|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
76.220.25.80
In Reply to: RE: " USB REGEN " by UpTone Audio ... any opinions ? posted by beppe61 on May 23, 2015 at 06:18:09
"Maybe one day i will discover that ethernet is much better than usb for this kind of trasmission."
Or WiFi!
Follow Ups:
My post listed four technologies. Of the four, only I2S was originally designed for streaming audio.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
USB is engineered for lowest possible cost over short distances with control from a central point. Ethernet is engineered for building wide distances and permits distributed control. Wi-fi is engineered for portability, not performance or reliability. I2S is engineered for communication within a box, especially on a circuit board.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Which one was designed for streaming audio again?
Sorry, I keep forgetting.
My post listed four technologies. Of the four, only I2S was originally designed for streaming audio.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
From your original post: "I2S is engineered for communication within a box, especially on a circuit board"
How is "on a circuit board" helpful for streaming audio? Or do you suggest that audio only be 'streamed' on a circuit board?
The audio streams from place A to place B. This can be from one part of a chip to another, from one part of a circuit board to another, from one circuit board in a box to another, from one box in a room to another, from one room in a building to another, from one campus to another, etc...
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
USB is engineered for lowest possible cost over short distances with control from a central point.
Perhaps more accurate to say USB is typically engineered for lowest possible cost . . .
My post listed four technologies. Of the four, only I2S was originally designed for streaming audio.
Well, yes and no. IIRC, USB had, for technically sound reasons, a separate Audio class, if not from the off, then very soon after.
Some argue that I2S was never really designed, just cobbled together as a temporary fix but that it sort of stuck around.
"Some argue that I2S was never really designed, just cobbled together as a temporary fix but that it sort of stuck around."
And was (at least initially) only aimed at on-board chip-chip communication. i.e. no ESD protection required.
R.
Hi and thanks for the very interesting reply
But i do not understand how i could connect the dac to the pc in wifi really.
For me to know that usb transmission is not reliable is a puzzling discovery.
Is this related to the length of the connection ? the quality of the cable ?
For now the best result i have achieved sonically is
Pc > usb to aes converter > dac
and the sound is decent.
But i wonder if a chain like Pc > usb dac could give better sound with more simplicity
Kind regards,
bg
Hi,
> But i do not understand how i could connect the dac to the pc
> in wifi really.
WiFi is just another serialised transport.
Like USB, IEE1394 and Ethernet it packs data into a serial stream. Even SPDIF/AES-EBU, AES-MADI and AES-67/Ravenna are ultimately the same thing.
Some systems are bidirectional on a single link, other are unidirectional. Some include flow control other do not.
But at the bottom we send 16 - 32 Bit Data serialised via some definition and and then decoded.
> For me to know that usb transmission is not reliable is a
> puzzling discovery.
Why should anything be perfect?
> Is this related to the length of the connection ?
> the quality of the cable ?
Both and many other factors too.
Any signal can be degraded sufficiently to cause problems.
This goes into some detail:
http://electronicdesign.com/boards/480-mbitss-signal-integrity-becomes-issue-usb-20-designs
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?
Hi and thank you very much indeed Mr. Thorsten for your patience and your very valuable support.However i wonder if a " PCIe to usb adapter card" could give a better usb signal out of the pc ...
i am buying one but not audiophile approved just to try it out.
I see someone recommending the ones with a NEC chip on board.
Maybe it will provide a cleaner/stronger signal on usb ?Thanks a lot again.
Kind regards,
bg
Edits: 05/24/15 05/24/15 05/24/15
Hi,
> I see someone recommending the ones with a NEC chip on board.
Beware, these have some compatibility issues with USB 2.0 devices, especially audio ones.
This chipset is used by an "audiophile" USB Card designed for PCIe add in.
I would suggest to avoid this chip-set.
Anyway, USB Standard compliant Chip-set's should deliver similar drive levels etc., so any material differences are unlikely in the direct USB physical interface. Again, implementation (especially PCB Layout) is crucial here.
The many stories of certain USB ports on PC's and Laptops delivering better/worse performance than others suggest that even major manufacturers do not always get it right.
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?
Thanks a lot again Mr. Thorsten.
I wonder if the regeneration of the usb signal is really the best strategy to get a good usb signal at the dac input.
I would try, instead, to start at pc level with an usb signal of the best quality possible and then try to transmit it without quality loss.Given that even usb ports on mother boards can suffer from design flaws the solution of an added pci card seems interesting and smart.
I know of the SOTM card ... but it is quite expensive at 350 USD.
Maybe out there a decently performing and cheap card do exist ... of course i am not looking for an high end outcome. Just decent.Thanks again for the very helpful and interesting advice.
Kind regards,
bg
Edits: 05/25/15 05/25/15 05/25/15
If you are looking for a card to improve the USB of you system I would highly recommend the PPA card. Much less than the STOM and better sounding IMO.
+1 on the PPA cards, good price also.
Hi,
There is no particular reason to presume that an add-in card provides electrically a different signal from a port build into the motherboard (this is not to say that other factors may not be involved of course).
Nowadays commonly you find a USB Hub integrated into the motherboard which tends to be located close to the USB Ports and act's there as "re-driver". It may be more worthwhile investigating the motherboards.
Non of this of course helps much if you use long USB Cables of questionable quality, in which case re-driving or re-generating the signal at the end of the cable may be the better choice.
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?
Thanks a lot again and very very interesting.
I just wonder if this issue of the poor usb transimission was already known because this device seems the first to address it.
I saw a lot of usb power isolation devices, but this is the first regenerator. I guess many will follow soon.
Thanks again.
Kind regards,
bg
Hi,
> I just wonder if this issue of the poor usb transimission was
> already known
Did you happen to read the article I referenced?
> because this device seems the first to address it.
USB Repeaters and active USB Cables have been long used outside audio (and by some within).
The first specifically High End Audio targeted use of a USB Hub Chip to re-drive the USB Signal (IIRC) was/is the Shiit Wyred.
That is unless we count the AMR CAS Setup, which was "dealer only - not for resale" Job (sold at our cost to our Partners - as most dealers/distributors of ours were unwilling to set up PC's for this themselves) and meant to demo AMR's DP-777 DAC.
This included a (commercial) active USB Cable, which we found to provide a material improvements over passive cables, but for a number of reasons we did not commercialise that system at the time for High End Audio. With hindsight perhaps a lost opportunity.
> I saw a lot of usb power isolation devices, but this is the
> first regenerator. I guess many will follow soon.
Well, I guess they all follow the Shiit Wyred, or the Belkin active USB Cable we used at AMR...
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?
" Did you happen to read the article I referenced? "
i see. It was well known but maybe considered not important.
Now i see some agreement that is indeed important.
So i expect other devices addressing the issue soon on the market
" USB Repeaters and active USB Cables have been long used outside audio (and by some within). The first specifically High End Audio targeted use of a USB Hub Chip to re-drive the USB Signal (IIRC) was/is the Shiit Wyred. That is unless we count the AMR CAS Setup, which was "dealer only - not for resale" Job (sold at our cost to our Partners - as most dealers/distributors of ours were unwilling to set up PC's for this themselves) and meant to demo AMR's DP-777 DAC.
This included a (commercial) active USB Cable, which we found to provide a material improvements over passive cables, but for a number of reasons we did not commercialise that system at the time for High End Audio. With hindsight perhaps a lost opportunity. "
well i understand that audio high end is a niche.
And pro users probably they just listen for major drops or issues.
" Well, I guess they all follow the Shiit Wyred, or the Belkin active USB Cable we used at AMR...
Ciao T "
Probably i just got carried away by the prose used in the reviews.
Sometimes they exaggerate about the actual improvements.
Thanks again for the valuable advice.
Kind regards,
bg
Nothing is perfect but if you read all the feedback on the Regen you will see almost unanimous belief/feedback that the Regen improves everything from low cost set ups to very expensive ones. I am very sure that Regen addresses something very fundamental in computer audio. Hey if it makes the Lampi better you have to believe Regen is a solid concept/implementation. At $175 how can you go wrong.
Hi and i feel the same. If many people say that it improves even high rez systems this means that it works indeed.
This is a cure. I am wondering about a possible prevention i.e. to improve the quality of the usb signal out at pc level and the quality of its transmission to the dac. I have also a weird idea in mind.
To buy a pci-s to usb card and disconnect the usb out power lines.
I mean, leaving connected only the data + and + pins.
Anyway i am expecting this device will flood the market.
Thanks again.
Kind regards,
bg
See my other post today. Regen addresses the noise created at 8khz by the usb protocol itself. Nobody else does this. Only the Uptone Regen does this, that's why it is unique.
Edits: 05/25/15
Hi and thanks again.
I am a little puzzled ... i read of a usb dac (the Lampi) that is the end of the world.
And i guess also extremely expensive (do not know the price).
and still a 175 dollar device elevates its performance.
They squeeze the last drop from a dac and then they connect it to the computer with an intrinsically flawed usb connection ????
I wonder we all know what is good and what is not ?
we really know what counts and what is marginal ?
Asynchronous usb transmission has been marketed as a godsend freeing from jitter and clock issues ?
and now we discover that is the classical can full or worms ...
I give up and for now i will wait ...
Maybe in some years time the scenario would be more clear.
Thanks again.
Kind regards,
bg
Edits: 05/25/15 05/25/15
I think you are over reacting. Nothing is perfect, yes even usb. But to the extent that a particular noise source can be identified and reduced significantly is not a condemnation of usb. It's a cause for celebration. Other means of connecting a pc -> dac (spdif/toslink/ethernet) all have their limitations as well. In fact since Ethernet sends data in packets like usb it is also subject to packet noise.
"Asynchronous usb transmission has been marketed as a godsend freeing from jitter and clock issues ?
and now we discover that is the classical can full or worms"
Never listen to the marketers. Listen to the engineers.
Hi and yes you are right ... i over-reacted
But you talk about " the noise created at 8khz by the usb protocol itself ". I guess this noise should show up in some measurements
Instead i see many usb dacs with impressive low noise throughout the audio band.
Should i not see some peaks around 8kHz in the noise graphs ?
Thanks again.
Kind regards,
bg
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: