[ricmat]

Hi guys!


I have a laptop Asus with onboard Realtek HDefinition Audio soundcard.

It has an SPDIF out, which I'm currently using.

Any good reason not to use it? would I improve my SQ if using USB or Firewire -> DAC?


Cheers!

 

[qwerty009]

I would also like to know cons of your current setup versus external soundcard. from what I've heard cheap parts of your on board card specially clock (quartz) will produce jitter to outgoing signal what isn't good for quality.

 

[zirgated]

This depends on the DAC - if it's using it's own clock, can read the data fine (your clock would have to be *very* wrong for it not to be able to) and always has data in reserve (sufficient buffer), jitter should make *NO* difference.

 

[ricmat]

Quote:

Originally Posted by zirgated /img/forum/go_quote.gif This depends on the DAC - if it's using it's own clock, can read the data fine (your clock would have to be *very* wrong for it not to be able to) and always has data in reserve (sufficient buffer), jitter should make *NO* difference.

well, i guess those are good news so...

 

[tbonner1]

Sound differences between digital outputs can be minimal as long a jitter is not a problem.

 

[radio]

Quote:

Originally Posted by tbonner1 /img/forum/go_quote.gif Sound differences between digital outputs can be minimal as long a jitter is not a problem.

Don't forget the upsampling cards like the audigy 2 series which upsampled downsampled everything to 48khz.

I think best would be to have a "bit perfect" card.

 

[Clutz]

Quote:

Originally Posted by zirgated /img/forum/go_quote.gif This depends on the DAC - if it's using it's own clock, can read the data fine (your clock would have to be *very* wrong for it not to be able to) and always has data in reserve (sufficient buffer), jitter should make *NO* difference.

That's highly debatable. Just because the DAC can get a lock on the incoming SPDIF signal, doesn't mean that there isn't a problem with jitter. It would largely depend on the DAC.

 

[zirgated]

Quote:

Originally Posted by Clutz /img/forum/go_quote.gif That's highly debatable. Just because the DAC can get a lock on the incoming SPDIF signal, doesn't mean that there isn't a problem with jitter. It would largely depend on the DAC.

Quite true. It also doesn't conflict with what I said.

If the DAC is using it's own clock and is using bit-perfect (it should be) buffered data, jitter cannot make a difference; the actual DA part is getting input not affected by external jitter (the DAC may, of course, have jitter itself) - since it's going through a buffer.

 

[Clutz]

Well, a PLL isn't going to get rid of all of the jitter, so jitter still does matter. It'd be better to have an SPDIF output that has less jitter. If there was such a thing as a perfect PLL for jitter rejection, no one would ever bother talking about it anymore.

 

[zirgated]

Quote:

Originally Posted by Clutz /img/forum/go_quote.gif Well, a PLL isn't going to get rid of all of the jitter, so jitter still does matter.

If you're relying soley on a PLL for jitter reduction.

Quote:

If there was such a thing as a perfect PLL for jitter rejection, no one would ever bother talking about it anymore.

Ah! So FLAC really is brighter than PCM after all. Look how many people are talking about it!

 

[Clutz]

Quote:

Originally Posted by zirgated /img/forum/go_quote.gif If you're relying soley on a PLL for jitter reduction.

What else are you going to do to get rid of the jitter? An ASRC?

Bottom line, you're going to get better performance with a source that has less jitter in it's SPDIF output.

Quote:

Ah! So FLAC really is brighter than PCM after all. Look how many people are talking about it!

That's not what I'm saying, and you know that.

 

[ricmat]

so, basically, I'm still "locked" with my SPDIF soundcard output "perfection" concerning it's jitter, right?

I'm considering the Apogee Mini DAC, so my question is - would it be resolved if I used:

USB -> MiniDAC ???

or

SPDIF-> MiniDAC is already a good solution?


Cheers!

 

[Clutz]

It really depends on the details of how the minidac handles USB. If it is going directly from USB to I2S, then you would be better going with USB, because you bypass the whole problem of dealing with SPDIF. Analysis done over at diyhifi suggests that this would result in less jitter than using SPDIF, particularly if the SPDIF receiver is the CS8416. If it is the 8414, the scenario is a bit different, in which case you might be better with the 8414, depending on their implementation of the SPDIF receiver (did they use a zobel network? an isolation transformer?) and depending on the amount of jitter coming from the source in the first place.

My best guess is that you are probably better off going USB direct, but that's just a guess.

 

[ricmat]

Quote:

Originally Posted by Clutz /img/forum/go_quote.gif It really depends on the details of how the minidac handles USB. If it is going directly from USB to I2S, then you would be better going with USB, because you bypass the whole problem of dealing with SPDIF. Analysis done over at diyhifi suggests that this would result in less jitter than using SPDIF, particularly if the SPDIF receiver is the CS8416. If it is the 8414, the scenario is a bit different, in which case you might be better with the 8414, depending on their implementation of the SPDIF receiver (did they use a zobel network? an isolation transformer?) and depending on the amount of jitter coming from the source in the first place. My best guess is that you are probably better off going USB direct, but that's just a guess.

well, I have to say I really dont know that kind of info, although I would like to have.

My point was that Apogee say that USB only promotes sample rates of 44.1/48KHz, while my pc SPDIF permits 96KHz output via SPDIF

(however, I read in MiniDAC's manual that its input via optical SPDIF only permits 48KHz sample rates!)

So, I'm a bit confused...

 

[zirgated]

Quote:

Originally Posted by Clutz /img/forum/go_quote.gif What else are you going to do to get rid of the jitter? An ASRC?

Anything that has sufficient buffering. An ASRC would work, though current implementations seem inflexible.

The algorithim that would immediately come to mind for me would be:

SPDIF reciever:
audio out => FIFO
metadata => DAC circuit

With the DAC circuit pulling data off the FIFO on-demand (in practice, it would likely be clocked to the DAC's internal clock once the DAC chip was configured).