Posts: 315
Joined: Mar 2013
Reputation:
7
2014-10-12, 23:41
(This post was last modified: 2014-10-12, 23:49 by Sunflux.)
Downgraded to 4.1.6... sorry to say the issue remains exactly the same.
Reading those links, I've never had it drop out during a normal session, only after waking up (although to be fair I haven't tested that too extensively, although definitely never "after a few minutes" like the one guy says). Also, it always works if I re-plug into the same USB port, unlike the second link. In fact I don't even have to unplug, I can just suspend/resume and it works. Really seems software-based to me, but it could be some quirk with this particular receiver.
Should I try that autosuspend setting?
Posts: 4,512
Joined: Nov 2008
Reputation:
224
not sure, I haven't heard of anyone else having your exact issue, with the same hardware - which I have here, and don't have any issues
Posts: 315
Joined: Mar 2013
Reputation:
7
2014-10-13, 05:25
(This post was last modified: 2014-10-13, 05:37 by Sunflux.)
So you have the IR605? Mine's probably 18 months old, could be a firmware difference or something.
I do have a HP 5188-1667 MCE receiver on the way to test, but it will be a week+ for shipping. I also have a classic Philips MCE receiver (one of the big black ones), but it's hooked up in a way that would be a real pain to remove for testing.
Unless you know of a way to easily get an old Streamzap IR receiver to work (doesn't seem plug-and-play). And, I can always buy a FLIRC, although does it emulate keypresses vs MCE button presses?
EDIT: It could be my imagination, but the problem might be significantly reduced in 4.1.6 (just testing turning the system on/off, on/off, on/off). I've also noticed that sometimes the receiver does not respond to the first press of POWER when in suspend mode (have noticed this from minute one), but does to the second - and when that happens, the chances that it will be non-responsive upon wakeup are greater. So I've added a macro that sends a non-functional MCE command just prior to power, which seems to entirely solve the double-press issue... and makes the system more reliable.
Posts: 4,512
Joined: Nov 2008
Reputation:
224
no, sorry, I have the HP MCE one that was used in those other threads I linked.
Posts: 315
Joined: Mar 2013
Reputation:
7
Ah. Well, it's big and ugly, but if it works...
With that said, at the moment I'm having great difficulty getting the problem to trigger in 4.1.6 with my improved power on command. Once I've spent a day or two with it like this, I'll let it upgrade to 4.2.1 and see what happens.
Posts: 315
Joined: Mar 2013
Reputation:
7
Okay, as weird as it sounds... so far with 4.1.6, if the unit's asleep, sending an unused MCE command *before* sending POWER seems to have 100% cured the issue with the system randomly being non-responsive upon wakeup.
No idea why this is, but that doesn't mean I won't make a wild guess. I'm not certain how USB communications work in this case, but based on that one wacky log entry I'm almost wondering if the receiver is waiting to finish sending the concluding "_UP" command, but the system is going into sleep mode before this happens. Since sending POWER while the system is off does NOT actually transmit that to the PC, when the system does wake up the receiver finishes sending the last valid command it had - or at least the tail "_UP" portion, or perhaps some garbled data relating to it... which in turn occasionally locks up LIRC. However, sending a random command while the system is off causes the receiver to flush its buffer so this doesn't happen. I'm probably wrong.
Anyways, I guess the next step is to see if this fix continues to work in 4.2.1, or if the issue you linked earlier also applies (nothing that says I can't be seeing TWO problems at once!)
Posts: 4,512
Joined: Nov 2008
Reputation:
224
when the system is suspended, IR receiver is in a low power state (D3), and any commands received outside of power toggle are ignored/dropped. When a power toggle is sent, the receiver passes it along, causing the system to wake from suspend. The only reason I can think of that your workaround shows improved behavior is buggy firmware on the IR receiver unit itself.
Posts: 315
Joined: Mar 2013
Reputation:
7
I don't think it passes along the raw IR code like it would any other normal command, so much as self-interprets it and then sends the USB wake command... which sounds like a very different process that may not impact in-progress transmission of normal commands. Don't get me wrong - almost certainly a firmware issue, but as far as I can tell there are only two possible reasons for what I'm seeing:
!) after successfully powering on the system, something causes the IR receiver to enter a state whereby it can no longer send commands to the system
2) something is confusing the IR receiver prior to or during sleep, whereby its first communications after wakeup is causing the software to no longer be able to interpret its commands
Since I can see no signs that the device itself is crashing or otherwise requires resetting, I'm almost willing to bet it's scenario #2. And again, I'm sure it's a firmware glitch... but it might be one that's aggravated by an already non-ideal situation.
And so far so good with 4.2.1.
Posts: 15
Joined: Aug 2014
Reputation:
0
I have a similar problem to Nitrous, but I'm not sure if it is to do with my XBMC setup or it's a problem with my ChromeBox setup.
Basically, if I am watching either TV or a movie (it seems regardless of which Add-on), the picture will randomly freeze after about 30 minutes or so. There is no buffering and the sound continues, so the stream hasn't dropped but the picture is completely frozen and the keyboard won't respond. The only thing I can do is unplug the whole box and reboot it.
Is there a log that I can turn on to see what might be the cause? (Please excuse my technical ignorance on this!)
Any help gratefully received.
Duncan
Posts: 20
Joined: Apr 2012
Reputation:
0
Hi Matt - awesome work - thanks .
just upgraded from my vintage but very well used ATV1 with crystalbuntu to the ASUS -CB - (M004U), the only mod I added was an extra 2gig memory card in the open slot. I set it up stand alone as its a replacement for the family's main XBMC box.
set up was very easy and took about 30 minutes
the Turtle beach USB digital audio adapter as per the WIKI works great and does pass DD and DTS to my 5.1 receiver via SPDIF
I "borrowed" my sons PS3 - bluetooth DVD remote and it works - good - plug and play.
I have a couple of question:
1) does your stand alone custom coreboot (firmware/bios) keep the CB(M004U) updated or do you have to manually update on a regular basis to keep up with your changes to the coreboot on GitHub ?
2) when i check the version of openelec it shows as 4.2.0 and auto updates are checked. I have pushed the ":check for updates" and nothing. Does the standalone openelec need to be manually updated to achieve 4.2.1 ?
look forward to your reply
Posts: 19,982
Joined: May 2009
Reputation:
451
nickr
Retired Team-Kodi Member
Posts: 19,982
After you chack for updates, is there a file in /storage/.update ?
Have you rebooted after "check for updates?
If I have helped you or increased your knowledge, click the 'thumbs up' button to give thanks :) (People with less than 20 posts won't see the "thumbs up" button.)
Posts: 1
Joined: Jan 2011
Reputation:
0
hello. im having an issue with the script. when i try to do stand alone setup. and openelec setup installer. i get openelec missing tool parted. tried several usb sticks no help, any help would be great thanks.
