the 8xxx serious was more power consuming than the 9xxx series, but I doubt this is the cause. The AC converter is sufficiently sized, and the board shall have no trouble either.

Try to underclock the GPU (nvidia-settings should now allow you to do that and you don't need much for desktop usage) and see whether that stabilizes anything.

So, the b*stard graphics kept freezing again and again so I couldn't work today and everything I researched (dmesg | NVRM -> errors xid) pointed to your first hunch about a dying GPU.

Got a GeForce 8400 GS V8402GL26 card on ebay for a few bucks from a guy nearby enabling me to pick it up and install it straightaway. I force-reinstalled the nvidia drivers and created a new xorg.conf and everything works like a charm, even the unpleasant flirring is gone.

Did I say everything? Well, when I start glxgears or firefox the screen goes black for a second and comes back - glxgears and firefox working all fine. Any ideas about that? The short black-out is probably not supposed to be there??

Probably not
Does it also happen with suspended compositor? (SHIFT+Alt+F12)
Does the cursor also disappear (maybe switch to a white theme for easy spotting)

So, after a few hours of running my machine froze again giving me similar dmesg xid errors. "Froze" is really not the proper word, because when I touched the (passively cooled) GF8400 card and the board I almost burned my fingers! The ebay seller (of the GF9500) told me that the GF8200e motherboard has massive heat problems and the occasional incidential failing of the USB ports can also be explained by that.

Reverted to the onboard GF8200 graphics card and nouveau (again no chance with nvidia), and either the messing around with config files during the last days or the switching off of all desktop effects made the annoying lagging disappear that I've encountered before. Well, at least so far. I'll see how it goes and probably need a new motherboard at some point. The nouveau graphics are just not quite as pretty.. but I'll survive as long as I can get my work done.

knightsu wrote:Found this last 4 lines of dmesg after i got "desktop effects were restarted due to a graphics reset" :
[ 57.831047] kactivitymanage[1427]: segfault at 7f47c3960d10 ip 00007f47c3ba769a sp 00007ffd4123f638 error 4 in libQt5Sql.so.5.5.1[7f47c3b92000+3f000]
[ 387.468677] perf interrupt took too long (2514 > 2500), lowering kernel.perf_event_max_sample_rate to 50000

dmesg | grep -iC 3 "NVRM"

Code: Select all

[    1.756672] hub 2-0:1.0: 2 ports detected
[    1.761717] vgaarb: device changed decodes: PCI:0000:01:00.0,olddecodes=io+mem,decodes=none:owns=io+mem
[    1.762252] [drm] Initialized nvidia-drm 0.0.0 20150116 for 0000:01:00.0 on minor 0
[    1.762259] NVRM: loading NVIDIA UNIX x86_64 Kernel Module  340.93  Wed Aug 19 16:49:15 PDT 2015
[    1.764327] nvidia_uvm: Loaded the UVM driver, major device number 250
[    1.780223] hub 3-0:1.0: USB hub found
[    1.780566] hub 3-0:1.0: 2 ports detected


journalctl | tail - http://dpaste.com/3TCRVZ5

Gotta ask, before Kwin restarts on you, did you switch to an activity that was "off" from startup?

Reason why I was asking because I'm also experiencing that same problem with my desktop running on GTX 1060. When I first saw that post, it made me realize that it usually happens when I used an activity that was "off" from cold boot. And it doesn't seem to happen when I switch back and forth to an activity that was "on" from boot.

Gonna have to check back in a few weeks to be sure, but I'll try to report back if it restarts even without switching to an activity that's "off".