|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
24.2.96.14
Who out there is looking for a low cost high speed (480Mbps) USB isolator? I am trying to determine market demand for such a product. I could have it ready for market by early next year.
Follow Ups:
I do think there is a market for a 100 to 150$ device, as long as manufacturers don't build such a functionality in.
I'd be a potential customer.
The Adnaco device is the only device I'm aware of that delivers 100% isolation - even on USB3.0. Most other deivces are more or less
filters that reduce certain noise types a couple of dbs.
To me such an isolation device would have to
1. 100% galvanic isolate - data,power and ground
2. Provide the option to inject an external highest quality 5V supply to
feed a DAC
3. Improve the timing jitter. (At least not making it worse)
4. Should come with flexible USB ports. A-> A B-> B. Because some want
to run devices like Dragonfly, others need it downstream on the
stationary DACs. Not to forget. It also needs to be hooked up
on mobile devices with micro ports.
Good luck. Keep us posted.
Cheers
All of the USB isolators that Abe displays in his response are full speed (12Mbps) USB isolators, which are fine for use with slower speed audio devices. Higher speed audio and video output and input devices use high speed USB (480Mbps). Many Audio Asylum inmates have expressed in the past the need for a high speed USB isolator to eliminate a ground loop between their audio device and their PC that is causing noise in their audio. Such a device is currently very expensive, usually requiring a card installed in the PC or an expensive optical extender hub (www.thewelltemperedcomputer.com/HW/USB_Isolator.htm). But the technology is just now becoming available to make a high speed USB isolator that could be priced in the $100-$150 range. This device would look like an oversized USB drive.
For $199, we have the iFi USB Power which has a ground lift function, a usb filter and a high quality regenerated 5V.
For $99 we have the Wyrd USB Decrapifier for which I do not have first hand knowledge but the information behind the product sounds good.
Both have been tested on by Audiostream and I agree with the finding for the USB Popwer.
An upowered fob will not be able to supply a clean and separate 5V. I would therefore not be happy with parting with $100-150 for one.
Except the Adnaco USB fiber device there are no real isolaters I am aware of. If talking about isolation I talk about 100% galvanic isolation of data, power and ground.
The devices you're bringing up here are filters that reduce certain noise a couple of db.
Just one of the factors. Optics have their own signal integrity problems.
Like most things in this world. 100% on one parameter can cause material degradation in others.
Noise is one of the key issues. That's gonna be covered. Much better then any of the filters you mentioned before. You can add Steve Ns filter and others to the list.
You can't compare Toslink and an asynchronous optical USB link.
The majority of professional data networks are optical.
Data integrity shouldn't be an issue.
The key will be the quality of the active transceiver unit.
How good will it convert and reclock the optical stream?
How good will will be its own power supply and its 5V source?
How much distortions will the transceiver inject into stream itself?
Cheers
you so certain about all this? Have you made measurements and conducted listening tests with high quality dacs and audio chain?
I've been running optical USB (1.0) for years. The receiving end was powered by a TeddyReg. I considered this a great solution. Unfortunately USB 1.0 wasn't sufficiant anymore.
iFi states 5db noise reduction with the iFi purifier!
You just need to read the specs.
They (those passive filters) get the noise that low that the bits are no longer that much affected. (That's what the marketing says). That doesn't mean though that the remaining noise -- whatever noise and other distortions or interferences we're talking about - won't creap into the DAC to affect other critical parts of it through the backdoor.
Obviously there are other effects with USB. Such as crosstalk between the lines. That's why you'll even find highpriced split USB cables to buy nowadays.
Or People cut the power line. Shorten the cable....
These problems are also gone with optical.
Cheers
A filter that colours sounds in high quality systems.
You're talking about iFi products. Same technology inside.
High quality devices (should) come with sophistcated filtering built-in.
If any of these devices respond to any of those tweaks in a positive way, I wouldn't consider them "sophisticated".
In case you own such a "sophistcated" device and you then apply "double" filtering, by e.g. introducing an USB filter, you can easily make things worse.
You might run into overdampening. You can't blame iFi for that.
Unfortunately you never know what a manufacturer has done about filtering.
That's IMO one reason why, for some those tweaks will work, for others there's no difference, for you it makes things worse. Fair enough.
With optical you wouldn't run into the overdampening trap.
Easy test. It worked on my Fireface UCX. Just add a small resistance to your the USB ground lead. And then you increase it.
You'll hear the difference.
There's no black or white.
Enjoy.
Absolutely different
There are plenty of theoretical reasons why Toslink (yes, yes, I know, it's not the same) should be superior to SPDIF. There are (were) plenty of people insisting that it's indeed better, citing galvanic isolation.
None of that changes the fact that it sounds thoroughly inferior - whether by design, or most likely by implementation. In case with optical USB, just citing reasons why it should be better is just as meaningless - and non-experiential - as with Toslink.
The problem with optical links lie in the imperfect transducer, due to signal conversion and refractive index and light scatter issues.
Toshlink just sounds inferior. I have only ever found one cable that seems to produce the goods. The touted glass ones are only just ok.
We need more transports, not computers. One can certainly make computers/USB sound pretty good, I am interested in more.When the big 3 start doing USB, I'll consider it. The answer to your question is not interested.
"Asylums with doors open wide,
Where people had paid to see inside,
For entertainment they watch his body twist
Behind his eyes he says, 'I still exist.'"
Edits: 09/07/14
Exactly correct. USB is not SOTA, despite what several small DAC manufactures were smugly claiming five years ago.
I will say this however, USB is worth looking at from small, purpose built, stripped down servers like the SOtM, the Auralic Aries, and a few others. The Aries gives you multiple digital outs, at a higher cost.
Here is very interesting news. Bryston has taken the BDP series and created the a model with just a USB out.
The reason USB, to me, sounds good on these devices is, to simplify, they are built for audio, and don't roll off an assembly line at Foxconn, and don't run bloated commercial mass market O/S's, etc.
Thanks for this info, and the info on the Bryston.
Once alternative transports start to be produced & compared to computer transports, I am hoping that folks will start to question certain assumptions: which will lead to more non-commercial computers as transports.
Cheers,
"Asylums with doors open wide,
Where people had paid to see inside,
For entertainment they watch his body twist
Behind his eyes he says, 'I still exist.'"
USB isolators already exist and there are at least hundreds to choose from. However, if you put yours in pretty enclosure and price it high enough to be 'audiophile approved', you'll probably get some takers. But there are some expensive 'audiophile approved' USB isolators already on the market. See link below the photos.
USB Isolators already exist. In what way would yours be different or better?
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Or what are they supposed to do? I don't have one and it sounds fine to me.
Cut-Throat
I hate that phrase, "Sound's fine to me."
When I was in college, I worked in the radio station as an engineer. One of the things we did was to listen to the signal that was going out over the air and detect when equipment was deteriorating. As most of the music came from LPs, one of the big problems was deteriorating or damaged cartridges. When we "wrote up" a bad cartridge, a "techie" showed up. He was supposed to replace or repair the defective turntable. There were a couple of techies we hated, because they could never hear the distortion.
"Hear that distortion?"
"Sounds fine to me."
On more than one occasion, we took care of the problem by physically breaking the disputed component so there could be no further question.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
nt
Cut-Throat
What it does is to provide electrical isolation. Among other things it will break ground loops, something that won't be of much significance if you don't have one. Another thing it will do is replace one source of power with a different source of power (hopefully cleaner power). It will also provide some common mode noise rejection and possibly some signal noise rejection. I don't believe any of these devices retime the bits in a USB frame and they certainly don't retime the start of the 1 msec ischronous USB frame rate.
I would expect that if USB noise were causing sonic problems then these devices might help. They would likely reduce the difference between clean USB ports on clean computer systems vs. noisy USB ports on noisy computer systems. I believe these may also be connected in a chain, in which case there may be additional levels of isolation possible. If a DAC came with good USB isolation then the effect of these devices might be minimal. They might also be minimal with a cheap DAC that has horrible sound, with or without the benefit of a clean USB signal. YMMV.
Because they don't affect the rate at which the USB frames are transmitted, these devices will not remove timing effects, i.e. if the computer sends frames at a different rate then they will arrive at the DAC at a different rate. If the DAC uses an async protocol then does not directly affect the timing of the DAC master clock. However, the processing of USB frames is fairly complex and all of the logic involved will create noise inside the DAC and the timing of this noise may affect the DAC clock and analog circuits if the DAC doesn't have its own isolation from the "evil computer like USB receiver" and the critical DAC circuitry.
OF course, one of these devices may make the sound worse, even while reducing the dependance of the upstream computer.
My personal belief is that these devices are a bandaid. The DAC should be built with a very low power USB receiver on a separate supply and then provide an I2S interface, clocked from the DAC to the receiver. This I2S interface is where the real isolation should be. The advantage of this is to keep all the noisy USB electronics away from the DAC proper. For this approach to work there has to be careful consideration of how power supplies and power wiring, ground planes, etc., are handled as well as how the clock architecture works.
With USB, unlike SPDIF, the signal path itself is potentially all digital, with no clock coming from the source. Unfortunately, with USB the signal path is much more complex than USB and this creates a lot of extra logic and complexity, all of which provide paths around any isolation coming from the signal path itself.
There are several DAC vendors who claim to have made DACs that are transparent to transports and some of them have customers (shills?) supporting their claims. Personally, I am skeptical. None have come up with any measurements to show how their products have more isolation than competitors and we know that bits aren't just bits. They have been supposed to have been just bits since the early 90's when two box transports first hit the market, but it has been known since these two box transports first hit the market they this was not the case. (As with my two box Proceed CD player from this vintage, where it was clearly evident that the coax interconnect sounded better than the ST fiber interconnect.)
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
was typically irrelevant. For info, goto the Audiostream website and look for the iFi USBPower and Wyrd USB Decrapifier reviews. Also in the Computeraudiophile website.
A good 'isolator' with clean 5V Power improves SQ by a large margin. Passive transformer coupled devices, however, are mainly aimed at hospital/industrial situations for spike and other interference suppression from typically noisy computer usb connections.
Here is an article I wrote that discusses some of these products.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: