[ironmine]

Response from one of eBay sellers:
 

>>Yes, it has high precision mode. I am not sure if the RCA is 75 ohm or not and the manufacturer hasn't announced any jitter measurements yet.

 

[leeperry]

well it doesn't have much point over the 01USD as far as I can see...and you can plug the 01USD directly onto your DAC using this kind of adapter: http://www.uauctionsource.com/picture/SKU004912_0.jpg
 
the 01USD is running in asynchronous mode, the cable shouldn't matter...and last time I checked, a good USB cable was cheaper than a good coax

 
plus the 01USD is using pulse transformers so using this adapter we shouldn't have to worry about interferences between the 01USD and the DAC.

 

[ironmine]

This card is similar to the Musiland card discussed here:
 
TEMPOTEC Fans Digital March
 

 
www.tempotec.com.cn/en/2-Product/23-Fans/DigitalMarch/index.htm

 

[leeperry]

oops, I've edited my post above....and last argument, the 01USD runs in its own faraday cage...we both know that a PC case is an EMI/RFI nightmare

 

[glt]

I'm not aware of any DAC that will display the sample rate to the 1 hz resolution. So other devices may or may not be precise at 44,100 hz. Low jitter and precision are not quite related. The variance from exact 44,000 hz is due to the precision of the crystal which is probably 100 ppm (parts per million).

 

[leeperry]

I'm not aware of any DAC that will display the sample rate to the 1 hz resolution.

 
well, this is your blog, right? http://hifiduino.wordpress.com/2010/04/08/reading-sample-rate/
 
it'd be nice to see what the HiFace feeds to this DAC

 
I've got some 10ppm clocks for my current transport, but I was told in the DIY forum that it'd be pointless to swap them..

 

[ironmine]

Quote:

well it doesn't have much point over the 01USD as far as I can see...and you can plug the 01USD directly onto your DAC using this kind of adapter: http://www.uauctionsource.com/picture/SKU004912_0.jpg
 
the 01USD is running in asynchronous mode, the cable shouldn't matter...and last time I checked, a good USB cable was cheaper than a good coax

 
plus the 01USD is using pulse transformers so using this adapter we shouldn't have to worry about interferences between the 01USD and the DAC.

What about reports that Musiland 01USD has an RCA output which is 90 Ohm instead of standard 75 Ohm? And what about complaints that the output voltage is only 0.2 V instead of standard 0.5-0.6 V?
 
One guy on a Russian forum tested this Musiland box and found it to be not bit-perfect...

 

[ironmine]

Quote:

well it doesn't have much point over the 01USD as far as I can see...and you can plug the 01USD directly onto your DAC using this kind of adapter: http://www.uauctionsource.com/picture/SKU004912_0.jpg
 

I cannot understand you. Why would I need this adapter? I can use a normal digital coax cable, if I ever buy 01USD, can I not?

 

[leeperry]

Well, this adapter avoids using a coax cable...so no interferences and no cable attenuation.
 
do you have a link to that test please? did he also try the BNC output?
 
some companies claim to be selling 75Ω BNC>RCA cables: http://www.stsi.com/assets/images/products/bnr-103.jpg
 
but most ppl agree that a RCA connector cannot be made 75Ω in the first place...

 

[glt]

Quote:

 
well, this is your blog, right? http://hifiduino.wordpress.com/2010/04/08/reading-sample-rate/
 
it'd be nice to see what the HiFace feeds to this DAC

 
I've got some 10ppm clocks for my current transport, but I was told in the DIY forum that it'd be pointless to swap them..

 
Sorry, I wanted to say that I've never seen a DAC show 1hz precision "besides the one I implemented" . That means that you may be getting deviations in other devices but you never know unless you measure it. most of the deviation in the sample rate can be attributed to the crystal in the musiland and some to the clock in the dac (20ppm) and some to roundoff error in using integer calculation. Deviation also depends on the production batch of the crystals since it is +/- ppm.
 
Since I don't have a hiface, I can't measure it
 

 

[leeperry]

so you're saying that the 44105 Hz we see might be a "worst case" scenario and that the Musiland might actually be outputting closer to 44.1 than that?

 

[glingrem]

Sample rate deviation can be no more 3% and it's no problem for connection.
It doesn't correlate with jitter in S/P DIF interconnect..

 

[leeperry]

Sample rate deviation can be no more 3% and it's no problem for connection.
It doesn't correlate with jitter in S/P DIF interconnect..

 
Yes, but it means that we're not hearing our music at the right speed and that everything's slightly out of tune?
 
that's what m2tech are saying at the end of that PDF, where they boast about very low clock inaccuracy: http://www.m2tech.biz/public/pdf/White%20Paper%20on%20hiFace.pdf

 

[glt]

Quote:

so you're saying that the 44105 Hz we see might be a "worst case" scenario and that the Musiland might actually be outputting closer to 44.1 than that?

Don't quite remember my calculations but most is due to the Musiland. So maybe 1 hz is due to the other contributions. But we are splitting hairs at this point.
 
Here is measurement of (another) Musiland I2S lines: http://hifiduino.blogspot.com/2010/02/musiland-driver-1082-fast-vs-precision.html
 
And if your Musiland was built from another batch of the oscillators, it could be better than mine

 

[glingrem]

Quote:

 
Yes, but it means that we're not hearing our music at the right speed and that everything's slightly out of tune?
 

Yes, You right. But can you hear this  speed difference?
I think, a problem was in inaccuracy at frequency synthesis.