Hi,
M2Tech actually use a pair of XO's. Musland uses a single PLL from 24MHz.
The funny thing, despite doing things "the right way", it does not really net them lower jitter than the Musiland, at the output of the DAC it drives.
As so often, it seems it is less what you use, but the how that makes a lot of difference.
Now properly modified the M2Tech unit does show up better, but so does a modded Musiland.
Ciao T
M2Tech actually use a pair of XO's. Musland uses a single PLL from 24MHz.
The funny thing, despite doing things "the right way", it does not really net them lower jitter than the Musiland, at the output of the DAC it drives.
As so often, it seems it is less what you use, but the how that makes a lot of difference.
Now properly modified the M2Tech unit does show up better, but so does a modded Musiland.
Ciao T
Using...
a pair of fixed oscillators is the way to go. As noted, the whole thing needs to be properly implemented for best performance though. The Hiface is limited by the power supplies, so mods to them lower the problems. Clean power supplies to the XO are paramount. This should be done the right way, without PLLs that just introduce additional jitter.
I still think it is possible to power this from USB power, given that great care is taken in cleaning it up, through filtering and regulation onboard. The phase noise numbers from the Audiophilleo unit suggest that USB power can be cleaned up quite well, and provide very high performance. The success (and jitter noise spectrum) of the Ayre QB-9 also supports this approach. Using USB power (with the choice to break out power from the board for paranoids) would make implementation into existing DACs much easier. I see no need for providing external clock sync, as long as onboard clocks are good, but perhaps there is no drawback in making this an option (trace impedance issues perhaps, RF guys?) External clocking seems superfluous if one gets the onboard clocks right.
a pair of fixed oscillators is the way to go. As noted, the whole thing needs to be properly implemented for best performance though. The Hiface is limited by the power supplies, so mods to them lower the problems. Clean power supplies to the XO are paramount. This should be done the right way, without PLLs that just introduce additional jitter.
I still think it is possible to power this from USB power, given that great care is taken in cleaning it up, through filtering and regulation onboard. The phase noise numbers from the Audiophilleo unit suggest that USB power can be cleaned up quite well, and provide very high performance. The success (and jitter noise spectrum) of the Ayre QB-9 also supports this approach. Using USB power (with the choice to break out power from the board for paranoids) would make implementation into existing DACs much easier. I see no need for providing external clock sync, as long as onboard clocks are good, but perhaps there is no drawback in making this an option (trace impedance issues perhaps, RF guys?) External clocking seems superfluous if one gets the onboard clocks right.
I think it is more to do with the costs. Material costs are minimal, but testing, manufacturing, marketing, sales, and everything that support them are very expensive, all for a product that will probably not drive many new sales. this is not to mention obsoleting the product they just spent a bunch of money on and have a large stock of.
Final performance of system will always depend on the whole design. In terms of jitter, both interfaces performs acceptable when heavily moded. But to take it another step further one has to start worry about things like DAC's driving clock phase noise, residual jitter distribution nature and general EMI/RF garbage. From my point of view, this is only achievable by introducing some galvanic isolation from pc data downstream and reclocking with local superclock. Nothing new, basically it's the same and proven direction as CD transport guys are going for some time now. Without slavable pc interface, I don't see this happening.
Yeah, read your comments on audioasylium about musiland, also was watching jkeny's work on m2tech.
Well, I perfectly understand this view from the other side of the fence, but c'mon, monitor 02 have been in market for year or so. I don't believe it would be such a financial tragedy by introducing new product. And for driving new sales, there is always marketing department
But what I'm talking, they know better...Now I must take the opportunity and ask about your x-mos based interface. Is it still under development ? Any preliminary release date ?
From: "Gerhard Sattel" <gerhard.sattel@rigisystems.net>
To: <webmaster@audiodesignguide.com>
Cc: "Marcel Dasen" <marcel.dasen@rigisystems.net>
Sent: Thursday, September 23, 2010 12:21 PM
Subject: RE: Rigi Kontakt
> Dear Mrs. Ciuffoli
>
> Our USBPAL evaluation/reference board incl. evaluation SW is available on
> stock, however we have to charge US$ 330.- plus shipping for it. Let me
> know if you want to order, you can pay also by PAYPAL.
>
> Best regards
>
> Gerhard Sattel
>
> RigiSystems AG
> Gerhard Sattel
>
> Managing Director
Incredible 330$ !
I think that my solution with EMU0404 USB is the best
Hi-Resolution System 192KHz 24bit with EMU 0404 USB
To: <webmaster@audiodesignguide.com>
Cc: "Marcel Dasen" <marcel.dasen@rigisystems.net>
Sent: Thursday, September 23, 2010 12:21 PM
Subject: RE: Rigi Kontakt
> Dear Mrs. Ciuffoli
>
> Our USBPAL evaluation/reference board incl. evaluation SW is available on
> stock, however we have to charge US$ 330.- plus shipping for it. Let me
> know if you want to order, you can pay also by PAYPAL.
>
> Best regards
>
> Gerhard Sattel
>
> RigiSystems AG
> Gerhard Sattel
>
> Managing Director
Incredible 330$ !
I think that my solution with EMU0404 USB is the best
Hi-Resolution System 192KHz 24bit with EMU 0404 USB
XMOS
Hello!
I'm wondering why nobody tried out an XMOS USB 2.0 audio reference design?
With the available schematics/pcb files it shouldn't be too hard to modify the reference design board to:
- have direct I2S output (remove the used DAC IC)
- connectors for external clocks (remove the ones on the board)
- connector for external 5V (do not feed the board from USB power)
As far as price is concerned I believe it wouldn't cost that much, being a two-layered board only.
Unfortunately I'm not accustomed with the used Mentor Graphics PADS cad tool.
Nobody tried this or did the above mods?
Hello!
I'm wondering why nobody tried out an XMOS USB 2.0 audio reference design?
With the available schematics/pcb files it shouldn't be too hard to modify the reference design board to:
- have direct I2S output (remove the used DAC IC)
- connectors for external clocks (remove the ones on the board)
- connector for external 5V (do not feed the board from USB power)
As far as price is concerned I believe it wouldn't cost that much, being a two-layered board only.
Unfortunately I'm not accustomed with the used Mentor Graphics PADS cad tool.
Nobody tried this or did the above mods?
Just recieved an AD newsletter presenting the following chip:
ADUM3160 | Full/Low Speed USB Digital Isolator | Digital Isolators | Interface | Analog Devices
Comments?
ADUM3160 | Full/Low Speed USB Digital Isolator | Digital Isolators | Interface | Analog Devices
Comments?
They say:
"Low and full speed data rate: 1.5 Mbps and 12 Mbps"
Maybe also 96KHz/24bit is feasible with, for example, board using Tenor TE7022L that it's capable of this resolution using USB Audio Class 1.0 on USB 2.0 full speed.
Ciao
Andrea
Hey, look at this Ethernet Streaming device:
I know it's not USB but this will solve the PC insulation problem, right?
An externally hosted image should be here but it was not working when we last tested it.
I know it's not USB but this will solve the PC insulation problem, right?
Interesting but three unanswered questions:
1. Does the computer see it as a soundcard or are you stuck with a special audio player
2. Does it accept two local clocks for 44.1 & 48 multiples
3. What is the bit depth and sampling limitation.
I do like the idea of ethernet over USB, but only when the device presents itself as a soundcard, none of the commercial units made today do that, so you are stuck with their limited playback software.
Hei,
some good news on this front. Take a look at the new thread I've started:
http://www.diyaudio.com/forums/pc-b...khz-usb-sdr-widget-collaborative-project.html
some good news on this front. Take a look at the new thread I've started:
http://www.diyaudio.com/forums/pc-b...khz-usb-sdr-widget-collaborative-project.html
Yes, that’s definitely great news, at least for Linux/Mac users. What’s makes me more confident about success of this project, is the fact that people with enough RF background (SDR is not what you would call newbie friendly) is behind the wheel.
In other words, the performance of SDR rx/tx is proportional to the dynamic range of AD/DA converters used (and their implementation). So in terms of acceptable noise floor, phase noise etc. they are on same path with diyaudio community, just on different (higher) level.
Now the bad news. I guess the reason why everyone is not jumping from happiness and posting here in excitement is still the same issue with M$ UAC2 support.
Not long ago I had some discussion with experienced programmer. He has done couple large projects of real time data acquisition systems with feedback in windows environment by means of USB 2.0, so he had good first hand experience of what it takes to develop specific software for AB communication of custom USB 2.0 device using ASYNC out. The words “massive teamwork” and “thousands line of custom code” were his words, describing the workload. So with all do respect to the radio guys working on this SDR project, I tend to be skeptical about their ability/willingness to develop their own UAC2 support in windows from scratch, when linux/osx is already working and it’s suits their needs.
So until no open source UAC2.0 drivers are showing up or M$ decides to support it, this is limited to linux/osx. Still, if the price of first beta SDR-Widget-Audio will be competitive and there will be option to order just interface card (1 of 2 in stack), it may end up good performer, even considering limitation to linux/osx. So it’s definitely worth keeping eye on it.
Btw, Happy New Year to all of you guys
Hello Sampler,
Good observations, indeed regarding this project. I somehow feel however that this might be the most economic stereo solution available. That if the UAC2 problems are being overcomed.
However I can't stop remarking that there are a number of products that seemed to program a driver for windows. Hiface comes to mind, maybe Musiland (not sure, I remeber that is isochronous) and probably Rigisystems
By the way, the more I look at Rigisystems board, the more I liket it..
Good observations, indeed regarding this project. I somehow feel however that this might be the most economic stereo solution available. That if the UAC2 problems are being overcomed.
However I can't stop remarking that there are a number of products that seemed to program a driver for windows. Hiface comes to mind, maybe Musiland (not sure, I remeber that is isochronous) and probably Rigisystems
By the way, the more I look at Rigisystems board, the more I liket it..
Hi guys and HNY,
There is a group buy that might interests you.
Although it is targeted to another group (radio amateurs, Software Defined Radio) it seems to me that you could benefit too.
The product is a USB to multichannel I2S board, it has a USB2.0 highspeed port on one side and can attach two codecs on the other side via I2S.
It provides concurrently (4+4) 4 mono input channels (2 stereo) and 4 mono output (2 stereo) channels streaming over USB, having all channels up to 192KHz sampling rate with 32 bit.
A necessity is the ASIO driver thats gets implemented right now, yet although not ready but more likely will be ready by the time the boards will arrive from the factory.
So, if you feel that this interests you, please pm me.
Best regards,
Christos
There is a group buy that might interests you.
Although it is targeted to another group (radio amateurs, Software Defined Radio) it seems to me that you could benefit too.
The product is a USB to multichannel I2S board, it has a USB2.0 highspeed port on one side and can attach two codecs on the other side via I2S.
It provides concurrently (4+4) 4 mono input channels (2 stereo) and 4 mono output (2 stereo) channels streaming over USB, having all channels up to 192KHz sampling rate with 32 bit.
A necessity is the ASIO driver thats gets implemented right now, yet although not ready but more likely will be ready by the time the boards will arrive from the factory.
So, if you feel that this interests you, please pm me.
Best regards,
Christos
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.