pc crash/freeze

[limber grove]

Hi there, i'm looking for some help of peaple that could know what's causing my problem. Sometimes when i launch a game, my computer just freeze, screen too, sound too (still "playing but blocked) and i have to restart my computer with the button.

I have an intel i7 12700kf
Intel arc770
16gb ram 3200 (with xmp profile on)
Asrock extreme z690
An hdd and sdd (old hdd starting to be tired my windows 11 is on that one)

The temp is always good so do you think the prob could be from my hdd ? Ram with xmp on ? Or arc ? (Every drivers have the lastest versions)

[limber grove]

[amber crow]

Have you tried DDU? Do you have errors in Event Viewer timestamped from when it happens?

It could be multiple things, so there's a lot of steps to check

Check the SMART from the HDD and screenshot it

[limber grove]

there for the smart

didnt try the ddu, don't know how to do
BTW after relaunching my pc (just had a crash) a auto repair on my external hdd launch, i eject him, will see if i still got crash

and in my event viewer, nothing, just the crit when I shut down my pc with the big button, and the last advert before that Connection ''error: Le nom distant n'a pas pu être résolu: 'api.intelsa.intel.com'
Network available: True'' literally : distant name couldn t be resovled

[limber grove]

Not my external hdd, seems like i have less perf when i'm gonna got the crash, some mini freeze, but in my task manager i couldnt see what would be the prob

that advert is still here 'api.intelsa.intel.com' could be the prob, gonna look on internet some info

[limber grove]

ok assuming that could be potential drivers not uninstalled that will cause the prob, i will do a ddu soon (i've got some prob to install 4091, and seems that it started from when i start tried to install it)

[limber grove]

Aaaaaand ddu, but, poof, that did it again, will try the beta drivers i think, cause everyrhing is on the best point to not make me crah (no xmp/oc, good temp, i've tried stress test
and nothing happend)

[limber grove]

i deactivated the isa intel, that was the last advert in my event viewer before the crash, but i still got one (i got them on some game only, here monster hunter world, ff14 online) i will try 4123 drivers, i hope this will repair my crashs : /

[limber grove]

ok some news here, i told that when i got a crash, after rebooting, i'm sure that i will NOT get a new one, till a last crash, i've just put my pc on sleep mode, so not off, aaaaaand, i didn't crash on the games, so i think it's an option/prog that launch after my pc was off, but don't relaunch after i force close my pc and reboot, will see if my bios is up to date

[rapid light]

Also make sure CSM is disabled in the bios

[limber grove]

Yeah, i put my bios up to date (reinstall windows cause of errors T_T) and csm is correctly disable, i will see if i still get those crash even with that fresh new windows and bios (and sdd cleared)

[limber grove]

with new windows (and fresh sdd) i don't have those crash anymore, i've reinstall visual++ (and vs code + others thing necessary to prog on unity) + directx for some games, hope that will not come back with those dl

[limber grove]

I GOT IT AGAIN aaaaaaaah

[limber grove]

Does directx could make those crashs ?

[limber grove]

Or a psu prob maybe ? 650w 80b+ is that enough ? (none of my comp are oc)

[limber grove]

Idk if this can be related. Sometimes, when i want to play to monster hunter world, u have an error (failed to launch direct x 11) i have to up to date the pilote of my arc directly in the peripherical manager (and choose the intel arc driver not the windows one or course) (i have 2 intel arc driivers showed but i have only the last one install till i format my ssd)

[sinful mountain]

Its a relatively new card issues are bound to happen when the release is without proper and updated drivers

[rapid light]

Do you have one or two PCIe power cables going from the psu to the gpu?

[limber grove]

One pcie power cable

I will put the W limit to 200 to see if that's change something

[rapid light]

Run two PCIe power cables to it. A single cable is pulling power from the same rail.

[limber grove]

k will try thx

[rapid light]

Also what power supply do you have?

[limber grove]

Cooler master 650w 80+ bronze

intel i7 12700kf
asrock z690 extreme
With this and the arc

[rapid light]

yea try running the second cable to it

[limber grove]

I will give news the next few days/week (i just hope this is not a faulty card)

[limber grove]

i have a bug that i was not talking cause i forgot about him, but sometimes too, when i turn on my computer, he could do 5 short bips, restarting auto, and booting like nothing happened (the bios may be got back to default) i have a UEFI bios with my asrock. sigh... i'm praying to get evrything working correctly soon

[sleek wind]

Five short beeps mean there has been a processor error. A damaged expansion card, the CPU, or the motherboard could be prompting this AMI beep code. Start by reseating the CPU. If that doesn't work, try reseating any expansion cards.

[limber grove]

I have already checked the cpu, nothing look broken, expansion cards ? What do you mean (i'm french)

[sleek wind]

everything that connects to the motherboard like a gpu

[limber grove]

Haaa k, well i assume it's the gpu

[rapid light]

try reseating the GPU and see if that fixes it. Sometimes its not installed all the way and causes issues

[limber grove]

k

[limber grove]

I just pluged the second pcie cable, aaaand, till all those years, i just saw that the pcie cable that was already here wasnt perfectly pressed, i just hope that was this missplay of mine that was causing instability, will test tomorrow

[limber grove]

Hey back again !! I'm still getting it :'). I played mh full high, without reducing the watt limit, and got it, gonna let the watt at 190 to see if i still got it (if not then my psu should be too old to give the best)

[limber grove]

Ok a second one really fast :/ humm

(I have desactivated the fast launch of windows cause i think ha was the bip boot problem)

And with the limiter, does the fast boost of windows could be related to the crash (i mean in the windows options) (i normally don't get a new one after a first, but now i am, and exaclty when i just get off the fast boot of windows)

[limber grove]

same, don't know if this can be related (i"m searching for each possible instability cause)
I have a screen (my principal) that is connected with a hmdi/display port cable (hdmi for the screen, and display port at the card)

[vernal moss]

try bios update... activate xmp, and rebar... set dram voltage to 1.36 or 1.37 volt (just to be safe)

then ddu + driver installation

if the problem is still there reduce max power step by step or get a new psu

[limber grove]

I will try to reduce it, but even at 190w limit i crashed, so i don't think thar will change

[limber grove]

I will try too to put only the hmdi to see if that's change something (it's strange but meh, trying)

[limber grove]

Do you think doing an occt psu check could be a good tool to see if this is the prob ?

[limber grove]

i did a power test on occt for 10 minutes (didn't want to burn my cpu) cpu + gpu at 100% after gpu reached the 90°C he just lost power to like 180w (cause of the 90C i think), but nothing alarming (fans was at 1200 max normal ?) but it think my psu is fine, and the gpu too, so it's probably an option in the bios or my windows or drivers (if you agree with that)

[limber grove]

(i did a session on mh, didn't crashed, i've lowered graph settings, and put my screen on hdmi (without displayport)

(i dk if it's the hdmi or the settings that make it not crashing, but i will vote for the settings one, and in that case, why)

[limber grove]

I will do a stress test on gpu only (cause power test wasnt using the graphic memory at all)

[limber grove]

doing a vram test, 10 minutes passed, nothing wrong, gonna let it a bit more

[limber grove]

@buoyant veldt

(I have the lastest drivers and done the ddu)

[buoyant veldt]

ok thats an old and big thread let me check it out

what are your specs

i need your mobo actually not the full specs

[limber grove]

I've tried a lot of thing, everything is report to see if i've got results

Btw, today i don't get the strange screen anymore, just loosing signal

Intel arc a770 le
I7 12700kf
16gb ram
Asrock extreme z690
Hdd
Sdd
Psu : cooler master 650 b+
Basic screen 60hz

Ho

[buoyant veldt]

have you updated your bios

[limber grove]

Yup, not the last last one, but just before the last (the last got my windows corrupt ^^')

So i'm a bit scared of the last

[buoyant veldt]

is virtualisation disabled?

csm off?

rebar on?

you know the drill

[limber grove]

Above 4 enable
Cam (rebar) enable
Every aspm disable
Virtualisation disable

Csm disable

[buoyant veldt]

ok so these are not the issue

i read that you tried reseating the gpu?

[limber grove]

Yup, if you hear reseating it by taking it off and back in of the mobo, yup

[buoyant veldt]

thats what i mean yes

are you sure your cpu or ram are not the issue

?

[limber grove]

I've got the ram with an other cpu and gpu without problem, for the cpu, can't told, i've took it back to look at the pins, everyone was perfect

[buoyant veldt]

you'll have to stress test the whole gpu so that we can be sure thats the problem

actually

open cmd as admin first

and type

sfc /scannow

[limber grove]

Wich gpu test if good for this, furmark ?

Gonna do the scan

[buoyant veldt]

yea do the scan first and then try furmark

what im afraid is that this could be the hdd

[limber grove]

What will be the options in furmark ? (And how many times, i've already done something but not for so long)
If it's the hdd it's ok now, i've put windows on my sdd, and i don't have much important thing (thanks to the corruption x) )

I will just buy a new one

The scan didnt find anything

[buoyant veldt]

ok good

[buoyant veldt]

just max it out

we need to see if its the load on the gpu

if it is it COULD be the psu

or a card that needs to be rmaed

[limber grove]

Actually launched

[buoyant veldt]

what do you mean

it couldn't launch before?

[limber grove]

I mean, i start the gpu stress test*

[buoyant veldt]

ah ok

[limber grove]

X)

[buoyant veldt]

run it for half an hour

[limber grove]

K

[buoyant veldt]

if it doesn crash try movid it to the hdd and run it there

moving*

[limber grove]

after 40 minutes, no crash, moving it into hdd so, gonna try

it was already into the hdd LUL

[buoyant veldt]

then it could be the game specifically??

weird

the card should be good

try running your game again

[limber grove]

it happens on others games than mh world, and rarely, but still, sometimes while i'm not playing and just hanging on youtube (on opera)

[buoyant veldt]

wait

try disabling hardware acceleration

on windows.

and opera

[limber grove]

lately, i got the crash last night on drl

ook

[buoyant veldt]

disable hardware accel on discord windows and opera

[limber grove]

done on discord and opera, looking to doing it on windows

[buoyant veldt]

good

[limber grove]

don't see it on windows (idk if it's common) but i will keep you informed if i crash on opera/while i'm not on a game

[buoyant veldt]

ok

[rapid light]

also try disabling fast boot in windows and the bios

[limber grove]

ty for the tips (i've already done it, resolving a prob doing this)

[limber grove]

@buoyant veldt i've got a new crash (i forgot to open arc control, so maybe it's this)

But in the same time it was a bit different here, my screen got black a few second, with still everything working, and came back like nothing, but after like 10 seconds, i've got the usual crash/freeze

I'm gonna try to re enable virtualization idk (how do i get the option of the hardware acceleration in my windows ? I don't have it actually)

[buoyant veldt]

weird sounds like a driver bu

g

[limber grove]

Maybe when one bug will be repair, that could be too, idk, i have it with every drivers actually 🥲 but i'm hoping

Still gonna try things, that could be only one option that is making this

[buoyant veldt]

does it happen frequently or once in a while

[fading creek]

in case you have the old arc panel, alt+M releases it
cannot find anything about it anymore, but the "(still "playing but blocked)" sounds like that issue

but i have a similair situation 11700k/A770le/psu:730W
im wondering if the power supply is just on the limit
1:load heavy "dyson sphere program" savegame
2: A770 max 190w 80c, 100% usage(as expected)
3: many browser windows spread over 2 screens(not just tabs) slowly building requirement(cpu usage)
4: multiple discord streams open
and then at some point either everything goes black, or freezes at the last known picture

im wondering if ripples in powerdrawing could trip cpu crash.... but thats speculation territory
im hoping i can fix this issue by buying an 1000W psu
(also had issues before the a770, just diffrent, with gtx970, setting cpu power to 20% in power managment was the temporary fix there)

[limber grove]

It can be 1 per day (playing the evening only) if i play big games, or, 1 per weeks if i'm not playing and just doing random things

It's actually pretty random

After a crash, i think (not sure) that i'm getting it lesser until i turn off my computer and came back the next day

[limber grove]

K got one on the desk/youtube

They are stranger now x)
Like in the game last time

I've lost the signal, still hearing things correctly and could do some things
And then, the sound got freeze, and the whole system too

I've again forgot to open my intel arc control

[limber grove]

One more in a game 🥲 i have deactivated virtualization to see if i got changes
I will try beta drivers too

[fading creek]

try a different hdmi cable if your able
heat expansion could cause rapid connect/disconnect with poor/old cables
or displayport if possible

my A770 hangs in there with paperclips instead of screws, becouse of bad match in the case, it was pulling the back end out with screws

[limber grove]

will try a different hdmi, and, i will look to the screws cause in fact they are not matching perfectly, so that will be the cause

it's not a big distance, but could be sufficient to pull a bit the card and making some instability

[fading creek]

when applying the paperclip treatment(having no screws), be extra carefull when moving the machine

[limber grove]

yeah, is it dangerous to put my tower standing (my tower is build to stand straight) without screws ? if yes i will try to lie her down

[fading creek]

well.... when i tried to put screws in, it was actualy slightly difficult, and pulled on my card too much

after you put the card in its slot, if the screws have no trouble going in, then it should be good with screws

if you want to try to make your pcie connection "as happy as possible", then laying it down to let the card be free of screws is an option

[limber grove]

Ho alright, i will still try for a session but i'm not forcing that much, so maybe not the prob here

[fading creek]

i did have issues with hdmi to dvi-d cables on intel connections, both onbourd and on a770
nvidia card never had issues with hdmi to dvi-d cables
(just to add information)

but thats for potentional hardware mis-connections
if that has been ruled out, then im looking back at power supply

[limber grove]

That was not the screws, got it again with the black screen, caming back, and crash

(Btw ihave alt f4 the game between the black screen and the crash, didnt see anything weird in the alt+o arc control, so i launched back the game started to play and then black screen and crash)

[buoyant veldt]

it could be a faulty card or psu

[limber grove]

🥲 i don't want this option

[buoyant veldt]

you can rma the card and request another if the issue persists its the psu

i seriously wish its neither

[fading creek]

what i did to gain stability:
1.set gpu power to 150w
2. turn cpu-turbo off
3. in power managment set max cpu power to 20% to 50%

if no problems at that point, then im thinking its power(psu)

[limber grove]

I'm caming back ! with some good bad news !!!!

i've got an error in the report event

WHEA-Logger 1
''Une erreur matérielle irrécupérable s’est produite. Un enregistrement décrivant cette erreur est inclus dans la section des données de cet événement.''

(google trad : An unrecoverable hardware error has occurred. A record describing this error is included in the data section of this event.)

Haha, now, what is the prob, trying to find the record

[buoyant veldt]

what... are the good news

[limber grove]

i know what to search x) that's the good news, at least i'm not in the dark anymore

[buoyant veldt]

thats a good start

[limber grove]

ok, re doing everything from the start, crystal disk told me that everything is good, so i think i can put appart my ssd and hdd

[buoyant veldt]

you have to check your psu

run a memtest

too while you are at it

[limber grove]

gonna check, i already test my psu technically with occt at full, but, will retry then (occt is good to test it right ?)

[buoyant veldt]

yea

[robust sable]

the problem was on gpu

[limber grove]

Yes and no, i don't have any clue to tell that it's my gpu at 100%

So i'm trying to know, now that i'm sure it's hardware and not software

(That's why i'm in the community supp, and not the arc one)

[buoyant veldt]

you did good dont worry

i guess you could stress test the gpu

[limber grove]

Don't seems to be the rams

Gonna wait till the end

[buoyant veldt]

should be fine honestly

so it comes down to the gpu? you could rma it

[limber grove]

Welp, i've done the occt power test, and got it, this black screen

So, i can point my psu, weirdly, look like pretty randomnly doing shit (maybe) shall switch her fast

[buoyant veldt]

whats your psu

[limber grove]

A 650w b+ cooler master, she have 5years

I don't get it x)
T_T

I think i should buy a new psu, i'm not loosing anything (well money of course)
But if it's not this, then i should rma the graphics card

Gonna still try the vram test at 90%, and if i got nothing i'm gonna get an old 1050 ti to do some more test

[buoyant veldt]

i guess

coolermaster is quite known for the medium ish quality of their psus

[limber grove]

ok so, i've done some more test, after a psu test so, and getting the crash, in the same day i've done a benchmark test with heaven, after 30 minutes, i've got nothing weird, but, cause my problem is weirder, i just launch the computer, launch heaven, the test, and boom, instant crash

gonna put an old graphic card, sadly she don't take as much power as the arc (old 1050 ti) but i think it's the graphic card so

[buoyant veldt]

@amber crow or @rapid light will have to confirm but this is more than likely the psu

[limber grove]

hoo ?

[buoyant veldt]

im not sure tho

your case is quite weird

[limber grove]

🥲 i've got the weirdest bugs everytime, i'm a professional

to resume for them ;

done the psu test and first heaven benchmark test was yesterday, while i've got the power test crash, and nothing with
heaven benchmark

Today i've instantly launch heaven, and instantly got the crash

[amber crow]

In court the next while so can't help until later

[buoyant veldt]

sorry

[buoyant veldt]

try with the 1050 ti i guess

[amber crow]

Does it crash when gpu is loaded? What about a cpu only load? Or a combined load

[limber grove]

can happen on youtube, no need a big load, never happen on launching the game, or app, it's always when i'm already on

i've done a stress test on the cpu, and a diagnostic, with the intel tool, and got nothing wrong

[buoyant veldt]

all you can do is rma the card and hope for the best

[fading creek]

i wonder if it still crashes if you set the power limit in arc control to 95W