Frequent HARD FREEZES after 1-2 hours of usage
Printed From: ASRock.com
Category: Technical Support
Forum Name: AMD Motherboards
Forum Description: Question about ASRock AMD motherboards
URL: https://forum.asrock.com/forum_posts.asp?TID=7036
Printed Date: 18 Oct 2024 at 6:44pm
Software Version: Web Wiz Forums 12.04 - http://www.webwizforums.com
Topic: Frequent HARD FREEZES after 1-2 hours of usage
Posted By: Mr Bungle
Subject: Frequent HARD FREEZES after 1-2 hours of usage
Date Posted: 22 Dec 2017 at 9:17am
|
Hi folks... Decided to build my first PC for my dad who wanted a budget gaming PC to use for himself and my son when he visits. I'd added components to my PC at home and figured I could handle it. I went with PC Gamer's budget cyber-monday build ( http://www.pcgamer.com/heres-a-budget-gaming-pc-built-with-cyber-monday-deals-for-522/" rel="nofollow - ) It included: CPU - Ryzen 3 1300X GPU - GTX 1060 3GB MOBO - ASRock AB350M Pro 4 RAM - GeIL Evo Potenza 2x4GB DDR4-2400 HD - Crucial BX300 240GB SSD PSU - Rosewill Valens 80+ Gold 500W So after installation, I put a fresh copy of Windows 10 on the SSD, and completed all updates (nvidia drivers, windows 10 updates, any basic hardware updates I could complete). I added Steam, VLC, AVAST, and Origin with no issues. I set Steam to install a game overnight (slow internet speeds) and when I woke up, the computer was frozen about midway through the install. Fans were still working but complete freeze. I reboot the next day and after finishing the install, I start the game and another freeze. This time, upon reboot, the computer locks up again within 5 minutes or so. If I let it sit for a few hours, I can get it to act normal for another 1-2 hours. (continued) |
Replies:
Posted By: Mr Bungle
Date Posted: 22 Dec 2017 at 9:18am
|
Now I have limited technical expertise, but here is what I've tried: I first searched forums and have found similar issues, but most of them don't seem to have been solved. Some links: http://www.tomshardware.com/answers/id-3535484/asrock-ab350m-pro-freezing.html" rel="nofollow - http://www.tomshardware.com/answers/id-3594663/asrock-ab350m-pro4-random-freezing-load.html" rel="nofollow - I removed the RAM and reinserted it. I've monitored the temps using CPUID HW monitor (nothing crazy even right before crashes) I've flash updated the UEFI I've made settings changes as suggested by this link ( https://www.drivereasy.com/knowledge/solved-windows-10-freezes-locks-up-randomly-easily/" rel="nofollow - ) the link includes things like - running memory check - reset virtual memory - run disk check - run system file checker - disable c-states Just ran MemTest86 and there were no errors. Took out GPU and put it back on. I'm usually pretty good at hunting out solutions to problems I've had, but this is the first time I've had to write a forum post. Any help or suggestions you guys could offer would be much appreciated. I'm hoping to fix this by Christmas for my dad. These frequent freezes are driving me crazy! |
Posted By: zlobster
Date Posted: 22 Dec 2017 at 9:56pm
|
After every HW removal/update/re-seating you MUST clear the CMOS. Clearing the CMOS forces the mobo to enumerate the installed hardware once again. Simply removing and then putting the RAM again does absolutely nothing. You could check Win 10 event logs to see if any errors pops up before the freeze. Honestly, it could be something pretty trivial. It may seem odd to you, but even clearing the CMOS has it own specifics. You could try enabling/disabling memory training; boost Vsoc to 1.1V; ------------- 1700X ZP-B1 (stock); X370 Taichi (UEFI 3.10); 16GB F4-3200C14-8GFX XMP; 256GB 960 EVO; RX 580 NITRO+ 8GB |
Posted By: Masardo
Date Posted: 26 Dec 2017 at 3:57am
|
" rel="nofollow - I have this motherboard and the exact same problem, which started a few weeks ago (after 2 months of using). Problem seems to go away for a few hours if the computer is left off for a while, then the freezing returns. Normally, the sound still continues for about 5 seconds. Have you solved the problem yet? There's mention of changing Vsoc voltage as well, but have no idea how to that within the BIOS.
|
Posted By: Mr Bungle
Date Posted: 26 Dec 2017 at 7:02am
|
I cleared the CMOS to no avail. Still getting same freezes. Here's what I got from the event log. Nothing directly proceeds the freeze, but here are some of the error messages I get after booting up... Event ID#360 - Windows Hello for Business provisioning will not be launched. Device is AAD joined ( AADJ or DJ++ ): Not Tested User has logged on with AAD credentials: No Windows Hello for Business policy is enabled: Not Tested Local computer meets Windows hello for business hardware requirements: Not Tested User is not connected to the machine via Remote Desktop: Yes User certificate for on premise auth policy is enabled: Not Tested Machine is governed by none policy. Event ID#10016 - The application-specific permission settings do not grant Local Activation permission for the COM Server application with CLSID
|
Posted By: zlobster
Date Posted: 26 Dec 2017 at 7:23am
|
" rel="nofollow - Hmm, typical Windoze cr@p in the log. Are you sure you have nothing else under 'Windows Logs -> System'? Mine is running fine and I'm still getting tons of errors. Did you check if the CPU & cooler are seated properly? Other cables? Did you install AMD chipset drivers? You could also try with just 1 stick of RAM. If issues still persist, swap it for the other stick. ------------- 1700X ZP-B1 (stock); X370 Taichi (UEFI 3.10); 16GB F4-3200C14-8GFX XMP; 256GB 960 EVO; RX 580 NITRO+ 8GB |
Posted By: Masardo
Date Posted: 26 Dec 2017 at 9:15am
|
Alright, I think I've got a solution: Open Device Manager -> Display Adapters, then right click your GPU, select Update Driver. Click the bottom option that allows you to browse your computer for driver software, and then the option that allows you to pick from a list of available drivers from your computer. Choose the 'Microsoft Basic Display Adapter' and proceed. By doing this I've ran into no freezes for a few hours, so it's not determined whether the freezes are completely gone, or just became more rare. I have also not tested any games, but I think they are going to perform worse as, by doing this, it probably won't use the NVIDIA drivers. |
Posted By: Mr Bungle
Date Posted: 26 Dec 2017 at 3:09pm
|
" rel="nofollow - So haven't been on long enough to see if it freezes, but gaming quality is horrible with the Windows display drivers. If my computer doesn't freeze up after switching, does this mean our culprit is probably the Nvidia drivers?
|
Posted By: Mr Bungle
Date Posted: 26 Dec 2017 at 4:53pm
| Computer just froze again... watching YouTube, screen froze, sound stuttered and then died. Right around the 2 hour mark of being turned on. |
Posted By: zlobster
Date Posted: 26 Dec 2017 at 8:36pm
|
" rel="nofollow - Seriously, default Win drivers for video? Come on! ------------- 1700X ZP-B1 (stock); X370 Taichi (UEFI 3.10); 16GB F4-3200C14-8GFX XMP; 256GB 960 EVO; RX 580 NITRO+ 8GB |
Posted By: Mr Bungle
Date Posted: 27 Dec 2017 at 2:08am
|
" rel="nofollow - So with the timing of the freezes (around two hours and the frequently if I reboot after that), wouldn't this most likely be a temperature or power issue? If I let the computer rest for awhile it goes back to two hours before freezing. |
Posted By: PetrolHead
Date Posted: 27 Dec 2017 at 4:04am
|
It sure does sound like a power or a temperature issue. I didn't recognize the PSU brand and did a quick google search. Looks like that Rosewill Valens -series isn't very good: https://www.hardocp.com/article/2016/05/05/rosewill_valens700_700w_power_supply_review/9 The link takes you to the conclusion page, but it seems like it's the DC output that's the weakest part. The summary of the DC output testing is: "After the very nice start the VALENS-700 had today, the overall DC output quality is lousy. And not just lousy, but failing in a massive right turn into a wall kind of way that hits so hard I should have been wearing a https://en.wikipedia.org/wiki/HANS_device" rel="nofollow - HANS device just watching today's chain of events.
In our testing today, the VALENS-700 started off with rather active traces on the 12v rail but quiet traces on the minor rails. However, during testing all of these values increased dramatically (and differently depending on the AC input voltage) to end out of specification on the 3.3v rail and very high on the others. How bad was it? Well, the 12v rail peaked at ~85mV of ripple/noise and the 3.3v rail did its very best to match that at ~75mV of ripple/noise (the ATX12v specification limit is 50mV). The 5v rail was no saint either as it peaked at ~40mV of ripple/noise during our testing. All things considered, this was just hugely disappointing given our other metrics today and this unit obviously trails the passing http://hardocp.com/article/2010/03/04/enermax_modu87_700w_power_supply_review/" rel="nofollow - Enermax MODU87+ 700W , http://hardocp.com/article/2011/04/06/lepa_g_series_700w_power_supply_review/" rel="nofollow - LEPA G700-MA , and the http://hardocp.com/article/2015/01/05/rosewill_capstone_750w_power_supply_review/" rel="nofollow - Rosewill Capstone 750W . Let's move on now to see how this all wraps up (you might be able to guess at this point)." The big reason high ripple is bad is that it can damage other components. Quickly too, if the ripple is bad enough. It can also make it harder to overclock. If I'm not mistaken, you're running your system at stock settings. At stock, my system's Vcore was bouncing all over the place, so if your system does the same, it would sort of makes sense that combining the half-mental auto setting of the BIOS with ripple from the PSU could cause instabilities. Especially if the ripple gets worse as the PSU warms up. Now, I know this is very "hand wavy", but if I were you, I'd try to install another PSU and see if the freezes still occur. ------------- Ryzen 5 1500X, ASRock AB350M Pro4, 2x8 GB G.Skill Trident Z 3466CL16, Sapphire Pulse RX Vega56 8G HBM2, Corsair RM550x, Samsung 960 EVO SSD (NVMe) 250GB, Samsung 850 EVO SSD 500 GB, Windows 10 64-bit |
Posted By: Masardo
Date Posted: 28 Dec 2017 at 1:47am
|
" rel="nofollow - What are the stock voltages on your RAM and CPU? Have you tried increasing/decreasing them? You might want to try changing your power plan from the control panel to 'High Performance' or 'Power saver' to see if there is any differences. |
Posted By: dane76
Date Posted: 28 Dec 2017 at 4:25pm
|
I also started swing hard freezes on my B350M Win10 build the last couple of weeks. What I have done so far: Clocked down Ryzen 1200 to default Update BIOS to 3.3 from 3.1 Update AMD chipset drivers and power plan Set memory voldtage to 1.35 Set VDDP to 1.1 Next set of action is CMOS clear. |
Posted By: dane76
Date Posted: 28 Dec 2017 at 10:02pm
Oh well. In the process of trying to remove my GPU to access the CMOS battery I tore of the PCI Express port. Damn you securing mechanism ! Thats a first for me  On to the online supplier to order a new motherboard.
|
Posted By: zlobster
Date Posted: 28 Dec 2017 at 10:45pm
|
OMG. That's bad, mkay! TBH, these new PCIe retention mechanisms prove to be troublesome for many. Now it's a chance to grab a Taichi!  ------------- 1700X ZP-B1 (stock); X370 Taichi (UEFI 3.10); 16GB F4-3200C14-8GFX XMP; 256GB 960 EVO; RX 580 NITRO+ 8GB |
Posted By: dane76
Date Posted: 28 Dec 2017 at 11:57pm
Instead of trying all kinds of trickery to get the board stable og go through a taxing RMA process I decided to take matters into my own clumpsy hands  A litttle shocked that I managed to break it so easy though  A new MSI B350M Gaming Pro has been ordered. Not a huge selection of M-ATX boards at my local dealer.
|
Posted By: zlobster
Date Posted: 29 Dec 2017 at 5:36am
Attaboy!  I usually use my 12ga shotgun for skeet shooting bad electronics/stuff I hate. No kidding. Someone behind me throws it in the air and I shoot it. Feeling is priceless!Hope you have more luck with the MSI as they are not particularly known for great products. Please do come back and do tell us how it went with the new mobo. ------------- 1700X ZP-B1 (stock); X370 Taichi (UEFI 3.10); 16GB F4-3200C14-8GFX XMP; 256GB 960 EVO; RX 580 NITRO+ 8GB |
dane76 wrote:Posted By: Mr Bungle
Date Posted: 05 Jan 2018 at 8:56am
|
" rel="nofollow - Plugged my own computer?™s PSU into the freezing computer and it froze again. Uggghhhh... temps look good... not sure what to do. CPU or MOBO issues??? I?™ve seen some of you suggesting adjustments to voltage... how do I do that safely? |
Posted By: php42
Date Posted: 06 Jan 2018 at 3:00am
" rel="nofollow -
i wouldn't go over 1.35v vcore, 1.1v SoC, or 1.4v DRAM voltage. it sounds like a ram issue IMO, i had similar issues while i was figuring out my overclock. ryzen sometimes has problems with ram even at stock settings. if you're running stock, DRAM voltage should be 1.20v by default, try bumping it to 1.30v ~ 1.35v. don't touch anything else, shouldn't need it if you're not overclocking. it's possible you may need to adjust ram timings, i don't know if that kit has an XMP profile, but XMP settings usually are not well optimized for ryzen. i would expect it to be fine at stock though, so save that as a last resort. it's also possible you just got bad ram. if you have any other ram laying around that you can swap it out for just to help narrow down the problem, that might be a good idea.
|
Posted By: Mr Bungle
Date Posted: 06 Jan 2018 at 3:23am
|
" rel="nofollow - I've taken out each stick of RAM and have had the computer freeze with just one stick plugged in. And why would the RAM work fine for two hours and then suddenly freeze the computer regardless of what programs I'm using? Doesn't the 1-2 hour countdown to freeze indicate it has to be some type of temperature related issue? The computer freezes much faster if I immediately reboot after a freeze. |
Posted By: php42
Date Posted: 06 Jan 2018 at 3:44am
|
well yes, chips perform worse (and thus need more voltage) when they're hot. once the computer has been running for a while and heated up, the stock voltage may suddenly not be enough. increasing the voltage a bit can help keep it stable even with the increased temperature. of course i'm just speculating, it could also be the memory controller, you could try raising SoC to 1.1v. but i had a similar problem with a RAM overclock becoming unstable with increased ambient temps. ------------- https://valid.x86.fr/7gfqhh" rel="nofollow"> 
|
Posted By: Masardo
Date Posted: 08 Jan 2018 at 1:30am
|
So I solved the problem, I have no freezes unless I run the computer for about 24 hours or so. What I did was lower my DRAM voltage to 1.1v at 2133 MHz, and also increase the VDDP to 1.15. Both of these can be changed in 'OC tweaker' in the BIOS. The DRAM voltage does seem low, but it has basically solved my problem as freezes are no longer a major issue. If 1.1v is unstable or you want a higher speed, you can probably try increasing the VDDP along with the DRAM voltage. Since the problem was fixed by messing around with RAM voltages, the problem probably lies somewhere with the RAM and/or the PSU.
|