[terente0081]

My new system is currently being built, so meantime, I am listening my music and movies and play games on the onboard sound of a 790i Ultra motherboard.
I link it to the Yamaha RX-V861 receiver via coaxial SPDIF.

My question is: How does SPDIF work? Is it's quality very dependant of the sound card used? Or a good discrete sound card is good because it has good DAC's that only show advantage in Analog connections?
Is there a big difference in sound quality between the onboard sound and a dedicated PCI Asus Xonar D2/PM card which I would also link to my receiver via SPDIF?

I know the Analog connection will show some big improvement, but will it be the same for SPDIF?

 

[1UP]

You can read about the standard on wikipedia. In practice, the box transmitting it IS the source for the DAC. Lots of variables (including the DAC's receiver implementation) affect the final sound quality you get.

SPDIF is generally disliked because it embeds the clock signal necessary for conversion/playback into the data stream, meaning clock recovery can be done suboptimally in the time domain, causing jitter.

Hence the popularity for certain USB devices recently.

You can't really make any assumptions about SPDIF being better on mobo vs sound card, whether internal or not. It always comes down to the particular implementation.

Except perhaps that I wouldn't feel it's going too far to say most mobo solutions will be crud.

Maybe someone with a Xonar can share their comments.

 

[justin2net]

Depends...I have tried outputting from various cards, including the Prelude, Audiophile 192, X-Fi Platinum, and some Realtek onboards. The difference is relatively minor, even up to 24/96 and 24/192.
Though my DAC does "dual reclock".

 

[MCC]

I've noticed a difference between my Chaintech AV-710 and onboard Realtek ALC885 for optical S/PDIF to my Stello. The $25 Chaintech card sounds fuller which leads me to believe that the Realtek chip might be resampling the signal somewhere despite my settings.

Your board has a similar Realtek ALC889A chip so I'd watch out.

Edit: I thought I'd mention how I came to this conclusion. I played the same song simultaneously in two separately installed but equally configured instances of foobar, one for each card. I used two monoprice toslink cables connected to my Stello and switched directly between both inputs. I went into the test expecting them to sound the same but that obviously wasn't the case.

I tested in the default output mode (not WASAPI) and switched the cables with the same result.