Mr. Exa has already described one reason why this is a bad idea. As a USB hardware designer, I can say that there is another very important factor: The circuit traces between the USB connect and USB chips should be designed with controlled impedances, avoiding vias and especially headers / jumpers. It's the sort of thing that amateurs get wrong, and they usually get it to work most of the time, but professional USB products will not allow modification of the D+ and D-. Jumpers on the power would not violate the USB recommendations, but as exa points out, these are completely unnecessary.
Ooops I forgot that Mr. rsdio was moderating the DIY related issues - if I have other questions they may need to be asked below the
....You can do whatever you want. That doesn't stop people from calling a bad idea a bad idea if that's what it is.Ooops I forgot that Mr. rsdio was moderating the DIY related issues - if I have other questions they may need to be asked below the
I merely summarized the recommendations of the people who created USB itself. If you want to go against the USB specifications themselves, hoping to somehow improve the performance by violating the specifications, then be my guest.
I won't try to stop you from implementing a bad idea if you won't try to stop me from calling a spade a spade.
P.S. If someone were moderating, wouldn't your post have simply disappeared? Why accuse me of moderating when I'm clearly doing no such thing?
Does anyone use this USB extender as a standard compliant galvanic isolator?
Icron Ranger 2224 Multimode Fiber Optic USB 2.0 Extender
Icron Ranger 2224 Multimode Fiber Optic USB 2.0 Extender
Hello Bunpei,
Thank you for the link. I will be very curious to have optic USB isolation compared to GMR isolation with respect to impact on sonic performance. The difference in price is obvious. Similar devices are expensive because isolation at the USB side requires high bandwidth capability.
In contrast isolation on the output side is less demanding in therms of throughput and therefore less expensive. The arguments against GMRs voiced on this thread are:
- Jitter caused by GMRs
- GMRs are not fast fast enough
One of the best articles on Jitter impact and evaluation can be found on the website of one of the best DACs - Anedio Affordable High-End Audio : Jitter measurement of USB and SPDIF interfaces
This is a highly recommended reading for audiophiles that are interested to develop independent opinion on the subject of Jitter. Ironically the objective information provided on the Anedio website is called "the best of advertising" by the Charlatans of the audio world.
Hi Leif,
I need at least 10 users interested to preorder the custom version of exaU2I to justify the development of the new design. I will need a small deposit upfront. The custom exaU2I will target PCM1704, however it will preserve the standard I2S compatibility. All users will benefit form the enhancements.
Best regards,
exa065
Last edited:
I rather prefer this approach by audiophilleo though I have not experienced the product.
jitter definition
phase noise definition
I wish we could measure phase noise profiles for three I2S clock signal lines using Symmetricom 5120A phase noise analyzer
5120A Test Set
and compare the profiles and sonic impressions.
We are considering to allow the use of VST plug-ins at the driver level. For now you can use the player plug-in features for DSP processing. I've posted my experience using Allocator with Foobar on the exaDevices Blog - Using Allocator with exaU2I and Foobar 2000
You are right. It will be done but it will take a while to develop a VST host and to integrate it with the driver in a convenient and simple way. I was just pointing out that you can do the exact same think within Foobar - on the VST input you have only 2 channels and the output is 8 channels.
I've just finished testing exaU2I with AlbumPlayer - a fine ASIO player with outstanding user interface and touch-screen support.
AlbumPlayer works flawlessly with exaU2I up to 384 kHz. You can see screenshots and setup instructions here - Using exaU2I USB-I2S 384 kHz/32bit Eight Channel Transceiver with AlbumPlayer.
AlbumPlayer works flawlessly with exaU2I up to 384 kHz. You can see screenshots and setup instructions here - Using exaU2I USB-I2S 384 kHz/32bit Eight Channel Transceiver with AlbumPlayer.
I'm still considering buying the u2i the only thing holding me back is that i want to be able to play movies with MPC-HC and watch TV using DVBViewer both are not ASIO compatible. I would need some software for it to do MME/WDM > ASIO. Does something like this exists?
What about Virtual Audio Cable? does exaU2I support this ?
On the website of VAC they say it is ASIO compatible, did any one test this? It should show up in the audiorepeater software from VAC.
What about Virtual Audio Cable? does exaU2I support this ?
On the website of VAC they say it is ASIO compatible, did any one test this? It should show up in the audiorepeater software from VAC.
Hello Leif,
I did more testing today and the results are excellent. The CPU usage is 2-3% at 384kHz. It may be a bit lower than Foobar. The AlbumPlayer volume control is disabled in ASIO mode and the LSB extensions requested by RayCtech are working as expected. I can confirm bit-perfect operation. To me AlbumPlayer sounds exactly the same as Foobar and J. River Media Center.
I wouldn't expect anything else from any bit-perfect player. This is the beauty of using bit-perfect signal processing chain - player, driver and USB interface. The jitter caused by the PC software and hardware is not a differentiating factor because of the asynchronous operation of exaU2I. And finally, the jitter levels of the exaU2I output are way too-small to be noticed with a jitter-canceling DAC like your Buffalo kit.
I think it is worth it to try AlbumPlayer. I would mention one issue to be careful with: The media library management capabilities are a bit too advanced for me. I wouldn't let it modify my media library in any automatic way.
Best,
exa065
Last edited: