XMOS XU208 USB BRIDGES - THE LATEST GEN HAS ARRIVED!
Jan 29, 2018 at 3:08 PM Post #3,796 of 3,865
Do you mean that pairing Matrix with Marantz wouldn't make any sense even with the coaxial rca-connection?

It just seems unnecessary and should make no difference at best. Can you decidedly tell the improvements from using the F-1 with the Marantz vs. the Marantz by itself? The Marantz already has a gavlanically isolated async hi-res capable USB interface and that is the most direct and jitter-free path to the actual D/A conversion stage from a PC.

Good reasons for a DDC: a) The DAC lacks a USB port b) The DAC only supports UCA-1 which maxes out at 96kHz/16bit but can do higher on SPDIF and you want to play hi-res c) The DAC has an inferior USB implementation compared to its SPDIF inputs d) The DAC's USB isn't galvanically isolated causing issues but the DDC is e) The DAC has an I2S input. None of which apply to the Marantz HD-DAC1.

The Marantz has its own reclocker, the value of up stream re-clock is a moot point here.
 
Last edited:
Jan 29, 2018 at 3:12 PM Post #3,797 of 3,865
It just seems unnecessary and should make no difference at best. Can you decidedly tell the improvements from using the F-1 with the Marantz vs. the Marantz by itself? The Marantz already has a gavlanically isolated async hi-res capable USB interface and that is the most direct and jitter-free path to the actual D/A conversion stage.

Good reasons for a DDC: a) The DAC lacks a USB port b) The DAC only supports UCA-1 which maxes out at 96kHz/16bit but can do higher on SPDIF and you want to play hi-res c) The DAC has an inferior USB implementation compared to its SPDIF inputs d) The DAC's USB isn't galvanically isolated causing issues but the DDC is e) The DAC has an I2S input. None of which apply to the Marantz HD-DAC1.

The Marantz has its own reclocker, the value of up stream re-clock is a moot point here.

Thanks for your reply. I will probably stick with the F-1.
 
Last edited:
Jan 29, 2018 at 4:26 PM Post #3,799 of 3,865
Hi Zowie1,
To understand : you have the Marantz HD-DAC1, and you use a Singxer F1 ?
if yes, it sounds better thru the F1 than using the USB input of the DAC ?
if yes, many DDC are told to be better than the Singxer F1 : Singxer SU-1, Shiit Eitr, the Matrix, & some others I don't know, so changing the F1 will lead to better SQ.

Usually, using USB input of DACs brings to lower SQ than using DDC>coax/AES>DAC. That's why DDC are popular.
This Marantz DAC may be the except to this rule. To be honest, I don't know because I never listen to it.
Rgds
 
Jan 29, 2018 at 11:56 PM Post #3,800 of 3,865
My point being, you likely don't even need the F-1, unless of course you can decidedly tell the difference between it and a direct USB connection to the DAC.

Actually using the coaxial connection with the F-1 compared to just regular USB-connection makes quite a difference in SQ with the HD-DAC1. I was just wondering if replacing the F-1 with the Matrix X-SPDIF-2 would make it even better.
 
Jan 30, 2018 at 12:17 AM Post #3,801 of 3,865
Actually using the coaxial connection with the F-1 compared to just regular USB-connection makes quite a difference in SQ with the HD-DAC1. I was just wondering if replacing the F-1 with the Matrix X-SPDIF-2 would make it even better.

Can you describe this difference? Is it THAT much different? Is it better or is it just different? Do you feel you would you be able to distinguish the 2 in a blind test? I just find it hard to believe Marantz engineers couldn't make the USB input right on a fairly expensive unit but sure there have been srew ups like this before. Perhaps there are other factors in play? Like not going through the same digital audio path, or perhaps not sending bit-accurate in one case vs. the other? Do you use ASIO or Wasapi?
 
Feb 3, 2018 at 10:10 AM Post #3,803 of 3,865
Has anyone found a good working driver for the breeze U8 DDC (usb to coax and opt)
 
Feb 3, 2018 at 10:56 AM Post #3,804 of 3,865
Feb 18, 2018 at 9:30 PM Post #3,807 of 3,865
USB data is received by the XMOS chip using asynchronous isochronous transfer mode, then sent out over differential I2S using stable word clock. Is that what you're after ?
Hem, I didn’t understand half of the words you used. Although I’m an engineer signal processing is not my field...
So does the conversion imply a reclocking of the signal and if so, which is the component that does that on the XU208?

EDIT: what I am ultimately trying to understand is the following: suppose I use a very clean, jitter free signal as an input to the XU208 module in my DAC (coming from a device with very good quality oscillators), will this signal be f### up by the good but not top quality oscillators in the XU208?
 
Last edited:
Feb 18, 2018 at 9:47 PM Post #3,808 of 3,865
Hem, I didn’t understand half of the words you used. Although I’m an engineer signal processing is not my field...
So does the conversion imply a reclocking of the signal and if so, which is the component that does that on the XU208?

EDIT: what I am ultimately trying to understand is the following: suppose I use a very clean, jitter free signal as an input to the XU208 module in my DAC (coming from a device with very good quality oscillators), will this signal be f### up by the good but not top quality oscillators in the XU208?

Not sure if the XMOS chip has internal oscillators but these converters or DACs typically have oscillators outside of XMOS, and if your oscillators are better than what's in a particular DDC or DAC then yes the final I2S output can be worse than what you feed it with over USB. That's my understanding anyway. The XMOS chip runs in asynchronous mode and thus is fully decoupled from the upstream clock.
 
Last edited:
Feb 18, 2018 at 9:52 PM Post #3,809 of 3,865
Not sure if the XMOS chip has internal oscillators but these converters or DACs typically have oscillators outside of XMOS, and if your oscillators are better than what's in a particular DDC or DAC then yes the final I2S output can be worse than what you feed it with over USB. That's my understanding anyway. The XMOS chip runs in asynchronous mode and thus is fully decoupled from the upstream clock.
Understood, thanks!
I’ll try to enquire the DAC manufacturer about how the conversion happens.
 
Feb 18, 2018 at 10:01 PM Post #3,810 of 3,865
Understood, thanks!
I’ll try to enquire the DAC manufacturer about how the conversion happens.

I don't think there is much to it conceptually, the data is received from USB, buffered in the XMOS, and then converted to I2S on a clock that's independent from the USB clock. Synchronous USB interfaces went out of fashion a while back due to the typically jittery USB clock.
 

Users who are viewing this thread

Back
Top