X370 Killer ac/sli Memory Problems & Freezing
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=5197
Printed Date: 23 Dec 2024 at 12:33pm
Software Version: Web Wiz Forums 12.04 - http://www.webwizforums.com
Topic: X370 Killer ac/sli Memory Problems & Freezing
Posted By: asdf2345
Subject: X370 Killer ac/sli Memory Problems & Freezing
Date Posted: 21 May 2017 at 4:15am
|
" rel="nofollow - First thing to note. I was an individual who had a x370 Taichi in which bios 1.94a did damage to some extent to the motherboard. It didn't brick it like others but it was glitchy and Windows threw errors. I replaced the mobo with a Killer ac/sli because I was refunded instead of replaced, which was fine because of the Taichi shortage. All other parts are still the same. Hardware AMD Ryzen R7 1700 x370 Killer ac/SLI (Bios ver 2.30) Nvidia GTX 970 4x8GB 32GB Team Dark 3000MHz Ram @2667MHz AMD R7 240GB ssd Toshiba P300 3TB hdd Nzxt Hale82 V2 700w (Also used a Rosewill Hive 750w psu) Win 8.1 Pro x64 Everything had been running fine for nearly a month. All of a sudden the computer started to constantly BSOD, Memory_Management and IRQL_Not_Less_or_Equal errors. I've been using memtest86+ for the last two days and found errors on 3 of the Team Dark sticks. The one good stick I have was further tested in the other DIMM slots. At this point I assembled everything and switched to the Rosewill PSU just to eliminate that possibility. Again, the computer still BSODs or freezes on the desktop. Memtest still reports that the single stick is good. Bad CPU? I just find it strange how out of the blue 3 memory sticks simultaneously go bad. Any further thoughts or insight is greatly appreciated. Thanks
|
Replies:
Posted By: asdf2345
Date Posted: 21 May 2017 at 5:14am
|
I swapped out the 970 for an old ati 2600xt just to make sure it's not a video card memory issue. And it's not, still getting the same BSOD errors. |
Posted By: moonstarmac
Date Posted: 26 May 2017 at 11:07am
|
I have noticed with my Team Group Vulcan that going over 2400 either doesn't boot or causes serious issues. I would suggest dropping the speed and seeing if it fixes the issue. Sadly most of the Hynix XMP profiles and DDR4 ram are build for Intel chips and are still having issues with AM4. |
Posted By: Robenger
Date Posted: 30 May 2017 at 9:22am
| I have the same board and all of my issues were the same and it had to do with the OC on the CPU and memory. One of those two are not stable. Run everything on default speeds and test again. |
Posted By: parsec
Date Posted: 30 May 2017 at 12:05pm
What was the DRAM voltage set to in the UEFI/BIOS? Or the SOC voltage and SOC LLC setting? Just using the Auto values? You can check their values in the H/W Monitoring screen. So the problems you had with your X370 Taichi were completely different than the BSODs you were getting with the X370 Killer SLI/ac? Did you happen to run Memtest when you had the memory in the X370 Taichi? My point is the memory might have had problems out of the box, and since you have 32GB, the bad memory chip/area was not used until recently. Or one at a time the DIMMs began having errors, and it was not until three DIMMs had errors that the BSOD issue began happening. Electronic component failures tend to happen early on in their usage if they have some internal defect. Did you happen to try running the memory at say 2400, and then run Memtest? You could test for a failing memory controller by putting the memory in another PC that uses DDR4 and running Memtest on it, if possible. I have the same board, with a 1700X with G.SKILL FlareX memory, and I do not get random BSODs. ------------- http://valid.x86.fr/48rujh" rel="nofollow"> 
|
asdf2345 wrote:Posted By: asdf2345
Date Posted: 05 Jun 2017 at 7:31pm
|
First, I had issues with the newest memtest not booting up at all. I was able to get memtest86 7.0 to work and tested the ram with that version. Despite what overclock values I set in the bios, memtest would always run the memory at 2133 @ 15-15-15-36. This was back when I first posted here. Not too long after my initial posts, I wanted to try ruling out OS issues (Win 8.1). I installed Win 7 and everything seemed to be stable and fine. The CPU was set to stock and has been for a while now as I switched to using the Wraith cooler, the Hyper 212 Evo covers up the first ram slot and was always in the way. Seeing as to how stable the system was, I added back one of the memory sticks that was previously throwing memtest errors. I reran memtest, and it was now coming up clean. I decided to stay with that configuration, A2 and B2 slots, and set the ram clock speed to 2667 (as shown in the Asrock Killer mobo pic) and using the voltage/timings recommended by Team Dark's website page 1.2 @ 15-17-17-35. http://www.teamgroupinc.com/en/product/T-Force%20Gaming/Gaming%20Memory%20Modules/DDR4%20MEMORY%20MODULES/DARK%20DDR4" rel="nofollow - http://www.teamgroupinc.com/en/product/T-Force%20Gaming/Gaming%20Memory%20Modules/DDR4%20MEMORY%20MODULES/DARK%20DDR4 For the last one and half weeks everything seemed to be good, gaming and otherwise. All of a sudden Deus Ex: Mankind Divided began randomly crashing, citing memory error 0xc0000005. Just Cause 3 will now also crash, same error. Prototype 2 can't even make it to the title screen without crashing and throwing that exact same error. I thought ok, maybe that one stick was bad (B2 slot). I removed it and I still get that same memory crash when gaming. I placed the stick back in and ran memtest, no errors. I then set the ram back to stock auto settings, and the games are still crashing with the same c0000005 error. I'm going to next try removing the A2 stick (the one that has never thrown any memtest errors). I also want to make clear that I haven't had any BSODs in Windows 7. The memory issues seem to be strictly gaming bound. |
Posted By: moonstarmac
Date Posted: 07 Jun 2017 at 1:44am
|
Try using all 4 sticks @ 2400. With Hynix based chips it seems to run more stable under the AGESA 1.0.0.4 BIOS support. After we get the AGESA 1.0.0.6 update you should be able to run it at 3000. |