Screen "resets" occasionally
I really don't know how else to explain it with that few words, but basically, my external monitor "turns off" for maybe 5 seconds, then comes back on like nothing happened. It does the same thing when the refresh rate changes, but I didn't trigger a refresh. The screen is not getting disabled, the application bar doesn't move, games don't switch screens during the "blackout". The audio doesn't static, and no system sounds are triggered to indicate that something happened.
https://paste.centos.org/view/3d903ba3
22 Replies
speaking of cursed, @tulip this has been happening for a few days now, so ig it's time to say something š
to be honest
how come you run into all this
are you cursed by an evil witch
or mage
the only thing I can really think of is that I noticed something like this happening when plugging in my webcam randomly into the #3 slot on my FW 16? So like, maybe the USB is resetting, which for some damn reason is making the... idk, screen or refresh rate do a thing?
I think I definately called the wrong duck a mean name as a child
unless something's in those bazzite logs though (fpaste), this one might be insane to try and debug, because it's truly random. Happens whether I'm playing games or just watching youtube, and only maybe once a day or so
that fpaste is also like 2 min after the event at most, so I'd expect the issue to be near the bottom if its there at all
where can I find KDE logs? Honestly, this sounds more like a KDE thing than anything else
this also seems to have gone away with latest bazzite
nope, it's back
oh well
I'm gonna lower the refresh rate to see if it does anything
@tulip do you know if the dmesg logs would have anything that would point to what's causing this? It's still happening. I was hoping it was just a cable thing, so I unplugged and re-plugged that, but it didn't fix it. I can't even tell if my PC recognizes its happening, since the monitor isn't actually getting disabled.
I've read through them multiple times, but I don't notice anything all that suspicious
actually, from what I can tell, the dmesg logs don't have anything during the time of the monitor reset
I've gone ahead and factory reset my monitor just in case it's somehow that
ok so so far, the factory reset worked lmao, so it was the damn monitor
ope nvm, it happened
darn
cc @Raevenant
I don't think this is the same issue unfortunately š
I'm trying out a new cable rn, maybe it's that
saw some stuff online that the stock cables are aweful
so here's the kicker
so far (it's been about a day) switching the cable to one that can only handle 100hz has fixed this issue
if I go a week without issue again, I'm going to get a new cable that can handle the 1440@165hz and see if that one also works
however it also looks like there was an update yesterday so I'll try putting my old cable back first to make sure I can even reproduce on latest bazzite
ok here's something interesting. It did happen again, but only the instsant I messed with a USB cable
which makes no sense at all, because my monitor isn't plugged in via USB, it's plugged into the dedicated GPU slot on the back
it's reproducible though
@tulip what do you even attempt to make of this
https://paste.centos.org/view/35fe5efd
captured
There's a bunch of lines in there about USB port 4 being problematic, which is interesting, because I don't have anything plugged into that slot. I do have an HDMI expansion card in there, so maybe there's a defect in that card and it's messing with the display processes?
I did more testing. This affects slots 3, 5, and 6 but not 1, 2, and 4
I've opened a support ticket with Framework, this points to a hardware failure
@Kyle Gospo the FW support engineer mentioned this might be intended behavior, and I have a hard time imagining that. Would you mind please confirming that this doesn't happen on your devices?
1. Monitor is plugged into the dGPU
2. Plug any USB device into slot 3, 5, or 6.
The monitor should turn off and back on again
I have a hard time imagining this is intended, but if it is, then it shouldn't be just my device doing it
whats up!
does this shit happen on windows?
this 1 billion percent doesnt seem to be indented behavior LOL
No idea, don't have a drive to test š I can live-boot Fedora to see if it's at least Bazzite
Or roll back to some older version
Be careful with cables. They will flat out lie about VESA certification on Amazon etc. This is the definitive source: https://vesa.org/displayport-developer/certified-components/ but it's search leaves a lot to be desired and they have certifications listed that aren't for sale. I still haven't found a 10ft cable that is certified.
admin662
VESA - Interface Standards for The Display Industry
Certified Components - VESA - Interface Standards for The Display I...
DisplayPort Certified Source AMD 9th Gen Ryzen Processor (STX, Type C) Family 12h Processors, Trinity APU, Radeon HD 7900XT, 6950, 6850, 6570, 6450, 5870, 5770, 5450, 4750, 4550, 3650, 3450, 3200, and FireGL V7700, Radeon R9 255, R9 350 Family 14h Processors Carrizo class APUs Mullins class APUs, including AMD A-series and Eā series Navi [ā¦]
I see similar issue as you describe, with 49inch super ultrawide at 5120x1440p at 240hz. If I drop the refresh rate (and there for the amount of data being pushed) issue goes away.
I can fairly reliabley cause it to happen, taking screenshots/selections makes the screen go black, usually it comes back, sometimes it says can't find signal. Switching input back and forth to the monitor fixes.
I need to test more with a shorter cable that's officially certified.
Oddly it almost never happens when playing a game.
But screen shot tool and full screen YouTube usually trigger it.
Yeah if the USB thing doesn't work, I'll check out a new cable, but I think the usb thing might actually be the culprit in my particular case, since it's not a software trigger but a hardware one (which admittedly does interact via software, so idk)
That solves the reproducible USB one, but I still don't know what was causing the random blackouts. I'll keep this thread open until I know it's actually gone. I have a feeling that it has something to do with the DP port, but I'll keep updated
This is the dmesg logs for the USB event. Beats me what's actually happening
oh dang wait there's a DRM error in there, go figure lmao
I wonder if usb-1-2-port4 is the monitor plugged into the dGPU USB-C socket
In fact, that's exactly what it is
so the bad cable is on the USB bus, and for whatever reason, plugging the 8bitdo controller into the non-display USB bus was causing it to freak, making the monitor's USB port do a power cycle
this is the most absurd problem i've seen yet from you
@Raevenant the witches curse really outdid itself on this one
Cables are the worst. I'm going to build a new desk eventually just so I can mount my tower underneath and not deal with it any more.
Standards used to mean something. Now they are just marketing and you can lie.
I tried two 10ft cables, the 3 and 6 foot versions are certified for both, but not the 10. They still say it can handle the data throughput, but the VESA certification is missing -_-.
Just gonna point this out, it's been 2 days, and so far it's gone
I went back to the stock cable, and switched back to DP 1
freaking weird lmao
Take the win I say lol