Greetings from Greece. First of all, sorry in advance for the huge post! I am just an average user but my budget is really tight and i can't afford or trust any local service stores (the fact that i bought most parts from different stores does not help either), so i will need all the help i can get to identify the issue myself before proceeding to a store and avoid it altogether if possible. I really hope you can help me and thanks in advance all those who make the effort to read it and troubleshoot.
I decided to build a new pc using some spare old parts and buying some new.
The parts list is;
Asus Prime a320m-k (new),
AMD A8-9600 (new),
Corsair CX 500 (used),
HDD WD1200JD (sata II, used),
Corsair Vengeance LPX 8GB DDR4-3000MHz (CMK8GX4M2B3000C15 ,new).
I have downloaded and installed the latest motherboard firmware.
At first i had bought different ram {1x Crucial Ballistix Sport LT 4GB DDR4-2400MHz (BLS4G4D240FSB) -new}. I had many problems trying to install windows 10. They were mainly caused by iGPU drivers as i assumed after research with various BSOD's after trying to install any chipset or video drivers, usually resulting to atikmdag.sys error or critical process died error. After many many trials i eventually managed to install windows properly on an external usb hard drive and then managed to copy the image from the external HDD to my internal WD1200JD. It seemed to work fine, but i was really worried with all these failed attempts; it just seems i got lucky once and if i ever had a problem i would end up with serious trouble trying to reinstall windows from scratch. So i contacted asus and their only response was that my ram is not in their QVL and that was the reason for my problems. Of course they were not able to provide info why it's not in QVL; because it's not tested or because it's tested and not compatible. I started researching with crucial website that has a compatibility app and found out that even though it said that a single 4gb BLS4G4D240FSB is not compatible, a kit of 2x4gb of the same ram IS listed as compatible. How strange... I talked with a crucial compatibility representative through chat and after a long discussion we concluded that my ram should be compatible with this mobo. I also ran memtest86 and had zero errors after 5 passes. Even so i decided not to mess with asus, since i could have a defective product and being stuck to their "not included in QVL" response. I bought the above mentioned corsair 2x4gb kit, which is in their QVL and can run at 3000Mhz when i upgrade to ryzen- the only downside is that with the a8 will run just at JEDEC 2133 XMP.
Unfortunately after installing the new ram things got even worse. First i tried to use my already installed windows 10, but i had a strange problem. When booting to windows, just as the windows logo begun to appear i would see some blue pixels flashing at some parts of the screen. When the windows logo appeared 4 pixels would lock in specific parts of the screen. Inside windows the system would randomly freeze every now and then (every few seconds or minutes) with various coloured lines appearing on my screen and my monitor restarting. Sometimes this would lead to system crash altogether (can't remember Bsod reason). I decided to do a fresh windows 10 install again. It seemed a little easier this time with less BSOD's (and less fresh install trials) when trying to install all the appropriate chipset and graphics drivers, but in the end all the times that i managed to correctly install all the drivers (i tried again and again fresh installing in order to do that with zero crashes-bsod's) the same freezing problem insisted to occur (with the zero crash installations). One of the BSOD's would be "attempted write to readonly memory" (it is one of the two single dump files i have managed to keep all these days). The pixel problem at boot screen seemed to persist with every fresh install, it even appeared when booting from usb stick to install windows. Same problems both for HDMI and VGA connection tested.
I decided to run memtest86 again and from the 1st pass it found something like 18 errors in hammer test. So i decided to end the test there and start testing each module individually. I started testing 1st module 2. At slot 1 it ended with 49 and 56 errors in hammer test at 4 passes. At slot 2 it ended with 35 errors at 3 passes and 364 errors at 23 passes (i left it testing longer last time just to check liability for other errors after long time use), all at hammer test. I took out the 1st module and inserted the 2nd. My pc would now even see it now, it was just dead. (I took my kit back to the store and confirmed that, see below) Since it was new year and stores were closed for another 2 days i had a lot of time trying to work things around with my remaining working 4gb module.
I tried again installing windows, only this time, it was impossible. Every time i tried installing them, i could only get to the part of the first restart. Windows had copied and setup all files (i had it disconnected from network for updates) and seemed to be almost ready. It would self reboot after 10 seconds, but after showing the windows logo for just one second it would reboot in a loop. I tried many times (the uefi boot order was ok i checked always and tried all boot options). No recovery options screen would emerge from HDD. Whenever i tried to boot from installation usb and auto-repair or reset windows it failed. I tried using usb stick to other ports, same results. I downloaded windows 10 many times with MediaCrationTool, same results. I tried doing installation at external hdd like before, with windows iso and wintousb, same results.
BSOD's at this point were System_Service_Exception dxgkrnl.sys sometimes, sometimes CRITICAL_PROCESS_DIED, sometimes 0xC0000021a and sometimes UNEXPECTED STORE EXCEPTION. (Maybe critical process died and unexpected store exception happened before, when i had the 2 rams kit "working", i can't remember for sure). I tried every troubleshooting guide i have found for these errors with no luck. After all those fails i tried to check what was written after fresh installing windows 10 on the HDD and i could not find any boot, system or windows folder, which could explain i think why sfc scan, dism or fixing the MBR could not solve any of my issues. Scandisk /f /r was always with zero errors.
I have tried all the above both using GPT and MBR when installing and using the appropriate UEFI settings (UEFI when doing GPT and legacy when doing MBR insallation, both for the usb drive and the HDD). I even tried all different settings each time, problems remained.
After all that i stopped trying to install windows and before returning my ram kit for replacement i decided to do some further hardware checks with ultimate boot cd. I used memtest86+ (not memtest86) with zero errors, i then used windows memory diagnostic and extended test and it failed at MATS+ (cache disabled) with many many errors (it did not say any number but i saw it live failing most if not all expected values).
I then tested my HDD with WD data lifeguard diagnostics dosdlg 5.27 extended test, no errors. The only alarming about my HDD was when it used HDDscan program when my system seemed to work properly, it was green in SMART test everywhere except the UltraDMA CRC Errors; Value 200, Worst 253, Threshold 0. But i imagine this count could be due to my previous old pc it was hosted, that died after 5 years of heavy use.
Next i tried running cpustress. I had it running mersene prime test 28.5 at setting "2" (for cpu and lots of memory test and psu heavy load) simultaneously with cpustress 2.0 and system stability tester, all running for about 14 hours with no crash (actually system stability tester must have finished sooner but it was with no errors, settings 64 turns, max threads-can't remember value, i think 64- and highest possible digits -can't remember value either, i think 1 or 16million-, zero errors). It could not use sensors command to watch CPU temp but when i rebooted after 14 hours of testing, the temp in UEFI was 39 Celsius.
Eventually i decided to return the ram modules and now i am waiting for the new ones to arrive in 5 days. So my questions are:
A. Is it ok if i find any hammer test errors on the new ram i will recieve?
B. Are the problems i described consistent with faulty memory modules?
C. Is it possible that my APU is the problem? Cpustress results seemed ok, but did it also check the iGPU cores? If iGPU cores were not tested, is there a way to test them? Should i run more tests for the "normal" CPU cores just to be safe?
D. Is it possible that my motherboard is the problem? Could it be the bus controller? Is there any way to test the motherboard?
E. Is it possible my hard drive fails and if so, is it a good idea to buy a teamgroup L7 EVO SSD 60GB just to have my OS there until i can raise some money for a better and larger SSD? (it is currently the cheapest choice i have at just 32 Euros). Do i also have the alternative to buy a sandisk extreme 64GB usb 3.0 (not the GO version) and have windows there? Is it supported by that UFD and will it keep the temperature at safe levels when using it many hours as a boot drive?
F. Is it ok to use windows 10 without activation for some time? I have not bought a key yet since nothing seems to work and i am on budget. Does it affect system stability and driver issues i encountered?
G. Should i be trying to install different OS like Linux if problems persist? (i have never used Linux before) I was thinking of Ubuntu game pack. My fear is about it's compatibility with my APU (drivers and stuff).
I hope someone manages to read all this and make some sense! Please help!!!
-- EDIT --
One thing that might be an important indicator and i forgot to mention is the info of the dram slots in UEFI environment. First time i noticed that when i had both ram modules "working" if i went on UEFI to see the info about them it would show up correct values on the main page and if i went in tools sections in asus spd information and selected the dram section it would show info correctly for the first slot (DDR4 - 2133MHZ and so on) but it would show nothing on the second slot, like it was empty (all 8gb showing up on main screen though). After i took out the one module (that ended up now working at all anyway) whenever i had the working one on slot 1 it would show up on all screens ok. Whenever i had it working on the 2nd slot it would show correct values in the main menu screen but it would not show correct values on the tool-spd info section; When selecting slot 2 to see the info i would get DDR3 (not even supported on this mobo) and wrong speed-something like 1*** with some more numbers i could not understand. All this info for that 2nd slot was showing up not aligned on the fields and it was blurry.
H. Is this an indicator of faulty motherboard? Is it possible my motherboard actually harmed the ram? Should i be cautious when i get the new ram or is it safe to insert it and continue testing?
I decided to build a new pc using some spare old parts and buying some new.
The parts list is;
Asus Prime a320m-k (new),
AMD A8-9600 (new),
Corsair CX 500 (used),
HDD WD1200JD (sata II, used),
Corsair Vengeance LPX 8GB DDR4-3000MHz (CMK8GX4M2B3000C15 ,new).
I have downloaded and installed the latest motherboard firmware.
At first i had bought different ram {1x Crucial Ballistix Sport LT 4GB DDR4-2400MHz (BLS4G4D240FSB) -new}. I had many problems trying to install windows 10. They were mainly caused by iGPU drivers as i assumed after research with various BSOD's after trying to install any chipset or video drivers, usually resulting to atikmdag.sys error or critical process died error. After many many trials i eventually managed to install windows properly on an external usb hard drive and then managed to copy the image from the external HDD to my internal WD1200JD. It seemed to work fine, but i was really worried with all these failed attempts; it just seems i got lucky once and if i ever had a problem i would end up with serious trouble trying to reinstall windows from scratch. So i contacted asus and their only response was that my ram is not in their QVL and that was the reason for my problems. Of course they were not able to provide info why it's not in QVL; because it's not tested or because it's tested and not compatible. I started researching with crucial website that has a compatibility app and found out that even though it said that a single 4gb BLS4G4D240FSB is not compatible, a kit of 2x4gb of the same ram IS listed as compatible. How strange... I talked with a crucial compatibility representative through chat and after a long discussion we concluded that my ram should be compatible with this mobo. I also ran memtest86 and had zero errors after 5 passes. Even so i decided not to mess with asus, since i could have a defective product and being stuck to their "not included in QVL" response. I bought the above mentioned corsair 2x4gb kit, which is in their QVL and can run at 3000Mhz when i upgrade to ryzen- the only downside is that with the a8 will run just at JEDEC 2133 XMP.
Unfortunately after installing the new ram things got even worse. First i tried to use my already installed windows 10, but i had a strange problem. When booting to windows, just as the windows logo begun to appear i would see some blue pixels flashing at some parts of the screen. When the windows logo appeared 4 pixels would lock in specific parts of the screen. Inside windows the system would randomly freeze every now and then (every few seconds or minutes) with various coloured lines appearing on my screen and my monitor restarting. Sometimes this would lead to system crash altogether (can't remember Bsod reason). I decided to do a fresh windows 10 install again. It seemed a little easier this time with less BSOD's (and less fresh install trials) when trying to install all the appropriate chipset and graphics drivers, but in the end all the times that i managed to correctly install all the drivers (i tried again and again fresh installing in order to do that with zero crashes-bsod's) the same freezing problem insisted to occur (with the zero crash installations). One of the BSOD's would be "attempted write to readonly memory" (it is one of the two single dump files i have managed to keep all these days). The pixel problem at boot screen seemed to persist with every fresh install, it even appeared when booting from usb stick to install windows. Same problems both for HDMI and VGA connection tested.
I decided to run memtest86 again and from the 1st pass it found something like 18 errors in hammer test. So i decided to end the test there and start testing each module individually. I started testing 1st module 2. At slot 1 it ended with 49 and 56 errors in hammer test at 4 passes. At slot 2 it ended with 35 errors at 3 passes and 364 errors at 23 passes (i left it testing longer last time just to check liability for other errors after long time use), all at hammer test. I took out the 1st module and inserted the 2nd. My pc would now even see it now, it was just dead. (I took my kit back to the store and confirmed that, see below) Since it was new year and stores were closed for another 2 days i had a lot of time trying to work things around with my remaining working 4gb module.
I tried again installing windows, only this time, it was impossible. Every time i tried installing them, i could only get to the part of the first restart. Windows had copied and setup all files (i had it disconnected from network for updates) and seemed to be almost ready. It would self reboot after 10 seconds, but after showing the windows logo for just one second it would reboot in a loop. I tried many times (the uefi boot order was ok i checked always and tried all boot options). No recovery options screen would emerge from HDD. Whenever i tried to boot from installation usb and auto-repair or reset windows it failed. I tried using usb stick to other ports, same results. I downloaded windows 10 many times with MediaCrationTool, same results. I tried doing installation at external hdd like before, with windows iso and wintousb, same results.
BSOD's at this point were System_Service_Exception dxgkrnl.sys sometimes, sometimes CRITICAL_PROCESS_DIED, sometimes 0xC0000021a and sometimes UNEXPECTED STORE EXCEPTION. (Maybe critical process died and unexpected store exception happened before, when i had the 2 rams kit "working", i can't remember for sure). I tried every troubleshooting guide i have found for these errors with no luck. After all those fails i tried to check what was written after fresh installing windows 10 on the HDD and i could not find any boot, system or windows folder, which could explain i think why sfc scan, dism or fixing the MBR could not solve any of my issues. Scandisk /f /r was always with zero errors.
I have tried all the above both using GPT and MBR when installing and using the appropriate UEFI settings (UEFI when doing GPT and legacy when doing MBR insallation, both for the usb drive and the HDD). I even tried all different settings each time, problems remained.
After all that i stopped trying to install windows and before returning my ram kit for replacement i decided to do some further hardware checks with ultimate boot cd. I used memtest86+ (not memtest86) with zero errors, i then used windows memory diagnostic and extended test and it failed at MATS+ (cache disabled) with many many errors (it did not say any number but i saw it live failing most if not all expected values).
I then tested my HDD with WD data lifeguard diagnostics dosdlg 5.27 extended test, no errors. The only alarming about my HDD was when it used HDDscan program when my system seemed to work properly, it was green in SMART test everywhere except the UltraDMA CRC Errors; Value 200, Worst 253, Threshold 0. But i imagine this count could be due to my previous old pc it was hosted, that died after 5 years of heavy use.
Next i tried running cpustress. I had it running mersene prime test 28.5 at setting "2" (for cpu and lots of memory test and psu heavy load) simultaneously with cpustress 2.0 and system stability tester, all running for about 14 hours with no crash (actually system stability tester must have finished sooner but it was with no errors, settings 64 turns, max threads-can't remember value, i think 64- and highest possible digits -can't remember value either, i think 1 or 16million-, zero errors). It could not use sensors command to watch CPU temp but when i rebooted after 14 hours of testing, the temp in UEFI was 39 Celsius.
Eventually i decided to return the ram modules and now i am waiting for the new ones to arrive in 5 days. So my questions are:
A. Is it ok if i find any hammer test errors on the new ram i will recieve?
B. Are the problems i described consistent with faulty memory modules?
C. Is it possible that my APU is the problem? Cpustress results seemed ok, but did it also check the iGPU cores? If iGPU cores were not tested, is there a way to test them? Should i run more tests for the "normal" CPU cores just to be safe?
D. Is it possible that my motherboard is the problem? Could it be the bus controller? Is there any way to test the motherboard?
E. Is it possible my hard drive fails and if so, is it a good idea to buy a teamgroup L7 EVO SSD 60GB just to have my OS there until i can raise some money for a better and larger SSD? (it is currently the cheapest choice i have at just 32 Euros). Do i also have the alternative to buy a sandisk extreme 64GB usb 3.0 (not the GO version) and have windows there? Is it supported by that UFD and will it keep the temperature at safe levels when using it many hours as a boot drive?
F. Is it ok to use windows 10 without activation for some time? I have not bought a key yet since nothing seems to work and i am on budget. Does it affect system stability and driver issues i encountered?
G. Should i be trying to install different OS like Linux if problems persist? (i have never used Linux before) I was thinking of Ubuntu game pack. My fear is about it's compatibility with my APU (drivers and stuff).
I hope someone manages to read all this and make some sense! Please help!!!
-- EDIT --
One thing that might be an important indicator and i forgot to mention is the info of the dram slots in UEFI environment. First time i noticed that when i had both ram modules "working" if i went on UEFI to see the info about them it would show up correct values on the main page and if i went in tools sections in asus spd information and selected the dram section it would show info correctly for the first slot (DDR4 - 2133MHZ and so on) but it would show nothing on the second slot, like it was empty (all 8gb showing up on main screen though). After i took out the one module (that ended up now working at all anyway) whenever i had the working one on slot 1 it would show up on all screens ok. Whenever i had it working on the 2nd slot it would show correct values in the main menu screen but it would not show correct values on the tool-spd info section; When selecting slot 2 to see the info i would get DDR3 (not even supported on this mobo) and wrong speed-something like 1*** with some more numbers i could not understand. All this info for that 2nd slot was showing up not aligned on the fields and it was blurry.
H. Is this an indicator of faulty motherboard? Is it possible my motherboard actually harmed the ram? Should i be cautious when i get the new ram or is it safe to insert it and continue testing?
Facebook
Twitter
Instagram