AB350m Pro4 couse graphics card issue?

Message Topic Search Topic Options Post Reply Create New Topic Printable Version Translate Topic

   Mine just crashed again too. Played some Star Trek Online for a couple solid hours, zero issues, tabbed in and out a couple times. Then, after 4-5 seconds at desktop right after exiting the game, display cut out and system froze.

Nothing abnormal on the temp readings, my voltages looked okay at a glance.

I'm still thinking of pulling the card and testing a different one, but I'm honestly leaning more toward just calling it a defective board. I'll probably try running OCCT later just to try to squeeze more information out of it.

Author	Message Topic Search Topic Options Post Reply Create New Topic Printable Version Translate Topic
Niccador Members Profile Send Private Message Find Members Posts Add to Buddy List Newbie Joined: 31 May 2018 Status: Offline Points: 44	Post Options Post Reply Quote Niccador Report Post Thanks(0) Quote Reply Posted: 01 Mar 2019 at 5:57am
	Mine just crashed again too. Played some Star Trek Online for a couple solid hours, zero issues, tabbed in and out a couple times. Then, after 4-5 seconds at desktop right after exiting the game, display cut out and system froze. Nothing abnormal on the temp readings, my voltages looked okay at a glance. I'm still thinking of pulling the card and testing a different one, but I'm honestly leaning more toward just calling it a defective board. I'll probably try running OCCT later just to try to squeeze more information out of it.

Niccador Members Profile Send Private Message Find Members Posts Add to Buddy List Newbie Joined: 31 May 2018 Status: Offline Points: 44	Post Options Post Reply Quote Niccador Report Post Thanks(0) Quote Reply Posted: 01 Mar 2019 at 12:46pm
	I'd like to apologize to mati998 for hijacking this thread with my similar issues. Here's hoping we can both get this sorted though. Crashed yet again. This time near the 30min mark on an OCCT GPU test...as before, locked up with no display. (This was following a 2hr OCCT CPU test.) I'm not even remotely convinced that the GPU is actually at fault yet. It performs flawlessly, right up to the crash. Temps, voltages, etc are all well within limits...no artifacts or other graphical hiccups. It simply works like a champ until the whole system goes down. The oddly high temp readings on sensors which are labeled (in some software) as Aux and/or VRM do not seem to be accurate. Unless the VRM heatsinks are not making any sort of proper contact, they don't appear to be remotely warm. Not all apps pick up these particular sensors, so I'm calling this a red herring for now. I've been reluctant to swap out the video card, mainly because I don't have an identical card to swap in. Changing out the card for one with drastically different power and/or bandwidth requirements might appear to be a fix at first, when in reality it could just be delaying me from finding an issue with the PCIE or power supply.... Which leads me to my next target of interest: the power supply. My PSU is a Corsair CX550M, which should be ample for this build...in theory it should cap out at about 400W under full CPU/GPU load. I'm trying to sort out how best to test it, although I think just swapping it out is probably the only way. I've read of some faulty power supplies causing all manner of bizarre behavior (see recent reviews on NewEgg for the Corsair CX550M), so I'm going to give this a shot. I'll update this tomorrow after testing.

mati998 Members Profile Send Private Message Find Members Posts Add to Buddy List Newbie Joined: 28 Feb 2019 Status: Offline Points: 13	Post Options Post Reply Quote mati998 Report Post Thanks(0) Quote Reply Posted: 02 Mar 2019 at 5:03am
	Niccador, it's OK that you write in this thread, because we may have the same cause of problems. Fortunately, my motherboard is still under warranty, power supply and processor too (but it remained 1 month) but I do not have time to look for the source of the problem and wait for repair :(

Niccador Members Profile Send Private Message Find Members Posts Add to Buddy List Newbie Joined: 31 May 2018 Status: Offline Points: 44	Post Options Post Reply Quote Niccador Report Post Thanks(0) Quote Reply Posted: 02 Mar 2019 at 9:40am
	Once again, the computer mocks my troubleshooting efforts. So after last night's crash, I swapped power supplies with an old Nspire 500W I keep around for stuff. Ran about 3 hours of tests with OCCT, with zero issues. Got my hopes up. Swapped the original power supply back in to confirm crashes... And it's been running for about 3 hours now, through the same or either tests, with no issues. Yet. Truly infuriating. Next, I'm going to just game tonight/this weekend with both power supplies until something breaks. Of I can't pin down a problem with the psu's, I'll try switching out the graphics card with an older one and a newer one, and see what happens. Will update tomorrow.

Niccador Members Profile Send Private Message Find Members Posts Add to Buddy List Newbie Joined: 31 May 2018 Status: Offline Points: 44	Post Options Post Reply Quote Niccador Report Post Thanks(0) Quote Reply Posted: 04 Mar 2019 at 7:42am
	Okay, so just a brief recap of everything I've done thus far. Hours and hours of memory testing with both Windows memory Diagnostic and MemTest86+, individual sticks and both sticks. Results: zero errors found. At least 8+ hours worth (total) of OCCT CPU testing, also no errors found. Tested three different Radeon graphics cards, multiple driver versions. Also tested with two different power supplies. Results: no change in crash frequency. Tested multiple audio driver versions, no change in crash frequency. No notable events appearing in Event Viewer during or near any of these crashes. The only observations I've managed to really make are these: 1) Audio crackles or distorts just before a crash, BUT, audio continues to play normally after display has crashed. This, to me, makes it appear like some manner of driver problem between audio and video, perhaps. But testing would seem to rule out drivers for both since it occurs with multiple versions known to be previously working. Likewise, it's not a defective video card, since it occurs with multiple known working cards. Could still be audio hardware, since I can't really change that out. 2) These crashes all started following a major Windows update to 1809 at the beginning of February. The only thing I feel I have yet to do is to just wipe and clean install Windows. Which I guess is my next step. It'll take me a couple days.

mochination Members Profile Send Private Message Find Members Posts Add to Buddy List Newbie Joined: 04 Mar 2019 Status: Offline Points: 6	Post Options Post Reply Quote mochination Report Post Thanks(0) Quote Reply Posted: 04 Mar 2019 at 8:28pm
	I also have problems with shutting down the PC where it won't go completely. http://forum.asrock.com/forum_posts.asp?TID=7316&PN=1&title=asrock-ab350-pro4-motherboard-problems ^^ These people also seem to have similar problems, at least with not being able to shut down the computer properly.

Niccador Members Profile Send Private Message Find Members Posts Add to Buddy List Newbie Joined: 31 May 2018 Status: Offline Points: 44	Post Options Post Reply Quote Niccador Report Post Thanks(0) Quote Reply Posted: 05 Mar 2019 at 5:26am
	So, I cleared CMOS yet again, and did a clean install of Windows 10 (109, build 17763.316). In the lead-up to this, while backing up my necessary data, the crashing intensified, and started happening every 5-10 minutes, or less. Interestingly, I was able to get it into Safe Mode, which didn't crash (at least for the hour or so it took me to wrap up). This may be purely coincidence though, since this thing has gone several hours between crashes before. After the clean install, everything seemed alright for a couple hours. Then it crashed same as before. And it continued crashing every few minutes...sometimes almost as soon as it got to desktop. Ran MemTest86+ again for a few long passes. Still no errors reported. Going to double check PSU using a third new supply, just to be certain. If that still fails, I'm going to try another clean install of Windows...this time rolling back to 1803 17134.619 and disabling further updates.

mati998 Members Profile Send Private Message Find Members Posts Add to Buddy List Newbie Joined: 28 Feb 2019 Status: Offline Points: 13	Post Options Post Reply Quote mati998 Report Post Thanks(0) Quote Reply Posted: 05 Mar 2019 at 7:04am
	I found something interesting. I accidentally kicked into the computer case, then it crashed. So I reboot computer and (I know it is dangerous and maybe quite stupid) I touched (very carefully) components. When I touched one of RAM stick then audio sound began to crackle,but still played and the screen crashed.I cleaned the RAM slots using compressed air and contacts on RAM sticks using isopropanol and currently works good, but I do not know if this is the cause of all problems, I will check

Niccador Members Profile Send Private Message Find Members Posts Add to Buddy List Newbie Joined: 31 May 2018 Status: Offline Points: 44	Post Options Post Reply Quote Niccador Report Post Thanks(0) Quote Reply Posted: 05 Mar 2019 at 7:13am
	Interesting, indeed, though I wouldn't go touching bare RAM while the system's running myself. Only reason I even considering reaching for mine is the ample heat spreader they have. Anyway. It seems interesting in your case, but also possibly coincidental? I would think if your RAM or slots had been dirty or loose, something would have shown up on a memory test. But who knows.

Niccador Members Profile Send Private Message Find Members Posts Add to Buddy List Newbie Joined: 31 May 2018 Status: Offline Points: 44	Post Options Post Reply Quote Niccador Report Post Thanks(0) Quote Reply Posted: 05 Mar 2019 at 12:22pm
	Okay. I'm done. Retested with a third power supply, no change in behavior. Ran additional memory tests, no change. Tried reinstalling both the 1809 build, and the previously working 1803 build of Windows, still with no change. I am at a complete loss. At this point, the only two parts I haven't swapped out are the CPU and the Motherboard. Unfortunately, the CPU appears to test just fine in OCCT, which leaves me with only one option: buy a different board. I don't want to talk trash on ASRock like some, but after what I've struggled with for the past month, and what I've read here and elsewhere, this was just a bad product. ASRock can make good boards. I've got a X58 Extreme that seems ancient now, but is still running like a beast, with a Xeon X5675. That was clearly a great product. The AB350M Pro4? Not so much.