I'm confused - why does setup/hold jitter matter when the data stream is source-synchronous with a clock? (Related question - why NRZ encode data if the clock is available...?)
Using I2S as a fake audio sink might be harder than I first thought - you have to fool ALSA into thinking that it's attached to a "codec". Also the intervening layers of software might try to do dither/scaling which would wreck the bitstream.
Using I2S as a fake audio sink might be harder than I first thought - you have to fool ALSA into thinking that it's attached to a "codec". Also the intervening layers of software might try to do dither/scaling which would wreck the bitstream.
Statistics: Posted by jdb — Mon Apr 28, 2025 6:59 am