• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Frequent display driver crashes and MEMORY_MANAGEMENT BSOD

eidairaman1

The Exiled Airman
Joined
Jul 2, 2007
Messages
40,435 (6.61/day)
Location
Republic of Texas (True Patriot)
System Name PCGOD
Processor AMD FX 8350@ 5.0GHz
Motherboard Asus TUF 990FX Sabertooth R2 2901 Bios
Cooling Scythe Ashura, 2×BitFenix 230mm Spectre Pro LED (Blue,Green), 2x BitFenix 140mm Spectre Pro LED
Memory 16 GB Gskill Ripjaws X 2133 (2400 OC, 10-10-12-20-20, 1T, 1.65V)
Video Card(s) AMD Radeon 290 Sapphire Vapor-X
Storage Samsung 840 Pro 256GB, WD Velociraptor 1TB
Display(s) NEC Multisync LCD 1700V (Display Port Adapter)
Case AeroCool Xpredator Evil Blue Edition
Audio Device(s) Creative Labs Sound Blaster ZxR
Power Supply Seasonic 1250 XM2 Series (XP3)
Mouse Roccat Kone XTD
Keyboard Roccat Ryos MK Pro
Software Windows 7 Pro 64
Joined
Oct 28, 2017
Messages
30 (0.01/day)
-------------------

CPU Vcore: 1.248-1.272V -up it to1.35v
CPU VDDP 0.900-0.936V up it to 1.25
DRAM Channel A/B Voltage 1.212V up it 1.3v
+3.3V 3.304V
+5V 5.040V
+12V 12.168V
VCORE SOC 0.852V that's idle. Fine.

XMP will auto-volt everything 'up'. I'd try i out, just to see the new auto power numbers. Is it 2133? at base? Are you OCing it to 2800?
Did you see the images of HWmonitor I posted a bit higher?

When I put XMP on profile 1 in the BIOS, without applying it, it seems to only increase the ram timing, not the ram voltage or any of the other voltages.
Also the voltage settings seem to use deltas rather than absolutes, so I'm guessing +0.1 for VCore +0.35 for VDDP and +0.09 for the DRAM?
Also are you sure about the VDDP to 1.25? from what I can find, people keep it between 0.9 and 1.0, though I can't even find where to set this in my bios. According to this, I think I need an X370 board to change vddp voltage
 
Joined
Oct 28, 2017
Messages
30 (0.01/day)
Laser... you can't have done W10 yet surely.
Not sure why you don't believe me but here. I even had to do the upgrade setup part in windows 7 itself a second time because the first didn't work for some reason

Laser, you tried XMP and adding a few mV to the mem?
No I'll try XMP with +0.1V on memory now I guess
If I turn on XMP apparently it'll use the XMP's voltage, and not the voltage I set. That should be a safe voltage right?
There is a Profile DDR Voltage a few lines below it that stays at 1.20V in both columns if I switch it between XMP disabled and XMP profile one so I'm not sure if it'll raise the voltage or not
 
Last edited:

TSR3

New Member
Joined
Nov 6, 2017
Messages
26 (0.01/day)
Oh I believe you, it's just you must have a decent download speed. Mines only 3MB/s.
VDDP is two things..1)CPU voltage and 2) CLDO_VDDP. I'm not good with AMD stuff so you should read up on 2) Seems important.
 

TSR3

New Member
Joined
Nov 6, 2017
Messages
26 (0.01/day)
not the voltage I set. That should be a safe voltage right?
But it might be the wrong V for your particular kit. Silicone lottery etc. My XMP (lpx3000 C15) gives me 1.356V. I ran it on 1.29 without issue. OC'd to 3466Mhz takes 1.39.
XMP changes lots of other Vs (up). So this is a test to see if you're pwr starved. That's how I always start anyway.

stays at 1.20V in both columns
Can you change them?
 
Last edited by a moderator:
Joined
Oct 28, 2017
Messages
30 (0.01/day)
But it might be the wrong V for your particular kit. Silicone lottery etc. My XMP (lpx3000 C15) gives me 1.356V. I ran it on 1.29 without issue. OC'd to 3466Mhz takes 1.39.
do you really think there's a plausible chance this will fix the problem or is just a "it might work so might as well try it" thing? Either way I'd rather go with some safer options first.
Can you change them?
No but I can change them in another tab. However it states in the manual "When using a non-XMP memory module or Extreme Memory Profile (X.M.P.)is set to Disabled, the value is displayed according to your memory specification. When Extreme Memory Profile (X.M.P.) is set to Profile1 or Profile2, the value is displayed according to the SPD data on the XMP memory." So I assume if I turn on XMP my manual setting won' matter. If you're interested, here's the manual for my motherboard.
 

TSR3

New Member
Joined
Nov 6, 2017
Messages
26 (0.01/day)
I'll give it read. Like I said upthread I'm looking at going AMD next cpu and board. On topic, you won't harm anything putting even 1.5V on your memory as it's a temporary test. Like 1.5V on CPU oc to see if it boots etc.
 
Joined
Oct 28, 2017
Messages
30 (0.01/day)
I'll give it read. Like I said upthread I'm looking at going AMD next cpu and board. On topic, you won't harm anything putting even 1.5V on your memory as it's a temporary test. Like 1.5V on CPU oc to see if it boots etc.
but won't a too low voltage prevent it from booting at all?
 

TSR3

New Member
Joined
Nov 6, 2017
Messages
26 (0.01/day)
If you boot and things (other than games) seem stable, Try-out x264 Stability Test. It's on overclock.net. It tests your CPU and Mem together in real world scenario. If you can do 5 loops then your game crashing troubleshooting could focus with the GPU not getting enough power. Or the GPU drivers need an overhaul.

but won't a too low voltage prevent it from booting at all?
Of course. Then hold down the pwr button, reboot into bios and turn it up some more. Our mem is the same and xmp rated 1.35V. You can go higher or lower.
 
Last edited by a moderator:

sneekypeet

Retired Super Moderator
Joined
Apr 12, 2006
Messages
29,407 (4.48/day)
System Name EVA-01
Processor Intel i7 13700K
Motherboard Asus ROG Maximus Z690 HERO EVA Edition
Cooling ASUS ROG Ryujin III 360 with Noctua Industrial Fans
Memory PAtriot Viper Elite RGB 96GB @ 6000MHz.
Video Card(s) Asus ROG Strix GeForce RTX 3090 24GB OC EVA Edition
Storage Addlink S95 M.2 PCIe GEN 4x4 2TB
Display(s) Asus ROG SWIFT OLED PG42UQ
Case Thermaltake Core P3 TG
Audio Device(s) Realtek on board > Sony Receiver > Cerwin Vegas
Power Supply be quiet DARK POWER PRO 12 1500W
Mouse ROG STRIX Impact Electro Punk
Keyboard ROG STRIX Scope TKL Electro Punk
Software Windows 11
If you boot and things (other than games) seem stable, Try-out x264 Stability Test. It's on overclock.net. It tests your CPU and Mem together in real world scenario. If you can do 5 loops then your game crashing troubleshooting could focus with the GPU not getting enough power. Or the GPU drivers need an overhaul.

Of course. Then hold down the pwr button, reboot into bios and turn it up some more. Our mem is the same and xmp rated 1.35V. You can go higher or lower.

Please learn to use TPU properly.... https://www.techpowerup.com/forums/...osts-and-use-the-multi-quote-features.234427/
 
Joined
Oct 28, 2017
Messages
30 (0.01/day)
If you boot and things (other than games) seem stable, Try-out x264 Stability Test. It's on overclock.net. It tests your CPU and Mem together in real world scenario. If you can do 5 loops then your game crashing troubleshooting could focus with the GPU not getting enough power. Or the GPU drivers need an overhaul.
guessing this is it? just making sure, might do a few laps with this tomorrow as well

Okay this is strange, I did a clean install of windows 7, motherboard disk drivers again, newest AMD driver again, fortnite again, launched it, very colorful corruption then crash. As usual I check the event viewer but this time there's no event about a display driver crash, so I decide to try something else and I ran furmark for 7 minutes with a max temp of 75 degrees Celsius no issues. So then I booted up Unigine's Heaven benchmark with default settings, went just fine.

So I guess OneMoar was correct in his assessment that this was due to incompatibility with Windows 7. Plus a weird issue with Fortnite itself.
Actually, I just started it again with the debug option enabled to see if I could get a debug message of the crash, but there still was none and this time the display driver did crash not once but twice. So this might have been just a Fortnite problem all along.
Then again there were the random display driver crashes and the BSODs so I can't tell anymore

EDIT:: Just had another display driver crash while just browsing, oddly enough it opened epic games launcher right after.
 
Last edited by a moderator:

TSR3

New Member
Joined
Nov 6, 2017
Messages
26 (0.01/day)
Okay this is strange, I did a clean install of windows 7, motherboard disk drivers again, newest AMD driver again, fortnite again, launched it, very colorful corruption then crash. As usual I check the event viewer but this time there's no event about a display driver crash, so I decide to try something else and I ran furmark for 7 minutes with a max temp of 75 degrees Celsius no issues. So then I booted up Unigine's Heaven benchmark with default settings, went just fine.

So I guess OneMoar was correct in his assessment that this was due to incompatibility with Windows 7. Plus a weird issue with Fortnite itself.
Actually, I just started it again with the debug option enabled to see if I could get a debug message of the crash, but there still was none and this time the display driver did crash not once but twice. So this might have been just a Fortnite problem all along.
Then again there were the random display driver crashes and the BSODs so I can't tell anymore

EDIT:: Just had another display driver crash while just browsing, oddly enough it opened epic games launcher right after.

Afternoon Laser. Had a read of your manual when I had a chance.
Page 1) Support for Windows 7 64-bit.
Manual says you've got dual bios. Nice. So nothing can go wrong :)

Downer that things sound worse. Have you run x264 stability? How did it go?
 
Joined
Oct 28, 2017
Messages
30 (0.01/day)
Afternoon Laser. Had a read of your manual when I had a chance.
Page 1) Support for Windows 7 64-bit.
Manual says you've got dual bios. Nice. So nothing can go wrong :)

Downer that things sound worse. Have you run x264 stability? How did it go?
I can't seem to access mega.nz due to some security hazard or whatnot. Had the same on my laptop. Thus I can't get the "package" as he describes it.
 

TSR3

New Member
Joined
Nov 6, 2017
Messages
26 (0.01/day)
I can't seem to access mega.nz due to some security hazard or whatnot. Had the same on my laptop. Thus I can't get the "package" as he describes it.

Strange. I can get there straight away. You English?
 
Joined
Oct 28, 2017
Messages
30 (0.01/day)
Strange. I can get there straight away. You English?
seems to be blocked by Belgium itself.
NET::ERR_CERT_COMMON_NAME_INVALID
Subject: *.services.belgium.be
Issuer: QuoVadis Europe SSL CA G1
Expires on: 15 jul. 2018
 

TSR3

New Member
Joined
Nov 6, 2017
Messages
26 (0.01/day)
seems to be blocked by Belgium itself.
NET::ERR_CERT_COMMON_NAME_INVALID
Subject: *.services.belgium.be
Issuer: QuoVadis Europe SSL CA G1
Expires on: 15 jul. 2018
Your English is perfect. Overclock.net don't host files. Perhaps one of the members here might look to host the .zip on TPU.

I could upload it but I would need permission from the original developers and admin here. I don't want to sign-up the OC.net.
 
Last edited:
Joined
Oct 28, 2017
Messages
30 (0.01/day)
I mean would there even be a point to x264 if I ran heaven benchmark? I imagine that stresses GPU CPU and RAM
 

TSR3

New Member
Joined
Nov 6, 2017
Messages
26 (0.01/day)
I mean would there even be a point to x264 if I ran heaven benchmark? I imagine that stresses GPU CPU and RAM

Exactly. You want to narrow-down the problem. If the CPU, mem, PCH systems are 100% stable you will know what to concentrate on.
 
Joined
Oct 28, 2017
Messages
30 (0.01/day)
Exactly. You want to narrow-down the problem. If the CPU, mem, PCH systems are 100% stable you will know what to concentrate on.
since furmark ran for like 7 min no issues it couldn't have been a simple GPU load error either, no? I'm getting kind of confused
 

TSR3

New Member
Joined
Nov 6, 2017
Messages
26 (0.01/day)
since furmark ran for like 7 min no issues it couldn't have been a simple GPU load error either, no? I'm getting kind of confused

Yes, confusing. But it's interesting solving problems like this. Try Prime95? Blend for a few minutes. Then small ffts.
 
Joined
Oct 28, 2017
Messages
30 (0.01/day)
Yes, confusing. But it's interesting solving problems like this. Try Prime95? Blend for a few minutes. Then small ffts.
During blend, on the first test, worker #5 #6 #9 #10 & #12 had
"FATAL ERROR: Rounding was 0.5, expected less than 0.4 Hardware failure detected, consult stress.txt file, torture test completed 0 tests in 1 minute 1 errors 0 warnings"
Stopped it after seeing this and a minute or two after that driver crashed.

Ran small sffts for ~8 min until it said "Self test 21K passed" No errors on any of the workers.

Put on the XMP profile in bios, booted fine, checked speccy and noticed it had correct speed and timings and decided to try the blend again.
Same deal but this time I kept running it when it got the fatal errors and one by one all the workers failed. Like 5-10 min after display driver crashed again
 
Last edited:

eidairaman1

The Exiled Airman
Joined
Jul 2, 2007
Messages
40,435 (6.61/day)
Location
Republic of Texas (True Patriot)
System Name PCGOD
Processor AMD FX 8350@ 5.0GHz
Motherboard Asus TUF 990FX Sabertooth R2 2901 Bios
Cooling Scythe Ashura, 2×BitFenix 230mm Spectre Pro LED (Blue,Green), 2x BitFenix 140mm Spectre Pro LED
Memory 16 GB Gskill Ripjaws X 2133 (2400 OC, 10-10-12-20-20, 1T, 1.65V)
Video Card(s) AMD Radeon 290 Sapphire Vapor-X
Storage Samsung 840 Pro 256GB, WD Velociraptor 1TB
Display(s) NEC Multisync LCD 1700V (Display Port Adapter)
Case AeroCool Xpredator Evil Blue Edition
Audio Device(s) Creative Labs Sound Blaster ZxR
Power Supply Seasonic 1250 XM2 Series (XP3)
Mouse Roccat Kone XTD
Keyboard Roccat Ryos MK Pro
Software Windows 7 Pro 64
During blend, on the first test, worker #5 #6 #9 #10 & #12 had
"FATAL ERROR: Rounding was 0.5, expected less than 0.4 Hardware failure detected, consult stress.txt file, torture test completed 0 tests in 1 minute 1 errors 0 warnings"
Stopped it after seeing this and a minute or two after that driver crashed.

Ran small sffts for ~8 min until it said "Self test 21K passed" No errors on any of the workers.

Put on the XMP profile in bios, booted fine, checked speccy and noticed it had correct speed and timings and decided to try the blend again.
Same deal but this time I kept running it when it got the fatal errors and one by one all the workers failed. Like 5-10 min after display driver crashed again


Bump ram voltages up, also get memtest 86
 

TSR3

New Member
Joined
Nov 6, 2017
Messages
26 (0.01/day)
Bump ram voltages up, also get memtest 86

What he said. Not enough juice. Seems we're back to where we where yesterday.
Also increase VDDO_SOC voltage by 0.25 (SOC is integrated memory controller power settings). I can't see it listed in the book. It's in M.I.T. somewhere.
 
Top