• These commercial threads are for private transactions. diyAudio.com provides these forums for the convenience of our members, but makes no warranty nor assumes any responsibility. We do not vet any members, use of this facility is at your own risk. Customers can post any issues in those threads as long as it is done in a civil manner. All diyAudio rules about conduct apply and will be enforced.

Hiface USB -> I2S 24bit/192Khz

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Jump to Latest
M

Marek

Member
Joined 2003
audiodesign said:
Chess Lim of the XtremeUSB are developping an USB interface called XD1014 EVB.

It will be available for the shipment in two weeks and It will work at 192kHz 24bit.

It has Optical in out for SPDIF up to 96kHz but and it have 192kHz if we wire for coaxial.

It support driver for Windows XP, Vista and 7 for both 32bit 64bit and Mac is for Snow Leopard only.

XD1014 EVB has oscillator to generate the Master Clock

I have asked to have a dip connector to get the I2S signals.

The price is still to define by XtremeUSB .
Click to expand...

To be sure I have a few questions:
- "no jitter anymore" - is it 100% asynchronous transfer?
- are there 2 separate XOs for 44,1, 48kHz and multiplies?
- is there any chances to integrace some ultra-low-jitter XOs?
- are XtremeUSB guys define output jitter of I2S signals? (not MCLK)
- what about power supplies? are they feeded from USB or external clean supply? is it possible to separate supply for each module? (XOs, chips)

Thanks
Marek
 
R

regal

Member
Joined 2004
audiodesign said:
Chess Lim of the XtremeUSB are developping an USB interface called XD1014 EVB.

It will be available for the shipment in two weeks and It will work at 192kHz 24bit.

It has Optical in out for SPDIF up to 96kHz but and it have 192kHz if we wire for coaxial.

It support driver for Windows XP, Vista and 7 for both 32bit 64bit and Mac is for Snow Leopard only.

XD1014 EVB has oscillator to generate the Master Clock

I have asked to have a dip connector to get the I2S signals.

The price is still to define by XtremeUSB .
Click to expand...

Need two oscillators, one 48khz mult + one 44,1khz mult. Otherwise better off hacking up an Hiface IMO.
 
UnixMan

UnixMan

Member
Joined 2005
audiofaidate.org
If we want to get a "definitive", jitter-free solution, then there are a few requirements which can not be avoided.

Asynchronous USB connection is of course required. But that's not enough.

Two separed low-jitter clock oscillators are required. But that's not enough.

Isolation form USB ground, supply & bus noise is IMHO another aspect which is paramount and must not be overlooked.

IME, all jitter reduction efforts are easily swamped if you let noise "get in" in any way.

Noise can come from (or pollute) common grounds, power supply lines, inputs, etc. We must find ways to stop it, or all our effort will be in vain.

Since it looks like isolating hi-speed USB2 is not easily (if at all) feasible, then we must isolate on the I2S side.

Our clean, low-jitter clocks must be placed on the "clean side", and sent back to the USB/I2S interface (which is on the "dirty side") through appropriate isolation (preferably optical: we must "open loops" and reject as much high frequency noise as possibe).

I guess I2S signal lines will then have to be "reclocked" too on the "clean side" after isolation.

That's surely a (little) bit of extra complication and cost, but failing to do so would make the whole effort almost pointless.

So please do not overlook this and try to get a proper design done once and "forever"!
 
O

orpheus

Member
Joined 2004
I posted a thread about this, but probably should have posted it here.

With a USB to SPDIF converter as small as the M2Tech, I think it would be better to reverse the gender of the connectors.

Female USB to Male BNC/RCA would be a better implementation, and eliminate the SPDIF cable and probably some jitter as well. Has anyone tried this? Any reasons not to do it?

Just curious,
Aaron.
 
EUVL

EUVL

Member
Joined 2003
xen-audio.com
This for me sounds like another Hiface.
Many things we do not need -- Toslink, SPDIF, ....
Costs more money, create more power supply noise, ...
Then easier to mod a Hiface.
At least that has been done and fully documented.
I am not sure it is a route I wish to follow.

No news at all from xmos ?


BR,
Patrick
 
UnixMan

UnixMan

Member
Joined 2005
audiofaidate.org
orpheus said:
With a USB to SPDIF converter as small as the M2Tech, I think it would be better to reverse the gender of the connectors.
Click to expand...
we're not really talking about USB to S/PDIF here. We are talking about USB to I2S. One of the main point about that is to avoid the (nasty) S/PDIF altogether.

orpheus said:
Female USB to Male BNC/RCA would be a better implementation, and eliminate the SPDIF cable and probably some jitter as well.
Click to expand...
you do not need anything special to do that. Just use a normal hiFace (get it with the BNC out), an USB2 "extender" type cable (a cable with a flat male on one end and a flat female on the other) between the PC and the hiFace and a "barrel" BNC connector on the s/pdif side. That's it.
 
Last edited:
A

agent.5

Member
Joined 2003
UnixMan said:
Our clean, low-jitter clocks must be placed on the "clean side", and sent back to the USB/I2S interface (which is on the "dirty side") through appropriate isolation (preferably optical: we must "open loops" and reject as much high frequency noise as possibe).

I guess I2S signal lines will then have to be "reclocked" too on the "clean side" after isolation.
Click to expand...


Which DAC are you planning on using? A ES9018 Dac uses a 80mhz clock. Maybe we can design a replacement clock module that feeds 80mhz to the Dac, divides the clock and feeds the 2 clocks that the USB/I2S interface requires.
 
S

sampler

Member
Joined 2010
mvaudiolabs.com/
Just summarizing all info.

The paramount things in product design we need:

- Asynchronous USB (device as clock master)
- Two separate low-jitter VCXO for 44.1k (22.579) and 48k (24.576) sampling derivation.
- Working bit-perfect (ks) drivers for Windows (XP,Vista,7).
- Good documentation/support from manufacturer
- 192khz 24bit I2S output
- Pricing range ~ 50Eu (75$)

Preferable things we need:

- As currently no technical solution for 480 Mbit/s USB2.0 isolation exists (only upcoming isolators from AD?), I2S isolation (optical with high speed cmos optocouplers like HCPL-7723 or digital/inductive as ADuM3440 or better ISO7240 – 1ns eye pattern skew jitter declared)
- Very customized drivers that are managing to stay within 150Mbit/s USB limit, and isolating at USB side (it is doable, at least in theory)
- No onboard PSU, connection for external
- Slaving to external clock (clocks) on “clean” side via isolation (if that is reasonable, as any isolation will introduce it’s own jitter)
- Drivers support for Mac, Linux.
- I2S modes selection

Current candidates:

Musiland 01 MINI
+ Price ~60USD
+ Good manufacture updates and support, HDCD decoding
+ Well documented mods
- Only one clock

XMOS USB Audio 2.0 Reference Design Kit
+ All paramount objectives satisfied
- Price

M2TECH Hiface SPDIF
+ Well documented mods
+ All paramount objectives satisfied
- Price
- Previous experience with manufacturer

XtremeUSB XD1014 EVB
- Only one clock ?

Let’s add if I missed something and start new thread.
 
A

audiodesign

Banned
Joined 2005
Two weeks are too long to wait so I have modified my EMU 0404 USB.

I have changed the power supply electrolytics near the DAC chip and I have got the balanced output on this chip.

All the opamp and all the capacitors on the signal path are bypassed .

The balanced signals with the +2.5Vdc has been connected on a Lundahl transformers like in the my DAc final (DAC Final).

Very simpe this modification, no more than 1 hour.
 
Last edited:
UnixMan

UnixMan

Member
Joined 2005
audiofaidate.org
audiodesign said:
Xmos still no answer, told me is necessary lot of time
Click to expand...
that's unfortunate. Hope they'll answer soon. IMHO to date "Xmos" seems to be the most interesting and viable solution.

BTW, why not trying to setup a collaborative effort to design our own custom board (+ FLOSS firmware and drivers) using that chip?

I believe the problem is so general and common that a great many people could (should) be interested in participating in such a project. And I'm more than sure that here at DIYaudio there are all the required human resources to address competently and effectively all of the hardware, firmware & software aspects of such a design.

That way we could address all the aspects of the problem in the best possible way in order to obtain a "definitive" solution to the digital source problem, leaving out only the freedom to choose whatever DAC & analog stage one may prefer to use.
 
S

sampler

Member
Joined 2010
mvaudiolabs.com/
UnixMan said:
BTW, why not trying to setup a collaborative effort to design our own custom board (+ FLOSS firmware and drivers) using that chip?
Click to expand...

x1 on that one. Basically only proper HW design (with onboard JTAG interface for fw programing) in form of complete PCB is needed, and we have a 100% working product using xmos reference fw and drivers. Software issues could be addressed later. So as I see it, it's not some "year or two" project.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Top Bottom