http://www.lessloss.com/computer_audio_usb.html
Aug 27, 2006 at 7:49 PM Post #77 of 85
Quote:

Originally Posted by audioengr
You know quite a bit about hardware for a software kind of guy. I've been doing digital design for about 30 years.


The software that I do includes drivers, clocks, protocols and measuring the outcome.

Quote:

This is why my I2S implementation is so good sounding. The BCLK has very low jitter. S/PDIF implementations are not as good. You can tell the difference after about 10 seconds of listening to music. It's not subtle. Not that my S/PDIF implementations are not good, they also have very low jitter. The I2S is just better.


Help me understand that. The I2S output from the USB audio chips are not any better than their S/PDIF outputs which is not very good. Unless you use a secondary PLL to remove the jitter from the clock line I can't quite see how that would work.

Cheers

Thomas
 
Aug 27, 2006 at 8:18 PM Post #78 of 85
I'm curious too... In the pcm2706, BCK (and WCLK) is derived from the (not so good) recovered MCLK. Just as in a typical CS841* spdif receiver. Is it different for the TAS1020 or TUSB3200 ?
 
Aug 27, 2006 at 8:48 PM Post #80 of 85
Quote:

Originally Posted by thomaspf
The software that I do includes drivers, clocks, protocols and measuring the outcome.



Help me understand that. The I2S output from the USB audio chips are not any better than their S/PDIF outputs which is not very good. Unless you use a secondary PLL to remove the jitter from the clock line I can't quite see how that would work.



Well, for one thing I dont use a USB audio chip that does USB to S/PDIF. I use the TAS1020A which does USB to I2S. The PLL on this is extremely good and there is significant "slop" in the way that it resolves the input and output frequencies. It seems to decouple very well from the USB clock. If the board design and loop filter is done well, the jitter is infinitesimal.

Not that it cannot be improved. I have another design in process that works in conjunction with it to totally eliminate any jitter from the PLL. This will be in my next generation of products. It is called the "Pace Car".

Steve N.
 
Aug 27, 2006 at 9:54 PM Post #81 of 85
I uderstand that the TAS1020 outputs an I2S signal I am just surprised that you think it is high quality.

What is the jitter on that pin and what is the jitter when you actually drive a bclk input with that over a cable to one of the rare DACs that accept it?

I am interested in your new DAC if it is stays reasonably priced.

Cheers

Thomas
 
Aug 28, 2006 at 12:25 AM Post #82 of 85
Quote:

Originally Posted by thomaspf
I uderstand that the TAS1020 outputs an I2S signal I am just surprised that you think it is high quality.

What is the jitter on that pin and what is the jitter when you actually drive a bclk input with that over a cable to one of the rare DACs that accept it?

I am interested in your new DAC if it is stays reasonably priced.

Cheers

Thomas


I dont have anything resolving enough to measure it accurately. I am looking at 2-4GHz sampling scopes to replace what I have. These should have the resolution. The noise on my scope is larger than the jitter.

I have two DAC's in process. The first will go head-to-head with the Cosecant. A USB TubeDAC, probably in the $3999.00 range. The first proto is in process. Still need to do front panel machining and logo etc.. Sounds awesome. Uses Siemens cca tubes from the 60's. Called the "Spoiler".

The second is my long-term project, a SS USB DAC with active preamp, balanced and SE outputs, remote controlled and battery powered. Inputs are I2S, USB, S/PDIF and AES/EBU. All DC-coupled analog. This one will have the Pace Car inside. I made I2S external so it can be used with a modified Northstar 192 transport. This one is called "Formula One". Probably more than $5K.

If that is all too expensive for you, then you should consider my $2650 solution. The Off-Ramp I2S coupled to a Benchmark DAC-1 with minimum mods and an I2S interface. It's a giant-killer at this price IMO. Nothing I have heard touches it.

Steve N.
 
Aug 28, 2006 at 3:09 AM Post #83 of 85
Quote:

Originally Posted by audioengr
I dont have anything resolving enough to measure it accurately. I am looking at 2-4GHz sampling scopes to replace what I have. These should have the resolution. The noise on my scope is larger than the jitter.

I have two DAC's in process. The first will go head-to-head with the Cosecant. A USB TubeDAC, probably in the $3999.00 range. The first proto is in process. Still need to do front panel machining and logo etc.. Sounds awesome. Uses Siemens cca tubes from the 60's. Called the "Spoiler".

The second is my long-term project, a SS USB DAC with active preamp, balanced and SE outputs, remote controlled and battery powered. Inputs are I2S, USB, S/PDIF and AES/EBU. All DC-coupled analog. This one will have the Pace Car inside. I made I2S external so it can be used with a modified Northstar 192 transport. This one is called "Formula One". Probably more than $5K.

If that is all too expensive for you, then you should consider my $2650 solution. The Off-Ramp I2S coupled to a Benchmark DAC-1 with minimum mods and an I2S interface. It's a giant-killer at this price IMO. Nothing I have heard touches it.

Steve N.


YOur own SS design sounds like a wavelength crimson match. In headphone amps we have kevin gilmore in the SS camp and Mikhail Rotenberg in the Tube camp. Maybe in dac we will have you and gordon rankin as a parallel.
 
Aug 28, 2006 at 3:50 AM Post #84 of 85
Quote:

Originally Posted by Konig
YOur own SS design sounds like a wavelength crimson match. In headphone amps we have kevin gilmore in the SS camp and Mikhail Rotenberg in the Tube camp. Maybe in dac we will have you and gordon rankin as a parallel.


I'm in this to win, not come in a dead-heat
evil_smiley.gif


Steve N.
 
You must log in or register to reply here.

Users who are viewing this thread

Total: 2 (members: 0, guests: 2)
Back
Top