Audio-GD NFB-11.32 & NFB-11 (2014) (Delivery & Impression Thread)!!!
Apr 13, 2013 at 7:55 PM Post #527 of 3,093
Quote:
[size=medium]Hi Kingwa,[/size]
 
[size=medium]reading about usb audio standard I think I know now where lies a problem with firmware #4.[/size]
 
[size=medium]Here for example is not too technical explanation about usb audio:[/size]
[size=medium]  http://www.edn.com/design/consumer/4376143/Fundamentals-of-USB-Audio?page=2[/size]
[size=medium]and you can read that feedback format should be 16.16.[/size]
 
[size=medium]This lead me to usb 2.0 specification and in point 5.12.4.2 you have:[/size]
[size=medium]“For high-speed endpoints, the value shall be encoded in an unsigned 12.13 format which fits into four bytes.[/size]
[size=medium] The value shall be aligned into these four bytes so that the binary point is located between the second and[/size]
[size=medium] the third byte so that it has a 16.16 format.”[/size]
 
[size=medium]Firmware #4 uses 14.18 feedback format. This must lead to incorrect calculation of momentary frequency,[/size]
[size=medium]when to send next portion of data. This is why Linux try to send 44.1kHz audio stream with frequency of  48kHz.[/size]
[size=medium]And this must cause buffer overflow in DAC.[/size]
 
[size=medium]Can someone check firmware #4 under Mac OS? Mac should expect standard 16.16 format too. [/size]
 
[size=medium]Kingwa could you prepare firmware #5 with standard implementation of usb audio with 16.16 feedback format?[/size]
[size=medium]Then we will be able to test it.[/size]


Hello. Was this issue resolved?
 
Apr 14, 2013 at 5:47 AM Post #528 of 3,093
If I am correct the post from Eryk which you posted is "valid" for experience with Squeezebox.
Later Eryk reported, that the problem did not re-appear when using Puppy Linux.
 
As far as I managed to dig into this with the help from Torstein Hegge at alsa-develop team the situation is such:
 
There was a problem with setting the correct sample rate in Linux kernel, that was reported in many cases for CM6631 chip and in at least two cases for the VIA VT1731.
More about this here -
 
http://permalink.gmane.org/gmane.linux.alsa.devel/106773
 
Why the difference with SB and Puppy linux - SB uses ALSA only, anda ALSA sets the DAC clock to highest available sampling rate. So if you start to play the 44.1kHz track, the sample rate will be wrong (48kHz and 14.18bit). If you stop and start the track the sampling would be set correctly.
 
Puppy Linux uses Pulse through ALSA. According to Torstein Hegge Pulse sets the DAC clock to 44.1kHz. That is why it seems that there is no problem when you play tracks with 44.1kHz. The problem would re-appear if you would try to play tracks with sampling rates other that 44.1kHz.
 
 
I can confirm that the patch from the above posted link solves this problem.
The same problem was discovered by Michele Pasciuti using Voyage MPD with Audio-GD DI-3 USB to SPDIF converter.
The patch solved the problem.
 
The Linux 3.9 rc6 kernel already includes the changes from the patch.
 
I don't know if it will be included in the trees of the older kernel versions.
 
Anyhow, to get rid of  the problem with older kernel, kernel source must be patched and the kernel recompiled.
 
 
Hope this helps,
Davor.
 
Apr 17, 2013 at 3:09 AM Post #531 of 3,093
Still have this problem King-wa, under Win XP Service pack 2.
But it's same problem under Windows 7
 
Only 24-bit playback is detected.
Something is wrong
(image missing)
 
No frequency rate control (the window with 44,1, 48, 88,2 and 96Khz)
The option is missing.
 


I've tried reinstall with all kind of drivers, but not sucess.
Seems installer don't detect it right, or my operating system... but both Windows XP and Windows 7? For for me seems, chipset problem. Try to tell VIA to test their drivers with p45 chipset, i have Asus p5q deluxe motherboard.
Thank you.
 
Apr 17, 2013 at 3:10 AM Post #532 of 3,093
Still have this problem King-wa, under Win XP Service pack 2.
But it's same problem under Windows 7
 
Only 24-bit playback is detected.
Something is wrong
(image missing)
 
No frequency rate control (the window with 44,1, 48, 88,2 and 96Khz)
The option is missing.
 


I've tried reinstall with all kind of drivers, but not sucess.
Seems installer don't detect it right, or my operating system... but both Windows XP and Windows 7? For for me seems, chipset problem. Try to tell VIA to test their drivers with p45 chipset, i have Asus p5q deluxe motherboard.
Thank you.
 
Apr 17, 2013 at 3:12 AM Post #533 of 3,093
Still have this problem King-wa, under Win XP Service pack 2.
But it's same problem under Windows 7
 
Only 24-bit playback is detected.
Something is wrong
 

 
 
 
No frequency rate control (the window with 44,1, 48, 88,2 and 96Khz)
The option is missing.
 

 

I've tried reinstall with all kind of drivers, but not sucess.
Seems installer don't detect it right, or my operating system... but both Windows XP and Windows 7? For for me seems, chipset problem. Try to tell VIA to test their drivers with p45 chipset, i have Asus p5q deluxe motherboard. The chipset name Intel P45/ICH10R
http://www.asus.com/Motherboards/P5Q_Deluxe/#specifications
 
Thank you.
 
Apr 17, 2013 at 2:47 PM Post #534 of 3,093
I was wondering if anyone has had any success using an android device to feed USB audio to this thing.  I was thinking about switching to an android device as my primary source.  If anyone has a Galaxy SIII or a note II, as well as a USB OTG cable they can test this out.  
 
Apr 18, 2013 at 2:53 AM Post #535 of 3,093
Quote:
I was wondering if anyone has had any success using an android device to feed USB audio to this thing.  I was thinking about switching to an android device as my primary source.  If anyone has a Galaxy SIII or a note II, as well as a USB OTG cable they can test this out.  

 
Hi Jlav,
 
I've got a Samsung Note 2 and the Compass 2 from Audio-GD. Need to grab a OTG cable somewhere and will test it one of these days. Will report back results when available.
 
Steve
 
Apr 19, 2013 at 1:25 AM Post #537 of 3,093
Quote:
 
Hi Jlav,
 
I've got a Samsung Note 2 and the Compass 2 from Audio-GD. Need to grab a OTG cable somewhere and will test it one of these days. Will report back results when available.
 
Steve


i think i tested it with the note2 and it did not work... i don't really remember if it did though, as it was 6 months ago and i tried only once for moments...
 

Users who are viewing this thread

Back
Top