs to regenerate clocks and reduce jitter. It's not like we just fell off of the turnip truck here.
