|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
85.19.92.6
In Reply to: RE: " only catching the high frequencies " posted by Bibo01 on June 25, 2015 at 20:13:09
Hi and i think this is a very good question and how usb interfaces should be designed.
I mean, a usb power supply should not be necessary at all with well designed usb dac/converters.
The very first thing that i read about usb was how dirty and weak is the power on the usb port ...
And the usb receiver reclocks the signal already.How many reclockings this blessed usb signal needs ???
It is a little bit funny that usb were presented as the solution of all timing problems afflicting spdif.
Thanks again for the question.
Kind regards,
bg
Edits: 06/25/15 06/25/15 06/25/15Follow Ups:
Congradulations, your comment hits the topic on the mark.
The promotion of usb audio above all else had it's roots in marketing and sales.
There were little justification of the over the top claims in power supply immunity and signal integrity.
I was one of the first people to implement Asynchronous USB in a high end DAC. Wavelength Audio-Gordon Rankin provided the update to the Crimson DAC in 2007. The result for me was improved sound.
I don't remember Gordon ever saying anything about Power supply immunity. As for signal integrety, Aysnchronous USB resulted in less jitter. Above and beyond that, Gordon never made the claims you have suggested. In fact, I don't know of any well recognized manufacturer that has made the claims you have suggested.
But I haven't seen everything out there in terms of promotion, so I am not saying you are wrong Fred. Just as it relates to Gordon Rankin.
When did I mention Gordon?.
I remember you posting how great the TD1543 dac sounded on usb.
I didn't say that you mentioned Gordon. I was just relating my experiences.
Perhaps we should look at what you stated in 2006 Fred; not that I'm really interested.
To help you on your experiences.
Gordon sold async USB as the holy grail in the early days.
You've been his closest follower over here.
You jumped on any bs they put in front of you.
Fact is, many of us were claiming that the async solution is all but perfect.
Even in the early days it was clear that a well tuned PC and power supply
and an isochronous DAC plus isolation sounded better then just a asynchronous DAC alone.
It took Gordon years to (very quietly) admit that there's something else going on.
His body Steve N. chimed in with his famous words "It's all common mode noise".
Bottom line. 10 years later the subject is still not under control.
And in my opinion manufacturers are to blame for these slow developments.
Most of them still build DACs where a simple filter on the USB bus shows improvements.
Hi, well on principle it could very well be.
Maybe it is just tricky to make it rightly.
But the principle sounds very good ... at least the principle.
Kind regards,
bg
Gordon sold async USB as the holy grail in the early days.
I fear you're overstating the case. I don't recall him claiming it was a panacea or that there was little need to pay attention to issues other than the transmission protocol.
You jumped on any bs they put in front of you.
I believe you've made this point before. Trust me, we get it.
Fact is, many of us were claiming that the async solution is [anything] but perfect.
Fact was that products that exploited it were too expensive for most of us so we tried more affordable approaches.
Even in the early days it was clear that a well tuned PC and power supply and an isochronous DAC plus isolation sounded better then just a asynchronous DAC alone.
Not sure how much (if any) was done by way of meaningful comparisons. One camp said one thing, another camp said another while you and fmak said the same thing again and again and again . . . Ah, those were the days.
It took Gordon years to (very quietly) admit that there's something else going on. His body Steve N. chimed in with his famous words "It's all common mode noise".
As I recall, the pair of them used to shout from the rooftops that, asynch or not, you still had to use decent PSUs, cables etc etc. These days, they don't post so much.
. . . in my opinion manufacturers are to blame for these slow developments.
Whereas my opinion is that maybe, just maybe, there's more to the topic than your black-and-white scenario suggests. See e.g. the designer's comments on the device this thread is nominally about.
D
I don't often agree with you..but you are spot on here.
You'll find a consistent thread on noise, power supplies, relocking and format
. . . I don't know of any well recognized manufacturer that has made the claims you have suggested.
Me neither but, hey, why let facts spoil a good argument?
beppe:
a usb power supply should not be necessary at all with well designed usb dac/converters
But a well-designed USB DAC pre-supposes a decent local PSU. Taking power from the bus is fine for non-critical devices but not so fine for others such as modest - let alone "well designed" - scanners, printers and so on. Do you complain about 3.5" USB drives needing their own PSUs?
The very first thing that i read about usb was how dirty and weak is the power on the usb port ...
Exactly. See above.
And the usb receiver reclocks the signal already
The USB protocol wouldn't work without a clock at both ends.
Hi and thanks for the valuable reply.
I start to hate a little the usb connection.
I wonder if ethernet connection could be intrinsically better.
It can transfer a signal up to 100 m i guess with accuracy
while usb cannot do that for more that 10 m ?
... this must mean something.
No doubt about the goodness of asynchronous transmission but more about usb connection.
Thanks again.
Kind regards,
bg
Edits: 06/27/15
Long Ethernet cables cause jitter. Saw a demo of this at CES
Alan
Hi and thanks a lot for the valuable information.
Let's say then that both usb and ethernet connections cause jitter, more or less depending on the length, cable quality, quality of the interface in the pc and receiver.
Then the problem is the quality of the usb receiver device that in the best case should suppress this jitter.
It could be that some designs are better at treating and then reclocking high jitter signals.
I think that the USB Regen is a very smart device and depending on the above mentioned parameters his impact on sound can be very positive.
Let's say from slightly positive from very very positive.
But always positive.
It is a very interesting device indeed.
Thanks again.
Kind regards,
bg
Edits: 06/28/15
Absolute nonsense. A post pulled out of your rear end.
You have such a way with words. Are you saying that Ethernet cables cannot produce jitter? The demonstration at CES was done by Prism Electronics who make professional digital gear for recording studios.
Alan
I am very well aware of who Prism is.But you don't give any information about how the demo was run, what product they were hawking, what their commercial interest was, and any other details.
Just "long ethernet cables cause jitter, I saw it a CES".
Audioquest was also "demonstrating" their CAT7 cables were "better" than the cheapest CAT5 with classic snake oil tactics.
I don't think Prism is in that category btw. But they were there at CES for a reason and there must have been an endgame.
Edits: 07/01/15
They were at CES because they had a prototype dac which they wanted to see if there was any interest in it. The demo consisted of a digital 1000hz signal that they sent to a dac (not there dac) over a 30 meter Ethernet cable. At the output of the dac they filtered out the 1000hz signal leaving a bunch of squealing sounds and tones which were very audible. The gentleman from Prism said that was the sound of cable induced jitter.
He said there can be jitter in the original recording, cable induced jitter and jitter caused by the dac.
Alan
This sounds like a BS marketing pitch. Beware demos. They are easily rigged. There is jitter on Ethernet packets, but this has nothing to do with audio jitter. The coupling of network jitter to audio jitter, depends on the protocols used and the quality of the implementation.
What protocol was used?
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Exactly. Demos in a commercial setting are to be viewed with the high levels of suspicion.
Classic marketing technique..create a problem, then sell the marks the solution.
Thank you for filling in the details. Interesting.
30 meters is exceptionally long and as the rep said, there are numerous causes of jitter, but there are also numerous causes of noise in a run that long.
My question would be how do they did know the sounds that were heard were the sounds of "jitter"? How about power supply noise? How about RFI/EMI?
Yes you are right about all the other factors that could be causing the sounds. Over at prismsound.com they have a video dealing with dacs and measureing the performance of dacs. It is over 1 hour long so I have not had a chance to view it. When I do ,if there is any additional info about jitter I will post again
Alan
Thanks for doing follow up.
I must say I have never, in years of reading CA forums and reading literature, heard of jitter manifesting itself as audible noise. I have always thought the conventional wisdom is that it cause timing errors.
If someone can point to a source that that identifies jitter as audible noise, I would be interested to see it,
20-30 feet?
I need about that to get from the internet router to the SONOS Connect in my main system.
That said, the router has about three city blocks of cheap old phone cable to the fiber connection down the street.
Talk about jitter!
Hi,
> a well-designed USB DAC pre-supposes a decent local PSU. Taking
> power from the bus is fine for non-critical devices but not so
> fine for others
This is true, but that local supply may very well be powered via the USB Bus and deliver extremely low noise. It is merely an issue to decide what you need and to design it into the circuit.
But that is just restating what I wrote earlier in a different way.
Ciao T
At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to untolerable pain. Really, what more could we need?
shove it into the front usb socket of a Mac Mini and the best computer audio can be had with 'audiophile' playing software?
"shove it into the front usb socket of a Mac Mini"
There are no front sockets on a MAC mini. I guess if you push hard enough you can make one...
Very few of us have your education or insight Fred. I have had to learn things the hard way with trial and error.
shove it into the front usb socket of a Mac Mini and the best computer audio can be had with 'audiophile' playing software?
You have to like the "telling you where to shove it" bit (OK, I have to) but, absent source, I can't say more . . .
Hi and thanks a lot for the kind congratulations.
On principle asynchronous transmission is indeed a good thing.
I am thinking for instance to those players that fill a buffer and then read with accuracy from that buffer.
The weak link appears the usb channel.
I bought an usb extender that uses ethernet cables and promises to work fine up to 30 meters, much more than usb i guess.
Maybe an ethernet like port could be more suitable for the task of transmitting the digital signal (i am way out my knowledge here).
Given that ethernet allows for much longer cables.
I am using ethernet cables of 10 meters between router and pc and nas.
Absolutely no issue at all ... never.For instance i am reading here
https://en.wikipedia.org/wiki/Category_6_cable
" When used for 10/100/1000BASE-T, the maximum allowed length of a Cat 6 cable is 100 meters (328 ft). This consists of 90 meters (295 ft) of solid "horizontal" cabling between the patch panel and the wall jack, plus 10 meters (33 ft) of stranded patch cable between each jack and the attached device "
it's a lot indeed. Way beyond usb standard capabilities.
Thanks again.
Kind regards,
bg
Edits: 06/26/15 06/26/15 06/26/15
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: