Thursday, March 8, 2012
Study: 250 nS is jitter audibility threshold
After getting excited about the possibility of obtaining the equivalent of an SACD transport/DAC combination with pure DSD (no PCM) playback (with a pure-DSD SACD), by combining a Pioneer DV610 disc-player with a Yamaha RX-V371 receiver, it occurred to me that I hadn't considered the combination's jitter performance. To make a long story short, I found that we aren't nearly as sensitive to jitter as some would have us believe. For instance, the developer of my favorite analog-to-DSD converter, the Grimm AD-1, has developed a clock with unmeasurable jitter, i.e. in the sub-picosecond range. Grimm's web-page about the clock features a review in which the clock was substituted for the internal clock on a high-end SACD player with an external clock input, and not surprisingly, it concludes that the substitution produced audible benefits.
However, a 2005 study entitled Detection threshold for distortions due to jitter on digital audio concludes that not even professional audio engineers can hear jitter below 250 nS, and that all consumer audio gear easily meets this standard.
I still consider the Grimm AD-1 converter to be the most transparent analog-to-DSD converter I've ever heard, and some professional recording engineers agree (go here for one example). The fact that I reached this conclusion while listening through a Sony CD/SACD player that cost about $250 in 2003, and which has a disappointing high end even after being left on for a few weeks, seems to indicate that its superiority is not due to the Grimm converter's incredibly low jitter.
Posted by nonastronaut at 12:37 PM