[SOLVED] Please help me identify if RAM or motherboard is faulty

Oct 9, 2020
27
1
35
Hi everyone,

I apologize for the long post, but I have been having stability issues with my Work/Gaming computer and would like to please ask for some help.

System specs:
CPU: I9-9900k Processor, [non-OC].
Motherboard: ASUS ROG Maximus XI Hero.
RAM: Kingston HyperX Predator RGB HX440C19PB3AK2/16 16GB (2x8GB) DDR4, [XMP-4000], installed in slots A2 & B2 as per manual.
PSU: Corsair AX860 Platinum 860W.
GPU: none, currently monitor is connected to MOBO HDMI.

Issue:
Yesterday while working on AutoCAD, suddenly my PC became slow, Task Manager showed high RAM usage by Autocad (95% +), ending Autocad via task manager solved high ram usage issue. However, when I opened Autocad to work on the same project, the same issue occurred, this time I didn’t end Autocad via task manager to see what would happen, so after a minute or two of high memory usage, The PC display went black for a moment and then back to the Windows 10 login screen, I logged in and no errors were showing in event viewer. So I opened Autocad for the third time, and the same problem occurred, This time my PC restarted, and once logged in I checked event viewer to see critical error: Event ID 41- Kernel power-system.

Diagnostics:
Windows memory diagnostics: No issue detected.
Prime95/Large FFT/AVX-512 Disabled/45 Minute run time: Fatal error in worker 10 “rounding was 0.49xx expected 0.4” (I did notice that 2 CPU threads were not working at 100% during the torture test)
Prime95/Small FFT/AVX-512 Disabled/36 tests 1 hour 7 minutes run time: no error nor warnings.
Memtest86: completed 2 passes, 6 errors found on first pass, 59 cumulative errors by end of second pass.

Question:
I am please asking for advice on what to do next. Is there a way to isolate anything to see whether the issue is from the CPU or ram or motherboard?

Thank you very much in advance for your help.
 
Solution
Go to Aida64 up on the menus where the stress test is and it says CPUID .... Click that and it will def tell you exactly what your memory timings are and Im positive it will show 1800Mhz which is 3600Mhz effectively dual channel. Also you did the smart thing as 4000Mhz or 3600Mhz is the same thing you will not notice a difference in performance. AMD systems love RAM and Intel is not picky. For AMD I suggest 3600Mhz the sweet spot. For Intel 2933Mhz is recommended but 3200Mhz will work and that its sweet spot anything higher is in a way a waiste. Keep it at 3600Mhz and have your stability my friend. Get back to me after you check in Aida64 up top menu go to CPUID and youll know exactly whats up and also let me know how the stress...

rgs80074

Distinguished
Oct 28, 2010
159
5
18,715
I am sure others will chime in but,

this is what i've always done when i thought I had ram issues.

first go down to 1 ram chip, if pc works then move onto another ram chip, if it don't work then put in another slot and try again.

just keep doing that for all the ram sticks. if they all seem to work, start over with one stick and test in each ram slot, if that still shows everything working

one ram chip at a time and run a test on the chip (can't recall what I've not had to do it in a while and every time I do i have to google search for a free tester).

anytime i've had a bad chip thats identified it.

I've never had a ram slot go bad on a motherboard or had a cpu go bad so can not help there.
 
Oct 9, 2020
27
1
35
I am sure others will chime in but,

this is what i've always done when i thought I had ram issues.

first go down to 1 ram chip, if pc works then move onto another ram chip, if it don't work then put in another slot and try again.

just keep doing that for all the ram sticks. if they all seem to work, start over with one stick and test in each ram slot, if that still shows everything working

one ram chip at a time and run a test on the chip (can't recall what I've not had to do it in a while and every time I do i have to google search for a free tester).

anytime i've had a bad chip thats identified it.

I've never had a ram slot go bad on a motherboard or had a cpu go bad so can not help there.

Thank you for your reply,
I have ran the memory test: memtest86 on my ram sticks individually, First i ran 2 passes on ram stick 1 in slot A2 and no errors were reported. Then i ran the test on ram stick 2 in slot A2 , no errors were reported either from 2 passes. From this information would it be safe to conclude that the ram sticks are not faulty ?
 
Oct 9, 2020
27
1
35
If you have errors as described in Memtest86 then RMA the full kit
Thanks for your reply,
So i have just finished running memtest 86 with 2 passes on my ram sticks individually. 2 passes on ram stick 1 in slot A2 reported no errors, and 2 passes on ram stick 2 in slot A2 reported no errors either, does this indicate that the ram sticks are not faulty?
could the error still be from the RAM if individually testing them provided no errors?
 

Turtle Rig

Prominent
BANNED
Jun 23, 2020
772
104
590
Your RAM usage pegs out what about CPU usage? Does that go to hundred percent as well? Also I don't know how your using AutoCad with a iGPU. If the screen goes black I doubt its a RAM issue as RAM issues give a BSOD IRQ NOT EQUAL OR LESS or other BSODs. As for event viewer by default in Windows 10 it does not log BSODs. So I need to know if your being pegged on the CPU as well or just the RAM? First thing I would do is go to BIOS and choose restore to optimized defaults. As for your RAM you first said you got errors then you said you don't get errors so I am confused. Also how long does the black screen last? Is it couple seconds or minutes or a instant and what not. For RAM usage were you pegged out meaning was all your RAM shown as in usage? RAM being pegged like that is very odd. At first I doubted it is the RAM but if you get errors which you said you did then you must RMA the RAM Kit period. Then take it from there. Also it might even be the PSU dying or capacitors going bad or not providing enough ampage on the 12v rail. What happens if you don't use AutoCad and mess around in the desktop all day long does this problem occur? First things first as you said you got RAM errors so you must RMA the RAM kit. Then get the new kit and see what happens. If problem persists then it can be the iGPU going faulty or it can be PSU but well tackle that when we get there. See if you can answer couple of my question then we will further troubleshoot this issue.🤷‍♀️👶
 
Oct 9, 2020
27
1
35
Your RAM usage pegs out what about CPU usage? Does that go to hundred percent as well? Also I don't know how your using AutoCad with a iGPU. If the screen goes black I doubt its a RAM issue as RAM issues give a BSOD IRQ NOT EQUAL OR LESS or other BSODs. As for event viewer by default in Windows 10 it does not log BSODs. So I need to know if your being pegged on the CPU as well or just the RAM? First thing I would do is go to BIOS and choose restore to optimized defaults. As for your RAM you first said you got errors then you said you don't get errors so I am confused. Also how long does the black screen last? Is it couple seconds or minutes or a instant and what not. For RAM usage were you pegged out meaning was all your RAM shown as in usage? RAM being pegged like that is very odd. At first I doubted it is the RAM but if you get errors which you said you did then you must RMA the RAM Kit period. Then take it from there. Also it might even be the PSU dying or capacitors going bad or not providing enough ampage on the 12v rail. What happens if you don't use AutoCad and mess around in the desktop all day long does this problem occur? First things first as you said you got RAM errors so you must RMA the RAM kit. Then get the new kit and see what happens. If problem persists then it can be the iGPU going faulty or it can be PSU but well tackle that when we get there. See if you can answer couple of my question then we will further troubleshoot this issue.🤷‍♀️👶
Thank you for your reply,

If i recall correctly,, when i noticed the ram usage being high, the CPU usage did not seem abnormal to me- somewhere around 30% when ram usage was 95%if i recall correctly- Regarding the RAM errors, i got errors when i ran memtest86 for 2 passes while having both ram sticks installed in mobo slots A2 and B2, when individually testing the sticks, by installing only stick 1 in slot A2 and running memtest86 for 2 passes, no errors were reported, i repeated the same procedure individually testing ram stick 2 in slot A2 and no errors were reported in memtest86 after 2 passes as well.

Regarding the black screen, the first time it happened i actually got a green screen for a moment, then a black screen, i previously had an amd gpu installed , i removed it around a month ago but i did not uninstall radeon software, after searching up green screen i came accross AMD users asking the same question and i assumed it may be from the radeon software that i did not uninstall yet, so i uninstalled it, and on the next freeze i got a black screen that lasted for about 3 minutes and then the pc restarted itself .When normally using my pc during the day watching youtube and browsing websites etc, i don't notice any hicccups. (Regarding RMA'ing my hardware, would the store i bought from think i'm a liar if i rma my ram then tell them i want to rma something else?)
 
Last edited:

Turtle Rig

Prominent
BANNED
Jun 23, 2020
772
104
590
Ok perhaps the B2 RAM slot is busted if both sticks work alone on A1. Apparently it seems like slot B2 is busted from what you wrote. So it can be the motherboard and if both sticks individually passed the test without error then its not the RAM its the motherboard or that B2 slot. Im leaning more towards the facts your motherboard is faulty. As for RMAing RAM you have a lifetime warranty on it and you don't RMA to store you deal with the company that makes the RAM. But it doesn't seem like you have to RMA the now that you tell me they work great one at a time on A1. IMHO I think your motherboard has gone south. Also I doubt it is the power supply. Its complicated I know, but first thing I would do is put in the single stick of RAM in A1 and torture the heck out of the PC with Prime95 and see if it crashes. Then try the other RAM stick and once again see if it crashes or not crash but give you the RAM usage issue your talking about. I put my money on the motherboard or iGPU especially if you see a black screen then it restores itself. So it looks like possibly you might have to RMA the motherboard. I hope this all makes sense. One thing you can try is get a dedicated video card and disable iGPU and see if your getting this same issue. Because the motherboard can go bad but also it could be the iGPU messing with your board and RAM slots. So once again put in a temporary video card in there and disable iGPU and then get back to me. Also to add to this you can also try a different HDMI cable because cables can go bad. So IMO I feel like after all you said that the motherboard is botched.🤷‍♀️👶💯✝
 
Oct 9, 2020
27
1
35
Ok perhaps the B2 RAM slot is busted if both sticks work alone on A1. Apparently it seems like slot B2 is busted from what you wrote. So it can be the motherboard and if both sticks individually passed the test without error then its not the RAM its the motherboard or that B2 slot. Im leaning more towards the facts your motherboard is faulty. As for RMAing RAM you have a lifetime warranty on it and you don't RMA to store you deal with the company that makes the RAM. But it doesn't seem like you have to RMA the now that you tell me they work great one at a time on A1. IMHO I think your motherboard has gone south. Also I doubt it is the power supply. Its complicated I know, but first thing I would do is put in the single stick of RAM in A1 and torture the heck out of the PC with Prime95 and see if it crashes. Then try the other RAM stick and once again see if it crashes or not crash but give you the RAM usage issue your talking about. I put my money on the motherboard or iGPU especially if you see a black screen then it restores itself. So it looks like possibly you might have to RMA the motherboard. I hope this all makes sense. One thing you can try is get a dedicated video card and disable iGPU and see if your getting this same issue. Because the motherboard can go bad but also it could be the iGPU messing with your board and RAM slots. So once again put in a temporary video card in there and disable iGPU and then get back to me. Also to add to this you can also try a different HDMI cable because cables can go bad. So IMO I feel like after all you said that the motherboard is botched.🤷‍♀️👶💯✝
Sorry, i forgot to mention previously that before i got a black screen on the second freeze, i first got a green screen followed by the display turning off and then back on to the login screen, after research i assumed it was related to amd radeon software that i did not uninstall after selling my gpu. So after uninstalling radeon software, on the second freeze that's when got black screen followed by a reset. I will now do as you have suggested and run p95 Large FFTs to see if any errors get reported, unfortunately i don't have any graphics card on me atm as i got let down by the 3080 launch. Thank you for your help, I'll run the tests and hopefully get back to you soon.
 
Oct 9, 2020
27
1
35
Ok perhaps the B2 RAM slot is busted if both sticks work alone on A1. Apparently it seems like slot B2 is busted from what you wrote. So it can be the motherboard and if both sticks individually passed the test without error then its not the RAM its the motherboard or that B2 slot. Im leaning more towards the facts your motherboard is faulty. As for RMAing RAM you have a lifetime warranty on it and you don't RMA to store you deal with the company that makes the RAM. But it doesn't seem like you have to RMA the now that you tell me they work great one at a time on A1. IMHO I think your motherboard has gone south. Also I doubt it is the power supply. Its complicated I know, but first thing I would do is put in the single stick of RAM in A1 and torture the heck out of the PC with Prime95 and see if it crashes. Then try the other RAM stick and once again see if it crashes or not crash but give you the RAM usage issue your talking about. I put my money on the motherboard or iGPU especially if you see a black screen then it restores itself. So it looks like possibly you might have to RMA the motherboard. I hope this all makes sense. One thing you can try is get a dedicated video card and disable iGPU and see if your getting this same issue. Because the motherboard can go bad but also it could be the iGPU messing with your board and RAM slots. So once again put in a temporary video card in there and disable iGPU and then get back to me. Also to add to this you can also try a different HDMI cable because cables can go bad. So IMO I feel like after all you said that the motherboard is botched.🤷‍♀️👶💯✝
Hello Again,
So, here's a summary of the tests that i've done soo far and a possible conclusion that i have come across (Note: P95 tests Done with AVX-512 Disabled):

Tests with both memory sticks installed in Mobo slots A2 & B2 running at 4000Mhz XMP:
P95 Large FFT'S/45M/Fatal error in worker 10.
P95 Small FFT's/60M/No error.
Memtest86/2 Passes/59 Cumulative errors.

Test with 1 stick (Stick 1) installed in Mobo slot A2 running at 4000Mhz XMP:
P95 Large FFT'S/60M/No error.
P95 Small FFT'S/60M/No error.
Memtest86/2 Passes/No error.

Test with 1 stick (Stick 2) installed in Mobo slot A2 running at 4000Mhz XMP:
P95 Large FFT'S/60M/No error.
P95 Small FFT'S/60M/No error.
Memtest86/2 Passes/No error.

Judging from the tests, i think it's safe to rule out faulty RAM sticks (Please correct me if i'm wrong),So there are two possible culprits that could be causing the issue, either the B2 RAM slot on the MOBO has gone faulty, Or there may be some iGPU/XMP/16GB RAM incompatibility that I may be unaware of?

Thank you once again for your help
 
Hi everyone,

I apologize for the long post, but I have been having stability issues with my Work/Gaming computer and would like to please ask for some help.

System specs:
CPU: I9-9900k Processor, [non-OC].
Motherboard: ASUS ROG Maximus XI Hero.
RAM: Kingston HyperX Predator RGB HX440C19PB3AK2/16 16GB (2x8GB) DDR4, [XMP-4000], installed in slots A2 & B2 as per manual.
PSU: Corsair AX860 Platinum 860W.
GPU: none, currently monitor is connected to MOBO HDMI.

Issue:
Yesterday while working on AutoCAD, suddenly my PC became slow, Task Manager showed high RAM usage by Autocad (95% +), ending Autocad via task manager solved high ram usage issue. However, when I opened Autocad to work on the same project, the same issue occurred, this time I didn’t end Autocad via task manager to see what would happen, so after a minute or two of high memory usage, The PC display went black for a moment and then back to the Windows 10 login screen, I logged in and no errors were showing in event viewer. So I opened Autocad for the third time, and the same problem occurred, This time my PC restarted, and once logged in I checked event viewer to see critical error: Event ID 41- Kernel power-system.

Diagnostics:
Windows memory diagnostics: No issue detected.
Prime95/Large FFT/AVX-512 Disabled/45 Minute run time: Fatal error in worker 10 “rounding was 0.49xx expected 0.4” (I did notice that 2 CPU threads were not working at 100% during the torture test)
Prime95/Small FFT/AVX-512 Disabled/36 tests 1 hour 7 minutes run time: no error nor warnings.
Memtest86: completed 2 passes, 6 errors found on first pass, 59 cumulative errors by end of second pass.

Question:
I am please asking for advice on what to do next. Is there a way to isolate anything to see whether the issue is from the CPU or ram or motherboard?

Thank you very much in advance for your help.
Obviously, you are running very high speed RAM. You do it by enabling DOCOin the BIOS ?
Have you tried disabling DOCP and run Autocad after that ?
If that correct the problem try to progressively increase the RAM speed (enable DOCP but manually enter a speed smaller than 4400 Mhz)
 
Oct 9, 2020
27
1
35
Obviously, you are running very high speed RAM. You do it by enabling DOCOin the BIOS ?
Have you tried disabling DOCP and run Autocad after that ?
If that correct the problem try to progressively increase the RAM speed (enable DOCP but manually enter a speed smaller than 4400 Mhz)
Thank you for your reply,
I get 4000Mhz ram speed by enabling XMP profile in the Bios, i haven't changed anything other than selecting XMP profile for the ram.
 
Thank you for your reply,
I get 4000Mhz ram speed by enabling XMP profile in the Bios, i haven't changed anything other than selecting XMP profile for the ram.
Ok but by disabling DOCP does the memory occupation problem by Autocad persist ?
Don't forget that you are using this PC at it's highest possible speed, and it is not really guarantee to work flawlessly near these borders,

Edit: sorry the RAM (or very similar ) is on the QVL, but nevertheless as you can read on it' specification sheet , nothing as guarantee as the highest speed
https://www.kingston.com/dataSheets/HX440C19PB3AK2_16.pdf
 
Last edited:
Oct 9, 2020
27
1
35
Ok but by disabling DOCP does the memory occupation problem by Autocad persist ?
Don't forget that you are using this PC at it's highest possible speed, and it is not really guarantee to work flawlessly near these borders,

Edit: sorry the RAM (or very similar ) is on the QVL, but nevertheless as you can read on it' specification sheet , nothing as guarantee as the highest speed
https://www.kingston.com/dataSheets/HX440C19PB3AK2_16.pdf
Hello again,
So, i restarted my computer and had a look in the BIOS, i couldn't see any DOCP mentioned anywhere, I think maybe DOCP is exclusive to AMD motherboards?, as i can't see it, i can only see XMP, i did notice however that my ram voltage was set to 1.35v in BIOS Setting, but it was displayed as 1.361v in the bios display,is that fine?
 
Last edited:
Hello again,
So, i restarted my computer and had a look in the BIOS, i couldn't see any DOCP mentioned anywhere, I think maybe DOCP is exclusive to AMD motherboards?, as i can't see it, i can only see XMP, i did notice however that my ram voltage was set to 1.35v in BIOS Setting, but it was displayed as 1.361v in the bios display,is that fine?
Ah yes XMP indeed, for Intel CPU.
1.35 or 1.36 are fine
Is your CPU overclocked ?
 
Oct 9, 2020
27
1
35
Ah yes XMP indeed, for Intel CPU.
1.35 or 1.36 are fine
Is your CPU overclocked ?
CPU is currently not over-clocked, running default intel turbo speeds, and Asus multi core enhancement is turned off in the BIOS.
However i did notice i had the option to choose between XMP 1 and XMP 2, do you know anything about that? I understand it's not a different speed because underneath that settings i have an option to either choose XMP 4000 or XMP 3600, i also noticed on CPU-Z it shows my RAM as XMP 2.0 , i wonder if it's related to XMP 1 and 2?
 
CPU is currently not over-clocked, running default intel turbo speeds, and Asus multi core enhancement is turned off in the BIOS.
However i did notice i had the option to choose between XMP 1 and XMP 2, do you know anything about that? I understand it's not a different speed because underneath that settings i have an option to either choose XMP 4000 or XMP 3600, i also noticed on CPU-Z it shows my RAM as XMP 2.0 , i wonder if it's related to XMP 1 and 2?
Yep, XMP-1 puts your RAM at 4000 MHz and XMP-2 at 3600 MHz
Two different profiles. You can choose the speed.
It's something else than the XMP 2.0 in CPU-Z which is the version of the standard
 
Oct 9, 2020
27
1
35
Yep, XMP-1 puts your RAM at 4000 MHz and XMP-2 at 3600 MHz
Two different profiles. You can choose the speed.
It's something else than the XMP 2.0 in CPU-Z which is the version of the standard
Yes i can select different xmp profiles which change the speed of the ram, but this settings is different, i'm not in my bios atm, but if i recall correctly, XMP 1 had settings that were optimised for Asus, whereas XMP 2 loaded default ram settings, do you think it has anything to do with my issue?
Thank you for your time.
 
Yes i can select different xmp profiles which change the speed of the ram, but this settings is different, i'm not in my bios atm, but if i recall correctly, XMP 1 had settings that were optimised for Asus, whereas XMP 2 loaded default ram settings, do you think it has anything to do with my issue?
Thank you for your time.
I am not familiar with you Asus board, but the XMP profiles are hard coded in the RAM sticks.
High end sticks as you have can have several XMP profiles from which you can choose.
The basic speed for you RAM is 2400 MHZ, and it has 2 XMP profiles saved in the sticks (3600 and 4000 MHz with their accompanying latencies and voltages).
This should have nothing to do with Asus, MSI, Gigabyte or any board brand.
It is simply there to allow a lower speed if problems occur whit the highest speed.
Which is exactly what I am trying to now by disabling XMP, and see if your memory occupation problem with AUTOCAD disappears.
 
Oct 9, 2020
27
1
35
OK . I see on page 3-15 of your manual what you are referring to. As I said, as I am not familiar with the board I don't know what appears on screen when you choose the various options
Hello,
No worries that's understandable, i Can take a photo of the different settings if you want, for now i am going to try running tests using XMP 2 and 4000MHZ and see how i go.
 
hello!
So i changed the XMP setting i told you about in the BIOS from XMP 1 to XMP 2 , with both ram sticks, i managed to do 2 passes in memtest86 with no errors, and it passed 1 hour on prime 95 large FFT, it seems like my issue was from the XMP 1 setting.
Could be if Asus is trying to push things beyond the official capabilities of the RAM.
Also the problem could be related to the memory controler inside the CPU if Asus tries to push things to far . Using the HyperX profile seems the prudent choice
 
Oct 9, 2020
27
1
35
Could be if Asus is trying to push things beyond the official capabilities of the RAM.
Also the problem could be related to the memory controler inside the CPU if Asus tries to push things to far . Using the HyperX profile seems the prudent choice
Yes you're right, do you recommend a method to test the CPU controller? Would a stress test on P95 Large FFT pick up any errors in the CPU controller since it stresses the controller and RAM together?