[nickr]

I have used several usb audio devices with linux in order to get spdif out for DTS/AC3 passthrough in linux - both with xbmc and mythtv - and thay have all worked fine.

Having said that, there are some discouraging posts about for this particular device. eg

http://forum.xbmc.org/showthread.php?tid=90318 (3 years old though).

This guy says he has one going:

http://forum.xbmc.org/member.php?action=...uid=114130

Perhaps proxzen would run up openelec on a usb stick or something and let us know?

[Bjur]

Any news on the IME driver??

[xbs08]

Anyone know a way of getting BT working in OE?
TIA

xbs08
Do you mean bluetooth?

If so go to Openelec Configuration / Openelec Setting menu.
Then select services and go into it, scroll down to the Bluetooth option. Select enable.
Then select Bluetooth from the Openelec side menu and pair your Bluetooth device. It should show in the screen. I've got my PS3 remote working on it.


RLTW

Yes, I mean bluetooth.
I didn't realize it was disabled in OE config

Thanks!

[kees667]

I have used several usb audio devices with linux in order to get spdif out for DTS/AC3 passthrough in linux - both with xbmc and mythtv - and thay have all worked fine.

Having said that, there are some discouraging posts about for this particular device. eg

http://forum.xbmc.org/showthread.php?tid=90318 (3 years old though).

This guy says he has one going:

http://forum.xbmc.org/member.php?action=...uid=114130

Perhaps proxzen would run up openelec on a usb stick or something and let us know?

Thanks. In an unrelated ouya thread the Turtle Beach Audio Advantage Micro II was recommended, prety cheap.
http://forum.xbmc.org/showthread.php?tid...pid1723586

[chaseme]

I'm at a total loss right now. I was running a nightly from Feb on my 2820 and all was fine. Had a power drop today which shut the NUC off. I fired it back up, booted fine but nothing I did would get the wired NIC to work. Wifi wouldn't enable. Finally figured it was time to move up to a beta, so I installed 4.0.1. All good BUT no wired NIC. Wireless connects but the performance is so bad it's unusable. Back to the wired, in BIOS and pre OS booting it's on. Link light etc is fine. As soon as Openelec fires up link light goes off and nothing I've tried will get it to connect! If I disable the wire network in the Openelec settings the link light will come back, enable the connection and link will turn back off. Has anyone dealt with this?

EDIT: Odd, I moved the link from a switch to direct in the router and it's all good now. Meanwhile other devices on the switch are still fine. Eh...whatever!

[nickr]

Could be power surge fubared one port on the switch

[pr0xZen]

I have used several usb audio devices with linux in order to get spdif out for DTS/AC3 passthrough in linux - both with xbmc and mythtv - and thay have all worked fine.

Having said that, there are some discouraging posts about for this particular device. eg

http://forum.xbmc.org/showthread.php?tid=90318 (3 years old though).

This guy says he has one going:

http://forum.xbmc.org/member.php?action=...uid=114130

Perhaps proxzen would run up openelec on a usb stick or something and let us know?

Thanks. In an unrelated ouya thread the Turtle Beach Audio Advantage Micro II was recommended, prety cheap.
http://forum.xbmc.org/showthread.php?tid...pid1723586

I'll see if I can't get around to it in a while - I'm in the midst of an "exam race", just popping by to get a small break from physics and integral equations.

But I let you guys in on a "secret" , that has been very convoluted by Turtle Beach:

I originally intended to go with that exact Turtle Beach USB dongle, for both some android compatability - and very decent Win & Linux driver support. but having a great but older receiver w/o HDMI, I (ideally) also wanted 5.1 analog out (possibility of software decoding of HD audio formats) in addition to optical SPDIF. AND a lower price.

The key component here is not the unit Brand - but the audio controller, and corresponding driver. Guess what audio controller is in the "Turtle Beach Audio Advantage Micro II" ?

Yep - The C-media CM106 - Same as in the DealExtreme one I have, that we've recently spoken of

To put a cherry on top - 2 links on getting this controller to work under linux, including XBMCbuntu (those issues might have been "fixed" by now - threads are old):
http://forum.xbmc.org/showthread.php?tid=95849
http://forum.xbmc.org/showthread.php?tid...pid1154799

You're welcome

[-DDD-]

Maybe interesting for the NUC Bay trail Users here -> http://forum.xbmc.org/showthread.php?tid...pid1732244

Could you retry with: https://github.com/FernetMenta/xbmc/tree/vpp

(MADI is non working for now, intel driver bug - but good for testing). It's not perfect yet, but it has heavy improvements also concerning the rendering.

[joshua.lyon]

I run Win8.1pro on the 4GB stick mentioned at the bottom here (finally got them to list it) - never seen any hint at RAM shortage. Only "low memory warning" I ever got was a "false positive", a confirmed Windows bug.

Interested in this statement. I have now a couple of times had windows 8.1 on my 2820 NUC displaying a "low memory" warning that takes focus away from XBMC, but task manager shows only 1GB of 4GB in use. Interesting you mentioned it's a confirmed bug, is there any resolution or more info you have seen?

In the last few week I get occasioanlly "low memory" pop up also on win 8.1 on NUC i5 haswell with 4Gb RAM.
Since I have changed nothing in the last year except I am up-to-date on XBMC nightlies, I would suspect that this is XBMC problem rather than Win 81, but this is only a speculation.
Does anyone knows more about this problem?

As mentioned it the post - not quite sure what triggers this "false positive", but as it doesn't affect all games and applications that would more likely eat up all available memory - there still has to be something specific that triggers it - and thus it must be a way to remedy it. I tried suggested solutions like bumping up swapfile size considerably, and making sure xbmc ran at the same resolution as windows itself, alternating "true fullscreen"/ fullscreen window - but saw no difference in the occurence of the errors. So I followed the instructions linked, and kept the swap file at ~2x RAM size. Have not seen a warning in 45 days and counting.

I would caution against assuming that this is a false-positive. While I am not saying it was not a false positive in your case, I was experiencing a similar set of circumstances as you were and it turned out there was a memory issue. Occassionally I would come back to my XBMC PC displaying a low memory warning and when I would check memory utilization I would find the system in a normal state (eg. 1 GB usage in your example). After experiencing this several times, I figured something must be wrong so I put a System Monitor trace on memory and found that when I wasn't using the system the memory usage would slowly creep up and max out... when the low memory warning would come up it would force XBMC out of full screen (eg. bring the Windows warning to the front) and would allow the memory to drain back to a normal value. So when I would go to check the memory usage everything would be fine but it was actually the history I needed to track in order to see the low memory event.

My issue ended up having something to do with the resolution on my XBMC PC changing whenever I would turn off the receiver, which somehow wreaked havoc with XBMCs memory utilization. I have the HDMI from my HTPC routed through my receiver and then onto the TV. It turns out that whenever I would turn off my receiver the HTPC was switching to a much lower resolution. I tried several different fixes and I don't recall which (combination) of them resolved the issue, but here they are:

1. XBMC: Use fullscreen window rather than true fullscreen: CHECKED
   
2. WINDOWS: Manually set page file to 200MB - 1 GB
   
3. HDMI: Tape pin 19 of HDMI cable
   
4. GPU: Force output resolution to be fixed to 1920x1080
   

[Crssi]

Sorry that I havent report back... I have forgot to.
I have solved this issue day or two after reporting here.
I did exactly one thing and this is what joshua.lyon did under number 1.

XBMC: Use fullscreen window rather than true fullscreen: CHECKED

After further researching I am almost 100% convinced that this is Intel Video driver problem. People have reported the same problem also when playing games (full screen) on Intel HD video cards.

[RideLikeTheWind]

Hi,
Having a few issues with getting my SSD to dual boot with Openelec and Ubuntu 14.04. I can get Ubuntu to run OK first then when I install Openelec afterwards are rerun update - grub it doesn't find it so I update the custom part of grub manually. After updating it puts Openelec into the grub.cfg file so I'm thinking all this is OK.
The problem starts when I reboot and there isn't any grub selection select at startup. I've removed the boot from USB option in the BIOS but am now at a loss? What other options do I need to select in the BIOS as well as anything else I'm missing?

Many thanks in advance.

RLTW

[nickr]

To get the grub menu displaying at startup you need to edit /etc/default/grub. Here is what the first portion of mine contains.

Code:

GRUB_DEFAULT=0
#GRUB_HIDDEN_TIMEOUT=0
GRUB_HIDDEN_TIMEOUT_QUIET=true
GRUB_TIMEOUT=10
GRUB_DISTRIBUTOR=`lsb_release -i -s 2> /dev/null || echo Debian`
GRUB_CMDLINE_LINUX_DEFAULT="quiet splash"
GRUB_CMDLINE_LINUX=""

and it brings up the menu. You have to run

Code:

sudo update-grub

after editing the file.

[RideLikeTheWind]

@nickr
Thanks for that it worked a treat!

RLTW

[scarecrow420]

I run Win8.1pro on the 4GB stick mentioned at the bottom here (finally got them to list it) - never seen any hint at RAM shortage. Only "low memory warning" I ever got was a "false positive", a confirmed Windows bug.

Interested in this statement. I have now a couple of times had windows 8.1 on my 2820 NUC displaying a "low memory" warning that takes focus away from XBMC, but task manager shows only 1GB of 4GB in use. Interesting you mentioned it's a confirmed bug, is there any resolution or more info you have seen?

In the last few week I get occasioanlly "low memory" pop up also on win 8.1 on NUC i5 haswell with 4Gb RAM.
Since I have changed nothing in the last year except I am up-to-date on XBMC nightlies, I would suspect that this is XBMC problem rather than Win 81, but this is only a speculation.
Does anyone knows more about this problem?

As mentioned it the post - not quite sure what triggers this "false positive", but as it doesn't affect all games and applications that would more likely eat up all available memory - there still has to be something specific that triggers it - and thus it must be a way to remedy it. I tried suggested solutions like bumping up swapfile size considerably, and making sure xbmc ran at the same resolution as windows itself, alternating "true fullscreen"/ fullscreen window - but saw no difference in the occurence of the errors. So I followed the instructions linked, and kept the swap file at ~2x RAM size. Have not seen a warning in 45 days and counting.

I would caution against assuming that this is a false-positive. While I am not saying it was not a false positive in your case, I was experiencing a similar set of circumstances as you were and it turned out there was a memory issue. Occassionally I would come back to my XBMC PC displaying a low memory warning and when I would check memory utilization I would find the system in a normal state (eg. 1 GB usage in your example). After experiencing this several times, I figured something must be wrong so I put a System Monitor trace on memory and found that when I wasn't using the system the memory usage would slowly creep up and max out... when the low memory warning would come up it would force XBMC out of full screen (eg. bring the Windows warning to the front) and would allow the memory to drain back to a normal value. So when I would go to check the memory usage everything would be fine but it was actually the history I needed to track in order to see the low memory event.

My issue ended up having something to do with the resolution on my XBMC PC changing whenever I would turn off the receiver, which somehow wreaked havoc with XBMCs memory utilization. I have the HDMI from my HTPC routed through my receiver and then onto the TV. It turns out that whenever I would turn off my receiver the HTPC was switching to a much lower resolution. I tried several different fixes and I don't recall which (combination) of them resolved the issue, but here they are:

1. XBMC: Use fullscreen window rather than true fullscreen: CHECKED
   
2. WINDOWS: Manually set page file to 200MB - 1 GB
   
3. HDMI: Tape pin 19 of HDMI cable
   
4. GPU: Force output resolution to be fixed to 1920x1080
   

As others have said, using a fullscreen window rather than true full screen seems to avoid the out of memory error. But that gives me a different issue. When using fullscreen window, and the AVR/TV are switched off, the HTPC resolution changes and XBMC gets "bigger" on screen. Then when everything comes back on again, windows has reset to 1080P resolution but XBMC is stuck in a zoomed in mode, taking up the full screen but all zoomed in (only seeing perhaps 25-50% of the top left of the screen, maximised to fll the display). This is the only niggling issue I have! Obviously I just want my wife to be able to switch things on and have it ready to go but instead it's either XBMC all large and weird, meaning need to close/reopen it or hit the fullscreen button twice... or windows popped up an out of memory error which takes focus away from XBMC and needs to be dismissed and then alt tab back into XBMC. Neither are really wife friendly I tried the various fixes to the zoom issue, such as using a tool to create an INF driver from static EDID data to force the monitor drier to always be in 1080P etc, and running a remote VNC console to the HTPC to try and set the default resolution when the AVR/T are off (it doesn't seem to work - VNC console never indicates the res changes but yet sometimes when you turn things back on XBMC is all zoomed in and weird). I wish the memory leak issue could be fixed then I would just run in true fullscreen which also eliminates problems with the windows taskbar taking focus from XBMC sometimes etc. I guess my hope is for Intel to release new graphics drivers, if that is the cause of the memory leak

[pr0xZen]

If changing to full-screen window is not a working solution due to resolution change - I would suggest trying out the Pin 19 trick . Its not difficult to do, and easily reversible. The concept is to avvoid, among other things, resolution change - through the NUC and AVR/TV being unable to tell each other that one of them "shut down" or went to "power saving" mode. As far as the NUC can tell, the AVR (or TVs) power state never changes.