|
Home
/ FAQ
/ News Classifieds / Events |
Audio Asylum Thread Printer |
Get a view of an entire thread on one page |
64.201.43.254
| '); } else { document.writeln(''); } } else { document.writeln(''); } } else { document.writeln(''); } } // End --> |
According to this review, their new USB DAC supports asynchronous mode. Is this similar to the way Ayre and Wavelength have implemented their USB inputs?
http://www.soundstagenetwork.com/ultraaudio.com/equipment/arc_dac7.htm
... ARC’s Dave Gordon told me that the DAC7 operates in "asynchronous mode"; i.e., a clock within the DAC controls the transfer rate of the digital signal. In a typical system, the timing of the digital-signal transfer rate is dictated by the computer, in what’s called adaptive mode. A USB connection via a computer is not ideal for this task, partly because of the number of functions a computer’s processor is required to simultaneously perform. When the processor in the computer is overtaxed, the digital stream ostensibly suffers, and the timing of the audio delivery won’t be perfect.
From what Gordon told me, the benefit of an asynchronous-mode DAC is less digital jitter (i.e., fewer timing errors within the digital stream itself), which, on paper, should result in better sound. ARC goes this one better, however: Once the digital data are inside the DAC7, they’re buffered (stored in memory), then re-clocked by an ultra-accurate second clock, before being fed to the DAC itself...
...and he is not a design engineer. So it is possible that he got some of the details wrong. The review of the DAC in Hi-Fi News (UK) said that the ARC DAC uses a TI USB-to-S/PDIF part. Then they feed the output of this through the S/PDIF receiver chip that has a low-pass filter on the PLL to remove a lot of the jitter. This is a nice improvement, but according to what they published is not true asynchronous mode.
Edits: 04/17/09
It looks like Gordon has cleared up the confusion.
"After a long talk with all involved the review has been upgraded to correct for the mistakes regarding the miss use of the term Asynchronous USB with the DAC7."
Unfortunately, rather than post a correction, they actually just removed both paragraphs, so that there is now no technical info.
Thanks for the response! I haven't seen that review, I will try to track down a copy to have a look.
The 'lock' LED certainly suggests there is a PLL being employed.
On the ARC website it says:
JITTER REDUCTION:
High-stability crystal-controlled re-clocking for all outputs
The word output is confusing. That aside, this suggests a VCXO is being employed? Would this be different than employing a 'low-pass filter on the PLL' you mentioned above?
I think the review is online at SoundStage.com.
As far as the specs page goes, clearly there is a misprint. They must have meant "inputs" and not "outputs". A PLL allows one clock to "lock" onto the same frequency as another clock. Most audio PLL's use a VCO (Voltage Controlled Oscillator) as the variable-frequency clock that "locks" to the incoming signal. Better implementations use a VCXO (Voltage Controlled Crystal Oscillator), which is still variable-frequency but uses a crystal for lower jitter.
For more on clocks, please read this post from Tony Lauck:
Thanks for the response, great link on clocking!
Found the review, no mention of the USB details other than "USB input receiver/CODEC from Burr-Brown"
> > Found the review, no mention of the USB details other than "USB input receiver/CODEC from Burr-Brown" < <
Well, that should tell you a lot right there. But that's not the full review. That was only Ken Kessler's text. The technical sidebar was not included in that excerpt on the importer's website.
Normally you can buy reprints online. But that particular one is not up yet.
Charlie,
Actually the part the PCM2902B has both a input ADC/SPDIF which is Asynchronous and an output DAC/SPDIF which is Adaptive.
This is where Dave Gordon got confused and why this started in the first place. In their application (since they do not use the input) the device is truly a 16 bit USB Adaptive device.
Thanks
Gordon
J. Gordon Rankin
I did not think there was another version of asynchronous data retrieval out there.
dCS has a couple of (very expensive) products that do asynchronous USB.
Post a Followup: