Burnin test "No operations reported for timeout period" also "bad_module_info"

[RainingTacco]

So i got weird behaviour in games where my display would sometimes go black and then turn on. Sometimes it would turn black and say "no signal", sound is still playing, and the one thing i can do is to reset the pc. Sometimes the game would just crash without any error log. Some games run fine. I see that before crashes i either have dwm.exe error, radeonsoftware.exe error, or amdkmdap driver error in event viewer.I don't have any BSODs or memory related errors in event log[maybe you could tell how they look like, because maybe im missing something?]. I had only one BSOD, when i've messed with Custom Resolution Utility, which happened when i've changed pixel clock too much, and tried to run Ghost Recon Breakpoint[which worked rock solid before that].This leads me to believe its GPU issue, but i wanted to rule out other things like CPU, RAM and PSU.

I have:
Ryzen 3600
2x8gb Patriot Viper Steel 3600 mhz CL17
MSI B450M Mortar Max
Gigabyte 5700XT Gaming OC
Everything stock except XMP setting on[i had to input some settings manually, because the default XMP settings were not the same as advertised]

So i've tried the prime95 test for an hour to stress test CPU and RAM:
No error, no warnings, no crashes, nothing in event viewer.
I've tried memtest86 with 4 passes[400 coverage] =no errors
I've tried memtest86 with 16 passes[1600 coverage, crazy over 13 hours] =no errors
I've tried memtest86+ with one pass =no error
I've tried HCI memtest with 6 workers 2000mb each for 100% coverage = no errors
I've tried memtest64 for 30 minutes for no errors
Ive tried windows memory checking utility =no errors

So i've also tried the BurnInTest from your company and this is what i got:
https://imgur.com/a/QiQKDVQ
I can run this test many times, and sometimes it breaks earlier like in 8 minutes and sometimes later like in 12 minutes[sometimes no at all, but ive only tested for 15 minutes each, so it could have crashed later]. The MBytes written/verified stops rising after some time, and at the same time "bad_module_info" event happen. After 6 minutes, the program throws error, and i end the test which shows the "No operations reported for timeout period". My question is, do i have a problem, or is your program borked?

[RainingTacco]

I forgot. I have Corsair RMx 850W PSU[2018 edition]. I've ran furmark with prime95 for 1 hour and occt power test for one hour and had no problems, crashes, freezes, black screens etc. So it seems it's not PSU. Since my post i also tried google stress app test testing my RAM for 1 hour with -W parameter[more heavy on controller] and it PASSed. Im going to test it overnight to see if there are any problems.

[David (PassMark)]

It looks like the memory testing process crashed.

Can you send us a debug log from BurnInTest so that we can take a look.

[RainingTacco]

It looks like the memory testing process crashed.

Can you send us a debug log from BurnInTest so that we can take a look.

I will definitely send it in friday or saturday, i don't have the access to the PC right now. I've done additional tests -14 hours of google stress app test, and then 14 hours of prime95 blend test, and had no errors, so im thinking that the problem is 99% on your side .

[David (PassMark)]

Also make sure you are using the latest BurnInTest release
https://www.passmark.com/products/bu...t/download.php
Version 9.1 Build 1001 at the moment.

[RainingTacco]

Also make sure you are using the latest BurnInTest release
https://www.passmark.com/products/bu...t/download.php
Version 9.1 Build 1001 at the moment.

Where are the log files stored?

[David (PassMark)]

See,
https://www.passmark.com/support/bitdebug.php

[RainingTacco]

Ok here's the log made with the newest version of the program (1001 revision)

[David (PassMark)]

So that is strange.
Maybe the hardware isn't stable enough to run the RAM test.
(i.e. the memory holding the program code gets corrupted causing a crash of the testing process).

Can you try Memtest86 instead. The program has a much smaller memory foot print.
https://www.memtest86.com/

[RainingTacco]

So that is strange.
Maybe the hardware isn't stable enough to run the RAM test.
(i.e. the memory holding the program code gets corrupted causing a crash of the testing process).

Can you try Memtest86 instead. The program has a much smaller memory foot print.
https://www.memtest86.com/

In my OP ive said that ive run the memtest86 first with 4 passes and then with 16 passes, and had no error. BurninTest can sometimes run for the 15 minutes on 100% memory test and pass, and sometimes it gets this "no operations reported for timeout". I don't get crashes in other programs, dont get BSODs, dont have memory related errors in event viewer. If 14 hours of prime95 didn't throw errors, then i think its stable enough for me. It might be problem on your side or on my side.
What kind of other errors your program can throw, when testing memory? Shouldn't it continue when it encounters error to find all the errors? If the memory gets corrupted ie. particular cell and the program can't continue because of that, shouldn't the error message say about that, instead of throwing "no operations reported"? I mean i understand that "no operations are reported for timeout period" because it literally stops writing to memory[program doesn't crash or hangs!], then after the 6 minutes the error is thrown. Why does it stops writing to memory? And if it stops why it doesn't show more detailed reason like "this memory cell is corrupted and couldn't be accessed for read/write cycle".

[RainingTacco]

Oh and im running the test on 100%[100% duty cycle?], while browsing chrome, can this influence the test? Duty cycle means that it tries to use 100% of resources? Meaning it wants to use all memory for test?

[David (PassMark)]

Can you on Level 2 debugging (from the Preferences window) and create another log.

[RainingTacco]

Can you on Level 2 debugging (from the Preferences window) and create another log.

Hi. Since i can't attach .trace file on the forum, and i won't upload it as .zip file[since you could say that you don't open unverified zip files] i've uploaded it to external site:
https://gofile.io/?c=nMg3qS
If something is wrong with the upload, please contact me and i will try to fix it.

[RainingTacco]

Ok i've disabled the XMP running the 2133mhz stock settings, as the last resort, tried the very same test, and still got the same error. Here's the log
https://gofile.io/?c=F4Giq3
So it's either my memory is broken and i should RMA it, or your program is wrong. I really need some answers, because i don't know what to do

[RainingTacco]

Ok, most peculiar thing. I've used Ryzen Dram Calculator, used the fast option, ive entered timings in the bios, and somehow after 4 or 5 tests i no longer get the error. Weird. I have yet to test if the system is stable.