Well i admit my mistake about the 50ps jitter thing, but everywhere i look they mention 50ps for those most commonly used spdif recievers like wm8805/dir9001 says intrinsic period jitter of 50 ps RMS, even the cirrus datasheet mentions it.
https://www.cirrus.com/en/mobile/products/pro/detail/P1262.html
sorry if it came across that way, i didn't really mean to spread misinformation.
Well you are correct regarding those older reciever chips - but a new class of chips are available. Also interesting is the new class of electo-optical transformers capable of 192k - as the older where limited to 96k.
In fact DIYinHK just released this http://www.diyinhk.com/shop/audio-kits/83-192k-optical-spdif-receiver.html
Of course their are FPGA and discrete SPDIF receivers available - both coax and optical.
That is what Chord implemented in DAVE - with 192k optical and 384k BNC coax.
Actually Chord has been doing ultra low jitter - ultra high throughput on SPDIF for some time - specs from the Hugo TT DAC - this was using the older FPGA Spartan implementation:
Inputs
- 1x Optical TOSLink 24-bit/192KHz-capable
- 1x BNC coaxial input 32-bit/384kHz -capable
- 1x HD/SD USB B-type input up to 32-bit/384kHz