|
|
Post by meridion on Feb 21, 2015 8:49:21 GMT -5
I have been having issues with the USB input with my Mac Mini - every so often, the Mac seems to think the DAC only supports 44 & 48k and it usually takes rebooting the XMC to bring it back. I also had an issue with the XMC reporting the word length as 17 bits (!?) on the info screen, even though the Mac SW thought it was sending out 16 or 24 bits. It took a reset on the XMC to clear that. It has gotten to the point that I have gone back to optical. I have a Auralic Aries coming, so I guess I'll see how that fares with the USB input (the Aries supports class 2 USB DAC's, so the XMC should work). Of course the Aries has S/PDIF and AES/EBU as well. I have also had issues where the MAC thought that it could only output 44 or 48K. As stated a reboot usually fixes the issue. It seems to be more stable on the windows side. I'm having the same issue sometimes with my Ubuntu/Linux PC. The linux driver does not seem to recognize XMC-1 USB correctly after a XMC-1 bootup. It sees just an 'Audio-1' device (capable of 44/48khz) instead of the normal 'XMC-1' USB hires audio device. Rebooting XMC-1 always helped. |
|
|
|
Post by ÈlTwo on Feb 21, 2015 11:57:14 GMT -5
I don't have any issue with my Linux PC (running AP-Linux), and it's some pretty old hardware.
What hardware are you using for your PC and what program are you using to send the music out to the XMC-1?
My set up uses MPD as the software, and ALSA, which is part of the kernel (well there is a dummy driver, which just points back to the kernel). I'm not sure about Ubuntu, whether it has ALSA as part of the kernel, or if it uses Pulse Audio.
|
|
|
|
Post by meridion on Feb 22, 2015 11:47:52 GMT -5
I don't have any issue with my Linux PC (running AP-Linux), and it's some pretty old hardware. What hardware are you using for your PC and what program are you using to send the music out to the XMC-1? My set up uses MPD as the software, and ALSA, which is part of the kernel (well there is a dummy driver, which just points back to the kernel). I'm not sure about Ubuntu, whether it has ALSA as part of the kernel, or if it uses Pulse Audio. It's lubuntu running on an Odroid (ARM) system (Ubuntu 14.04.1 LTS - GNU/Linux 3.8.13.23 armv7l). Ubuntu has ALSA as part of the kernel. My setup consists of Squeezelite using ALSA. The Odroid system is always on, but the XMC-1 is powered off (to standby with video off) when not used. XMC-1 does not get recognized properly about every 5th to 10th power-on of XMC-1. |
|
|
|
Post by ÈlTwo on Feb 22, 2015 19:05:10 GMT -5
You could try to remove pulse audio:
sudo apt-get remove --purge pulseaudio alsa-base
and then reinstall alsa:
sudo apt-get install alsa-base
Logout from Ubuntu and log in again...
I'm not sure, but I think this may be an ARM processor issue.
|
|
|
|
Post by michaelhifi on Apr 3, 2015 19:16:35 GMT -5
It's been a marathon. I know I've run over 50...
I haven't bitched too much although my Win 7 had issues as did my win 8 pro, but, so typically me, forged ahead with Win 10. I got Dirac LE to work but it was painful. I gave up on Dirac Live in Win 10. Stupid pushing stupid. 
So I formatted my drive and installed Win 8.1 pro. It's raw except for update and MB drivers. So, I would have hoped, or maybe thinking I would actually get lucky, to get the USB driver loaded once and for all! You know, I bought into the DIRAC Live software and wish to try it 
NOT. I get "unknown error" after the install. "System has not been modified". I just can't seem to get past step "a" without serious intervention. 3 OS's - 0 luck.
Ran as administrator. Tried the normal setup then again using Win 8.1 setup. XMC is plugged in. I generally fault user error, the ID10T error, being X IT. I know I'm getting old and feeble but... help me out here. |
|
|
|
Post by rocky500 on Apr 3, 2015 20:54:25 GMT -5
It's been a marathon. I know I've run over 50...
I haven't bitched too much although my Win 7 had issues as did my win 8 pro, but, so typically me, forged ahead with Win 10. I got Dirac LE to work but it was painful. I gave up on Dirac Live in Win 10. Stupid pushing stupid.
So I formatted my drive and installed Win 8.1 pro. It's raw except for update and MB drivers. So, I would have hoped, or maybe thinking I would actually get lucky, to get the USB driver loaded once and for all! You know, I bought into the DIRAC Live software and wish to try it
NOT. I get "unknown error" after the install. "System has not been modified". I just can't seem to get past step "a" without serious intervention. 3 OS's - 0 luck.
Ran as administrator. Tried the normal setup then again using Win 8.1 setup. XMC is plugged in. I generally fault user error, the ID10T error, being X IT. I know I'm getting old and feeble but... help me out here. It could be PC has some sort of fault? Faulty ram, ports etc. Network cable to XMC-1 or port it is using faulty? XMC-1 network cable goes to router/switch? I use Win 8.1 Pro and all is well. Does Dirac install perfectly fine? Maybe run an update on line to get windows current. I use the Schiit drivers to install for connecting USB to the XMC-1. schiit.com/drivers/ |
|
|
|
Post by michaelhifi on Apr 4, 2015 11:44:42 GMT -5
Feeling like mile 21. Getting tired and wish to finish - the install.
I got the Schiit drivers to install. Everything seems to work except I get no response from the mic. I guess that means nothing works...
In manage audio devices I get the C-Media SPDIF in audio and HD microphone in microphone. If I turn the gain to max I see the bar move but it's not related to the mic. I get sweeps but they are not seen.
Now what?
|
|
|
|
Post by rocky500 on Apr 4, 2015 19:55:00 GMT -5
Feeling like mile 21. Getting tired and wish to finish - the install. I got the Schiit drivers to install. Everything seems to work except I get no response from the mic. I guess that means nothing works... In manage audio devices I get the C-Media SPDIF in audio and HD microphone in microphone. If I turn the gain to max I see the bar move but it's not related to the mic. I get sweeps but they are not seen. Now what? Are you using an extension cable with the mic? These have been known to cause problems. If using one, try without, just to see if it works. I use a USB3.0 extension cable that is 2m long. |
|
|
|
Post by michaelhifi on Apr 6, 2015 13:22:30 GMT -5
I tried with no extension. Didn't change anything.
I'm going to call Emotiva is a few minutes. I may have them send me a mic even though the one I have did work on LE.
Thanks for trying to help.
Michael
|
|
|
|
Post by michaelhifi on Apr 7, 2015 8:48:48 GMT -5
Fixed!
|
|
|
|
Post by rocky500 on Apr 7, 2015 20:22:56 GMT -5
Great!. What was the fix? |
|
Chris
Emo VIPs 
Posts: 424
|
Post by Chris on Apr 21, 2015 19:51:20 GMT -5
Hi,
I've been trying out my RPi running PicorePlayer 19i hooked up via USB to my XMC-1 Firmware 3.0.
I have tested playing files up to 192k/24b and everything is running okay. I also can playback DSD files by using the DSD/DoP plugin which uses LMSto decode to PCM to send to the RPi and out to the XMC-1.
I just noticed today that while the XMC-1 is not displaying the correct bit rate? The Sample rate changes correctly on known files of various rates (44k, 96k, 192k) but the bit rate seems to be stuck saying 24 bits when I know the file is 16 bits. I'm not sure if this is an issue with the PicorePlayer code? But, I seem to remember an earlier thread for firmware prior to 3.0 that noted a problem with bit rate display? When I test using an S/PDIF input (Squeezebox) Touch, I am also seeing it stuck on 24 bit for a known 16 bit file.
Is this still a known issue (e.g bit depth display) with the XMC-1?
-CB
|
|
|
|
Post by rocky500 on Apr 21, 2015 22:20:36 GMT -5
Hi, I've been trying out my RPi running PicorePlayer 19i hooked up via USB to my XMC-1 Firmware 3.0. I have tested playing files up to 192k/24b and everything is running okay. I also can playback DSD files by using the DSD/DoP plugin which uses LMSto decode to PCM to send to the RPi and out to the XMC-1. I just noticed today that while the XMC-1 is not displaying the correct bit rate? The Sample rate changes correctly on known files of various rates (44k, 96k, 192k) but the bit rate seems to be stuck saying 24 bits when I know the file is 16 bits. I'm not sure if this is an issue with the PicorePlayer code? But, I seem to remember an earlier thread for firmware prior to 3.0 that noted a problem with bit rate display? When I test using an S/PDIF input (Squeezebox) Touch, I am also seeing it stuck on 24 bit for a known 16 bit file. Is this still a known issue (e.g bit depth display) with the XMC-1? -CB Could it be PicorePlayer is the problem? I did a quick search and someone else was stuck in 32bit all the time. forums.slimdevices.com/showthread.php?99253-Help-on-some-ALSA-squeezelite-raspberry-piCorePlayer-issues |
|
Chris
Emo VIPs
Posts: 424
|
Post by Chris on Apr 25, 2015 21:23:20 GMT -5
Hi, I've been trying out my RPi running PicorePlayer 19i hooked up via USB to my XMC-1 Firmware 3.0. I have tested playing files up to 192k/24b and everything is running okay. I also can playback DSD files by using the DSD/DoP plugin which uses LMSto decode to PCM to send to the RPi and out to the XMC-1. I just noticed today that while the XMC-1 is not displaying the correct bit rate? The Sample rate changes correctly on known files of various rates (44k, 96k, 192k) but the bit rate seems to be stuck saying 24 bits when I know the file is 16 bits. I'm not sure if this is an issue with the PicorePlayer code? But, I seem to remember an earlier thread for firmware prior to 3.0 that noted a problem with bit rate display? When I test using an S/PDIF input (Squeezebox) Touch, I am also seeing it stuck on 24 bit for a known 16 bit file. Is this still a known issue (e.g bit depth display) with the XMC-1? -CB Could it be PicorePlayer is the problem? I did a quick search and someone else was stuck in 32bit all the time. forums.slimdevices.com/showthread.php?99253-Help-on-some-ALSA-squeezelite-raspberry-piCorePlayer-issuesAnything is possible but after further research and testing with another DAC. I think I might understand what is going on. Using the Alsa drivers that come with the Picoreplayer I can set various output settings for each DAC. In the case of the XMC-1, it would appear that only the 16 bit driver mode (Alsa -output 80:4:16:) works by default with the XMC-1. Using any of the 24 bit settings result in playback with noise and jitter (unlistenable). I think by default the PicorePlayer sees the XMC-1 as a UAC1 16 bit device so higher sample rates are being down converted. I would prefer bit perfect if I can attain it. When I was testing with my Benchmark DAC2 HGC, I was having trouble getting any sound until I researched an determined that I could force the Benchmark into UAC2 via a remote control setting documented in the manual. Once I changed to UAC2 mode I was able to get it working at all bit and sample rates. I'm not sure what more I can do on the Alsa drivers of the Picoreplayer (I will enquire) but I was wondering if it would be possible to add a setup item under the USB streaming menu that forces UAC2 mode on the XMC-1 just like the DAC2 HGC is able to do? This would be very helpful for people like me that are kind of on the cutting edge of USB streaming. So, it would be a very nice addition if Emotiva could manage it? Pretty please Emotiva? Thanks. -CB |
|
Chris
Emo VIPs
Posts: 424
|
Post by Chris on May 9, 2015 10:16:30 GMT -5
Yesterday, I had a breakthrough with USB streaming to the XMC-1. As I continued to test, it became evident that I could only USB Stream from my Picoreplayer Raspberry Pi in 16 bit mode. I had to use -o HW:0,0 -a 80:4:16:0 which are the Alsa settings. Any other settings resulted in noise and distortion on the output. I thought I was at a dead end but due to the tireless efforts of Steen (author of the Picoreplayer distro) he uncovered a possible solution that has brought to light an OS bug related to CMedia USB chips that now is being worked on. Here is the fix that has worked for me: If you are willing to edit the cmdline.txt file you could try this:
1. Delete your string looking something like this: dwc_otg.fiq_fsm_mask=0x7 (the 0x7 can be a different value - doesn't matter simply delete it all).
2. Then add this: dwc_otg.fiq_fsm_enable=0 to your cmdline.txt file.
Note:
If you insert the SD-card in a windows computer you can easily see and edit the cmdline.txt file.
Then save the changes and reboot your raspberry.
Regards
SteenUSB Audio bug discussionI can now use alsa settings of -o HW:0,0 -a 80:4:: which means I am now running in 24bit mode. The discovery of this bug and this workaround until a real fix is available will probably help a number of DACs that use the the Cmedia chipset including Emotiva and some of the Schitt DACs. -CB |
|
|
|
Post by makutaku on May 19, 2015 15:39:59 GMT -5
I am stuck with an issue that appears to be a bug in how XMC-1 handles changes to USB Stream input:
I have my HTPC connected to XMC-1 via HDMI1 and USB Stream.
Only JRiver plays thru the USB in exclusive mode, all other applications via HDMI.
I also have my Oppo connected via HDMI2.
Input buttons:
* Input1 => Audio=HDMI1 Video=HDMI1
* Input2 => Audio=USBStream Video=HDMI1
* Input3 => Audio=HDMI2 Video=HDMI2
What I noticed is that often (but not always), changing from Input1 to Input2 and playing a track in JRiver, no sound comes up, despite JRiver saying it's playing.
However, by changing to input3 and then back to input2, I can finally hear what JRiver is playing.
Basically, forcing a HDMI handshake forces XMC-1 to finally play what is being streamed.
|
|
|
|
Post by millst on May 19, 2015 16:21:39 GMT -5
Yeah, the Raspberry Pi USB implementation is awful. It's been problematic since the start. Overall, the software side has improved things over time, but there is only so much that can be done due to the hardware limitations.
USB audio class I doesn't have enough bandwidth to go higher than 96kHz. Well, maybe if you went mono. I didn't think people ever had much luck getting class II devices working. I wouldn't worry about seeing 24-bit [or 32-bit] instead of 16-bit. It's still bit perfect. The extra bits are just padding for a 16-bit signal.
-tm
|
|
|
|
Post by qdtjni on May 19, 2015 17:36:49 GMT -5
Seems like RPI works fine with XMOS based USB-DACs but not well at all with CMEDIA stuff at least in my experience.
Then again the only CMEDIA based USB-DAC I have is the XMC-1 so not that much of sample size.
Maybe it works better with RPI2?
|
|
|
|
Post by millst on May 19, 2015 22:22:18 GMT -5
Maybe the XMOS has better driver support, but can't say I've done much research here. A $35 computer designed for education wouldn't be my first choice for trouble-free driving of a high-end audio system. I'm using mine for background music at 16-bit and 44.1/48kHz. It works great for that so far.
The USB implementation is exactly the same on the Pi2. Maybe a few less dropped packets with the extra CPU cycles.
-tm
|
|
|
|
Post by qdtjni on May 20, 2015 4:46:18 GMT -5
Maybe the XMOS has better driver support, but can't say I've done much research here. A $35 computer designed for education wouldn't be my first choice for trouble-free driving of a high-end audio system. I'm using mine for background music at 16-bit and 44.1/48kHz. It works great for that so far. I have tried with at least 5 XMOS based DACs and with they all work fine with highres, worst case with -a 80:4:: Agree that there are better choices. I'll try Chris suggested workaround, when I found some time and spirit for it. |
|
