[SOLVED] Memory Management BSOD after a while

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Mar 14, 2019
33
1
35
0
Rails are within spec at least - doesn't mean the PSU isn't faulty, just that the rails don't appear to be doing anything obviously out of the ordinary.
Will wait for feedback following Colifs post.
I haven't ever run it myself - https://appuals.com/how-to-run-a-cpu-stress-test-using-prime95/

my case filters probably need to be cleaned. Winter here so over heating unlikely.
So my PC just froze, heres a log file from the HW thing no BSOD but was weird - log
 
Mar 14, 2019
33
1
35
0
Did it freeze during Prime95 testing?

All voltages are in check at least. 12V is fluctuating a bit, but nothing unusual.

Guessing where this log ends is where the freeze occurs?
I haven't ever run it myself - https://appuals.com/how-to-run-a-cpu-stress-test-using-prime95/

my case filters probably need to be cleaned. Winter here so over heating unlikely.
I don't think it was running and the log most likely ended there. Good news and bad news tho I just ran a stress test with prime 95 and in one of the tests it said hardware failure detected, rounding was 0.5 when the expected needed to be less than 0.4. I believe this was worker #3 so 2nd or 3rd core.
 
Reactions: PC Tailor
I feel its hardware, I am just not sure which. Freezing with no bsod is not a windows habit.

I have asked for a 2nd opinion on thread, I will look at those reports again later

@PC Tailor he has a I3 - 8100, pretty sure you can't overclock those?
Well it's a locked CPU, but usually the base clock can be modified (why anyone wants to beats me). But it can also be undervolted from my understanding.
Wondering why max speed on BIOS DUMP says 8300MHz (not the first time I've seen this).

Processor Version Intel(R) Core(TM) i3-8100 CPU @ 3.60GHz
Processor Voltage 8ah - 1.0V
External Clock 100MHz
Max Speed 8300MHz
Current Speed 3600MHz
The Prime95 error can often come from unstable CPU configs, which is why i had to clarify.
I believe this error CAN also be thrown out by RAM.
 
Reactions: REVLX
Mar 14, 2019
33
1
35
0
Well it's a locked CPU, but usually the base clock can be modified (why anyone wants to beats me). But it can also be undervolted from my understanding.
Wondering why max speed on BIOS DUMP says 8300MHz (not the first time I've seen this).



The Prime95 error can often come from unstable CPU configs, which is why i had to clarify.
I believe this error CAN also be thrown out by RAM.
Would it be best to load BIOS and load optimized defaults just in case? or am I just suggesting stupid things xD
 

Colif

Titan
Moderator
turbo boost of 5ghz, thats a neat trick

i think thats a bios default max speed, as I can see it all over the place but no one is questioning it.

EHarr3, Thank you for posting in the Intel® Communities Support.

Just to let you know, some models of motherboards have a setting in the BIOS that is enabled by default and that overclocks the processor base frequency, the Intel® Core™ i5-9600KF has 3.70 GHz of clock speed and 4.60 GHz when Max Turbo Frequency gets activated as you can see here:
https://ark.intel.com/content/www/us/en/ark/products/190884/intel-core-i5-9600kf-processor-9m-cache-up-to-4-60-ghz.html

So, the best thing to do will be to get in contact directly with ASRock for them to verify the settings in the BIOS and make sure there is nothing in there forcing the processor to run at speeds it does not support:

Once you do that please install the Intel® Processor Diagnostics Tool, it will run a test on the unit and it will provide a report where you will be able to verify the clock speed detected, if it passed the test it should be working fine:
https://downloadcenter.intel.com/download/19792/Intel-Processor-Diagnostic-Tool

In reference to the question about the warranty, normally, most of the stores have a 30 day warranty period for replacement on their products, but you will need to check that information directly with the place of purchase, depending which product is the one that got defective, that will be the fastest way in the case there is a hardware failure with any of the components of your platform.

Regards,
Alberto R.
https://forums.intel.com/s/question/0D50P00004KjQSJ/i-need-to-help-identifying-the-source-of-the-default-clock-ratio-being-8300mhz?language=pt_BR
 
Reactions: REVLX and PC Tailor

Colif

Titan
Moderator
My take on 8300 MHZ is its just an upper limit set for all Intel bios. I don't think it has anything to do with this problem. I don't think anyone has managed to overclock a CPU to that speed, even with extreme overclocking techniques, so the speed might just be set at that as max for all Motherboards... its not something I have seen before.

I am not sure what the problem you are having is, though if you can't complete the Prime95 scan without a BSOD it could be the CPU still...
 
Mar 14, 2019
33
1
35
0
My take on 8300 MHZ is its just an upper limit set for all Intel bios. I don't think it has anything to do with this problem. I don't think anyone has managed to overclock a CPU to that speed, even with extreme overclocking techniques, so the speed might just be set at that as max for all Motherboards... its not something I have seen before.

I am not sure what the problem you are having is, though if you can't complete the Prime95 scan without a BSOD it could be the CPU still...
Hmm see I didn't get a BSOD during the test it just reported it as an error, but it doesn't provide like any further advice apart from running a memory test which I've already done
 

Colif

Titan
Moderator
I don't think it was running and the log most likely ended there. Good news and bad news tho I just ran a stress test with prime 95 and in one of the tests it said hardware failure detected, rounding was 0.5 when the expected needed to be less than 0.4. I believe this was worker #3 so 2nd or 3rd core.
See, this is what I get for not noticing...

The algorithm, in a rough sense, uses a transformation that produces floating-point results when a precise answer is an integer. So, each decimal result is rounded to the nearest integer, and the rounded-off decimal is a "rounding error" in the FFT algorithm. When results come out below nnn.4 or above mmm.6, it is obvious which way to round and the rounding error is considered "safe". But if a calculation comes out nnn.48, is the error 0.48 and should be rounded down, or is the error 0.52 and should be rounded up? The calculation is no longer trustworthy.

When the algorithm produces an error above 0.4, it repeats the particular calculation that produced the error. if the error is repeatable, it is a sign that the precision of the transformation is not high enough for the size of calculation being attempted, and a slower/larger/more precise transformation is then used to reduce error below the 0.4 cutoff. However, if the error is non-repeatable, the computer has a hardware error- calculations are no longer deterministic. This is where you're at.

Slow down the overclock, reduce heat (by cleaning out dust, improving airflow, routing cables better, reseating CPU heatsink with less thermal paste, moving the GPU a slot or two away from CPU, etc....), check memory for errors
https://www.mersenneforum.org/showthread.php?t=24129

I see why overclock mentioned. many of the people with error are in fact overclocked.

there should be a file called stress.txt in either prime95 folder or documents, it might reveal more about the error.
 
Mar 14, 2019
33
1
35
0
See, this is what I get for not noticing...



https://www.mersenneforum.org/showthread.php?t=24129

I see why overclock mentioned. many of the people with error are in fact overclocked.

there should be a file called stress.txt in either prime95 folder or documents, it might reveal more about the error.
Yeah that's what I read but it was just generic do a memtest etc, I mean I might re read it to see if I missed something
 

Colif

Titan
Moderator
If it was me, I would take PC to a repair place and ask them to test it as something isn't right, and they should have parts they can substitute into place (known working parts) so as to test what remains. and narrow down on what is broken.

One of the fixes for the Prime error is voltages on CPU, its possible its the PSU at fault, but instead of wasting money to confirm that blindly, I would get a shop to make sure. I don't like telling people its X without being sure. Currently I am not.

too many could be's
 
Reactions: REVLX
Mar 14, 2019
33
1
35
0
If it was me, I would take PC to a repair place and ask them to test it as something isn't right, and they should have parts they can substitute into place (known working parts) so as to test what remains. and narrow down on what is broken.

One of the fixes for the Prime error is voltages on CPU, its possible its the PSU at fault, but instead of wasting money to confirm that blindly, I would get a shop to make sure. I don't like telling people its X without being sure. Currently I am not.

too many could be's
I don't know if this is relevant but ever since this pc was built, when it turns on it always lights up then lights off almost like it turns itself off and then turns back on completely.
 

Colif

Titan
Moderator
One solution of the Prime95 thing is increase voltages to CPU - I am not a hardware guy, I can't tell you how to do that or how much.

Rounding errors in Prime95 usually are caused by low CPU or RAM voltage. Or incorrect ram timings or speed.


The reason for the ram voltage increase is that Gigabyte has a issue of out putting lower voltage for the ram.

Example I put 1.500V in for memory voltage but only get 1.488. If I change that to 1.525 I end up with 1.504 Volts. So the step up in voltage should help with memory speed.

it all sounds like power of some sort. that start up act is interesting.
 
Mar 14, 2019
33
1
35
0
One solution of the Prime95 thing is increase voltages to CPU - I am not a hardware guy, I can't tell you how to do that or how much.



it all sounds like power of some sort. that start up act is interesting.
It could be due to me turning off the plug at night, I don't know what to increase the voltage to in my case since he mentions that if it was 1.5 for that that other guy to change it to 1.525, but Bios says my ram is on 1.2V on default and CPU is 1.020V on default
 

Colif

Titan
Moderator
I would get it looked at, if the PC acts like that at startup for a repair place, they should at least try to fix that. That isn't normal. We tested almost everything in PC apart from motherboard and PSU and those 2 are difficult to do without spare parts. Tell them the results of Prime 95, if they clued in they might know how to fix it.

I wouldn't change voltages without knowing what I was doing... I haven't done it myself so am not prepared to tell you to do it.. though it might be perfectly safe. I don't know. I gave you the quotes as examples, not as suggestions :)

I had hoped someone else would comment in here after asking for help yesterday but so far not.
 
Reactions: REVLX

ASK THE COMMUNITY

TRENDING THREADS