Jitter

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Not sure this post is in the right section ...
However If I connect a Music Player which streams the songs through my NAS - will this reduce jitter in conjunction with my DAC in opposite to just sending the signal directly from my HDD to the DAC?

Any thougths on this greatly appreciated. thank you.
 
Not sure this post is in the right section ...
However If I connect a Music Player which streams the songs through my NAS - will this reduce jitter in conjunction with my DAC in opposite to just sending the signal directly from my HDD to the DAC?

Any thougths on this greatly appreciated. thank you.

I think we need a bit more information. What DAC? How is it connected, and to what? When you say "sending the signal directly from my HDD to the DAC", are you saying that you have a DAC with a controller that can read a hard disk directly (using USB, I assume)?
 
Okay. My setup is a NAS with UPnP where the music is stored. I have connected a RaspberryPi to my NAS (through ethernet). My BRIK DAC is connected through USB to my Raspberry running MPD.

DAC (USB) --> RASPBERRY --> NAS. My DAC is doing all of the decoding completely bypassing the onboard sound of the raspberry. All the Raspberry does is buffer 50% of the song before playing it.
 
Okay. My setup is a NAS with UPnP where the music is stored. I have connected a RaspberryPi to my NAS (through ethernet). My BRIK DAC is connected through USB to my Raspberry running MPD.

DAC (USB) --> RASPBERRY --> NAS. My DAC is doing all of the decoding completely bypassing the onboard sound of the raspberry. All the Raspberry does is buffer 50% of the song before playing it.

There should not be a significant difference in jitter.
 
Copy that. So it doesn't really matter in terms of jitter if I'm accessing the files through ethernet with flow-control or if I'm accessing the files directly on the hard drive of the nas.

No, doesn't matter. Both the hard drive interface and the ethernet connection are fully buffered and independently clocked, and the only possible influence on jitter is the CPU/IO load they cause, and that is rather theoretical (and the load is pretty similar in any case).
 

TNT

Member
Joined 2003
Paid Member
NAS [Ethernet] --> Raspberry [USB] --> BRIK DAC [AUX] --> AMP

http://webbuilder3.asiannet.com/ftp/2407/BRIK Spec - DA Converter.pdf
6moons audio reviews: Brik Audio DAC and integrated amplifier

I'm using kernel streaming on the Raspberry. The stream's supposedly untouched.

That the stream has bit integrity is paramount. If it isn't I would not call the errors necessarily "jitter". The important in this case is that it is altered i.e. not good.

I can assure you that how the kernel in a NAS is configured has no impact on jitter. Loss of data, alteration of data OK, but not jitter as in bit/word clock timing error at the DA chip. No.

//
 
Rather? Please explain how, theoretically, a NAS could affect word clock jitter in the D/A?

I was leaving some space for the "everything matters" crowd. :)

The NAS itself can't really affect it in any way, but the network traffic causes varying load on the RPi. We can go through theoretical scenarios of how that would have an effect on clock jitter, but for practical purposes it does not matter.
 
Since MPD is buffering the audio before sending it to the DAC how could playback on my DAC actually be affected by network traffic with final destination RPi?

You wouldn't believe some of the more far-fetched theories I have come across. But then again, there are those who claim to hear a difference between different C compilers used to compile the player software. Personally I would describe them using a word for small kitchen appliance for cooking soups and stews that is also the name a village in North Yorkshire. Interestingly the two parts of the word are also slang words for psychoactive drugs... :)
 

TNT

Member
Joined 2003
Paid Member
Ok Ok, so the "nature" of the incoming data to the DAC could have, via power feeding etc., effect on the properties of the Rasberry spdif generation.

Galvanic isolation is good. Toslink very nice if clock re-generation is performed downstream of the opto link.

Toslink/spdif -> PLL -> DAC ... no.
Toslink/spdif -> PLL -> memory buffer -> *new* clock -> DAC ... yes.


//
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.