Monarchy DIP/Scott Nixon Dac vs. USB Scott Nixon Dac?

The only USB devices that implement asynchronous USB audio that I know of are the old emagic EMI2/6 and the Creative Audigi 2NX.

All other devices that I know of do not implement a low jitter version of USB audio.

Cheers

Thomas

I thought the isochronous adaptive mode also allowed low jitter through data flow control?

Yes. In adaptive mode you have two isochronius streams. One from the PC to the sound card to transport the audio samples and one from the sound card to the PC to send the flow control information.

Cheers

Thomas

Scott Nixon USBTD is not asynchronous, but it does have internal USB-to-I2S structure, which is nice. Users report that USB version sounds better than spidf TubeDAC, so why not give a shot?

What does the USB-I2S buy you. You basically get a bad clock on a separate line?

Cheers

Thomas

Has anyone bought (or at least heard) a Scott Nixon USB dac? Could anyone compare a Nixon dac (even not USB) to the Storm D02 or the zhaolu?

I would still like to know whether anyone could comment on the nixon USB dacs. Is the tube version worth the extra money? Would the power suply upgrade be more worthwhile? How does it compare with other popular dacs like storm or zhaolu through optical or coaxial inputs?

That is what I was thinking as well

Using the direct I2S reduces jitter by avoiding the conversion to S/PDIF and the assocated encoding of the clock into the serial data stream. It also usually eliminates a second PLL for recovering the clock from the data stream.

Steve N.

Okay, let's rehash this another time.

The single chip USB audio controllers are all isochronous except for a few products that have other issues. You feed them a clock but internally they buffer the data asynchronously and need to estimate the incoming data rate.

They do that by multiplying and dividing their clock signal under software control watching the fill level of their buffers. The output clock from any of these USB chips is not a low jitter signal no matter how high the quality of the input clock is. A high quality input clock helps but does not provide for a really high quality output.

The output from various chips can be a variety of formats. Most do S/PDIF some do I2S. The amount of jitter in the S/PDIF signal or the seperate clock signal for I2S coming from these chips is the same!

Now let's look at the next step. compared to S/PDIF the receiving DAC can use the sperate clock signal directly with I2S. however, remember where that came from. Unless you add a secondary PLL to that clock signal you simple get what the USB chip generated which is not of very high quality.

For the Highest quality conversion you will not want your DAC to follow either a I2S clock signal or a S/PDIF input. You will want a crystal oscillator driving the converter chips directly and then pace the source to match your speed. This is what asynchronous USB audio and master clock DACS are all about.

A practical solution short of that are the various asynchronous resamplers which translate jitter into broadband noise. Another route are the synchronous reclocking schemes whic are basically a secondary PLL. A typical example of an asynchronous resampler would be a Benchmark DAC1 and a synchronous reclocker would be the Lavry DA10.

I2S is no magic bullet and USB does not help this either. Having said that, I do believe that some of the USB -> digital adapters are working pretty well and you should look into that if you plan to buy a DAC that is not already equipped with a dejitter stage.

Cheers

Thomas

hey thomaspf, i found your post really informative. i have a question, what makes one usb-digital converter better than another? for example, i was looking at the HagUSB from hagerman tech; how would that compare to something like an m-audio transit?

Would the squeezebox be a superior solution for pc audio to a usb interface?