Jitter SPDIF bit-perfect thoughts ...

0 Members and 1 Guest are viewing this topic. Read 2148 times.

soren b

  • Newbie
  • Posts: 1
Jitter SPDIF bit-perfect thoughts ...
« on: 20 Dec 2009, 03:10 am »
I have done a small test myself in order to reveal how different digital sources was able to produce bit-perfect delivery.
What I did was preparing a wav (approx 7 min of music)  put in a mark in the file; I used the third sample in the right and left channel which I altered to be max in right and min in left, giving me a easily recognizeable mark in the stream. I burned the wav on to an audio cd (regular data CD treated as audio). played the CD on my CD player (Nakamichi CD2) using the SPDIF output which I grab'ed at my PC using a ESI Juli@ soundcard.
The ESI was configured to be clock'ed externally.
Using Audition to grab the audio. I found that in ten test runs I was able to get bit perfect delivery on a third of the test runs. The remainder tests revealed less than a hundrede differences scattered around in the stream.
(after grabbing the music, I found the mark I out in and used that as a fix point in order to cut the right amount of samples. Then I did a mix paste inverse with the original wav. I also used the Wav tool in EAC to the test).
Now, to me it seems as if the SPDIF transport is able to performe quite good  in that case; The Nakamichi used has been imroved using a Lclock and that might improve the situation?

I did the same test just playing the CD from another PC using the PCs CD-ROM and built-in sound card. Using Foobar as the player. The same result as above.
(I have discovered that iTunes (before 7 which I haven't investigated) and WMP is doing funny stuff with the data, which FooBar doesn't)

 I did the same test using the Squeezebox receiver, and got the same result  once again.

 My question is whether my test reveals anything at al,l other than things is  working pretty ok?

 I mean is the problem in the receiving part of the SPDIF signal?
 I don't know whether I am able to have the equipment configured to be  clock'ed externally ... will it make a difference?
Btw if I grab the music using internal clock everything seems to be pretty  much messed up; is that what happens when the SPDIF hits the receiving  equipment. If that is the case a master clock that everything referes to  might be the solution.

Any questions or comments on this is much appreciated.

audioengr

Re: Jitter SPDIF bit-perfect thoughts ...
« Reply #1 on: 20 Dec 2009, 07:49 pm »
I have done a small test myself in order to reveal how different digital sources was able to produce bit-perfect delivery.
What I did was preparing a wav (approx 7 min of music)  put in a mark in the file; I used the third sample in the right and left channel which I altered to be max in right and min in left, giving me a easily recognizeable mark in the stream. I burned the wav on to an audio cd (regular data CD treated as audio). played the CD on my CD player (Nakamichi CD2) using the SPDIF output which I grab'ed at my PC using a ESI Juli@ soundcard.
The ESI was configured to be clock'ed externally.
Using Audition to grab the audio. I found that in ten test runs I was able to get bit perfect delivery on a third of the test runs. The remainder tests revealed less than a hundrede differences scattered around in the stream.
(after grabbing the music, I found the mark I out in and used that as a fix point in order to cut the right amount of samples. Then I did a mix paste inverse with the original wav. I also used the Wav tool in EAC to the test).
Now, to me it seems as if the SPDIF transport is able to performe quite good  in that case; The Nakamichi used has been imroved using a Lclock and that might improve the situation?

I did the same test just playing the CD from another PC using the PCs CD-ROM and built-in sound card. Using Foobar as the player. The same result as above.
(I have discovered that iTunes (before 7 which I haven't investigated) and WMP is doing funny stuff with the data, which FooBar doesn't)

 I did the same test using the Squeezebox receiver, and got the same result  once again.

 My question is whether my test reveals anything at al,l other than things is  working pretty ok?

 I mean is the problem in the receiving part of the SPDIF signal?
 I don't know whether I am able to have the equipment configured to be  clock'ed externally ... will it make a difference?
Btw if I grab the music using internal clock everything seems to be pretty  much messed up; is that what happens when the SPDIF hits the receiving  equipment. If that is the case a master clock that everything referes to  might be the solution.

Any questions or comments on this is much appreciated.

If you are not getting consistent results with S/PDIF, then there are likely signal integrity problems causing errors.  I have seen a lot of poorly designed S/PDIF interfaces in modding lots of gear over the last 10 years.  The cables you use can also make or break it.

Steve N.