Is jitter an issue with usb signals ?

Hi !
i have read that usb signals, like spdif signals, have a certain amount of jitter
This jitter can be increased by a bad connection (bad cable, poor impedance matching, etc.).
And that usb receiver chips (like Tenor, Xmos, ect.) can have problems in processing high jitter signals .

I have the following questions:
1) is this true ?
2) is there a way to reduce usb signal jitter ?

Thanks and regards, gino
 
Last edited:
If the jitter is bad enough, you'll get drop-outs or pops. If you don't get dropouts and pops, it's good enough.
ANY competently engineered USB device will be good enough in this respect unless you do something silly interconnecting it.

Hi and thanks a lot for the very useful advice.
So if i do not experience clicks or pops it is fine.
That is what i needed to know. I have no instrument to measure jitter.
I will rely on my ear. :rolleyes:
Thanks again, gino :D
 
USB isochronous has jitter issues so go with USB asynchronous, this eliminates jitter totally.

Hi and thanks a lot for the helpful explanation.
I was wondering because i read that sub receiver chips can have problem in processing high jitter signals.
From this the sensitivity towards cable and impedance matching and so on.

Note though that USB normally has issues with common-mode noise which some call 'jitter'

this sounds interesting. What is it ? is there a way to suppress it ?
Thanks again, gino
 
USB signals are inherently high jitter in themselves, but this is unrelated to the audio clock, its the nature of USB data to be transmitted in bursts.

Common-mode noise usually emanates from a computer's switched-mode power supply. Its only an issue when it finds its way into your audio kit via your cable screens. You can prevent that by using all balanced connections for the audio interconnects (where there are no pin1 design problems to undo part of the advantage of balanced).
 
USB signals are inherently high jitter in themselves, but this is unrelated to the audio clock, its the nature of USB data to be transmitted in bursts.
Common-mode noise usually emanates from a computer's switched-mode power supply. Its only an issue when it finds its way into your audio kit via your cable screens. You can prevent that by using all balanced connections for the audio interconnects (where there are no pin1 design problems to undo part of the advantage of balanced).

Thanks a lot again. Very interesting.
Just one last question ... the use of a balanced digital connection AES/EBU can have the same effect of suppressing this noise coming from the pc ?
I have this option.
Thanks again, gino
 
If your PC has an AES/EBU output, or an optical (TOSLINK) output, these would be likely a better choice than USB yes as they'll practically eliminate the CM noise. If you're going to use the AES/EBU you might want to check that the wiring of the XLR input to your DAC does follow best practice in terms of how pin1 is terminated. Best practice isn't too frequently followed in my experience.
 
If your PC has an AES/EBU output, or an optical (TOSLINK) output, these would be likely a better choice than USB yes as they'll practically eliminate the CM noise.

No. But i have a usb to AES/EBU converter that i could use.
Optical is usually of low quality on cheap products.
I prefer not to use it.

If you're going to use the AES/EBU you might want to check that the wiring of the XLR input to your DAC does follow best practice in terms of how pin1 is terminated. Best practice isn't too frequently followed in my experience.

I will do for sure. Thanks a lot again for the very useful advice.
The pc is too handy as a source to be overlooked.
I am the opposite of a turntable guy ...
Kind regards, gino
 
I have never had any problems with USB.
I use a USB DAC and it has never popped or gone wrong.
The only thing that might cause problems is excessively long USB cables.
Or the USB bus being overloaded

Hi and thanks a lot for the vauable reply
Does your dac use usb power to work ?
I think we can agree on the fact that usb bus is weak and with dirty voltage.
This is not good in general.
Regards, gino
 

YashN

Member
2013-12-04 12:55 am
Canada
Thanks again. I think i should visit a doctor ... :( i am a little obsessed these days with issues that are no issues at all.
Regards, gino

Gino, not every issue is jitter, even with USB. There is a lot of info to digest over at CA, so take time to do this and really dig into it, otherwise you'll get conflicting and confusing answers, especially by those who think erroneously that async USB solved everything for audiophilia (it hasn't).
 

diyaudnut

Member
2015-02-24 5:41 am
USB 2.0 uses NRZI encoding for digital data transmission and is based on clock recovery. The standard provides a jitter specification. But as someone said, if the digital data is safely retrieved there is no noticeable effect, else the pops and clicks. Any reasonable USB interconnect should provide reliable data transmission.

Where i have personally noticed jitter to be a problem is a jittery clock on I2S interface causing a receiving DAC to not achieve a lock on the stream.
 
Gino, not every issue is jitter, even with USB. There is a lot of info to digest over at CA, so take time to do this and really dig into it, otherwise you'll get conflicting and confusing answers, especially by those who think erroneously that async USB solved everything for audiophilia (it hasn't).

Some links would be better as there is so much rubbish surrounding any real relevant information that it is hard to determine real facts from the scaremongering and rubbish...
Abraxalito has nailed the problems in his posts... why complicate things, unless you want to jump on the USB hub path that many are going down (oh sorry they are not USB hubs they are Reclockers or Regens or whatever they are calling them, USB hub to normal people, of course those supporting and selling these things will fill your head with horror stories regarding the USB interface.....
 
Gino, not every issue is jitter, even with USB. There is a lot of info to digest over at CA, so take time to do this and really dig into it, otherwise you'll get conflicting and confusing answers, especially by those who think erroneously that async USB solved everything for audiophilia (it hasn't)

Hi ! thanks again. I tried to reply but the site was under maintenance.
I am going to do some tests with isolators and other things. I have already heard some differences.
Main and maybe only goal is to block the power coming from the pc and replace it with a cleaner voltage.
5VDC clean are always better than 5VDC dirty.
Thanks again, gino
 
Last edited:

YashN

Member
2013-12-04 12:55 am
Canada
Abraxalito has nailed the problems in his posts... why complicate things, unless you want to jump on the USB hub path that many are going down (

Things are much more complex than that when you want great results as an audiophile.

If you think USB was already perfect to begin with, then you don't need the aforementioned devices and you don't need additional information.

The biggest issue with the usual USB is the normal cable: it has an analogue signal, and therefore is subject to signal integrity issues. When the SI issues have to be dealt with to determine the bits, the USB receiver processing creates additional noise which affects the DAC (most of them)...

Look for our friend Gino's post over at computeraudiophile.com and/or read John Swenson if you really want to know more, or since this is diyaudio, do your own tests, as I do.

On the other hand, if you think you know everything there is to know about the audiophile subject, then don't bother.
 
Hi !
it is quite clear to me that usb has not been a good choice .
Personally when i find something satisfactory i do not look for something else.
And actually from what i read a new very promising solution is Gigabit Ethernet:

[IMGDEAD]https://cdn1.bigcommerce.com/server3000/b3f72/products/428/images/1218/photoviewer_dawsystem_dante_accelerator_angle__17041.1382376298.1280.1280.jpg?c=2[/IMGDEAD]

https://us.focusrite.com/ethernet-audio-interfaces/rednet

https://us.focusrite.com/ethernet-audio-interfaces/rednet-am2

The prices are still very very high but likely to come down at least for more basic unit.
I am sure we will see pretty soon adapters and interfaces using this standard.
Kind regards, gino

P.S the sensation that usb is a joke starts from the quality of the connector.
Like HDMI compared to DVI ... there is an abyss of quality. Because DVI is for serious professionals.
Good quality Ethernet looks much more serious that the average usb plug.
Someone told me that in the end usb was born for mice, keyboards and printers and there should be used.
Maybe he is very right.
 
Last edited: