weird problem - identify the source ?

[MCCOPRA]

Hi, I have been using the system in my signature.

 

My problem is at random times i get Black Screen Restart with no Bluescreen.

 

Sometimes systems auto restart and others i need to manually restart.

 

I usually get 1-2 day free and later. i crash system.

 

I test my system every day first boot with 3d Mark Firestrike Ultra Test

 

What i Tried -

 

Windows 10 with all updates - multiple nvidia drivers

W10 with no updates -  multiple nvidia drivers

HB Bridge

LED Bridge

1 Ribbon

2 Ribbon

GPU Stock Bios

GPU OC Bios

GPU Overclock

GPU Downclock

850w Gold PSU

1000w Platinium PSU

CPU Overclock

 

Same happened with GTX 1070 SLI Setup too

 

1 Stick of Ram to 8 Stick of ram 8gb HyperX

Fresh Install of Windows 7 with Multiple Nvidia Drivers.

Re-Flashed Motherboard Bios. (Currently)

Stock Motherboard Settings No OC the CPU (currently)

 

Currently Waiting for tomorrow to test the system if anyone has any ideas please Share

 

if it crashes again prob will get a new mobo and try or a 6800k or get rid of system and get a Ryzen 1700X + x370 Mobo?

 

[FIXXX]

No BSOD or GPU artifacting smells like a possible PSU issue (note that not only big load can cause it it fail, but a longer session of use). That's what i'd try to replace and test.

[MCCOPRA]

1 minute ago, FIXXX said:

No BSOD or GPU artifacting smells like a possible PSU issue (note that not only big load can cause it it fail, but a longer session of use). That's what i'd try to replace and test.

Tried with 850 Watt Gold psu was having same problems and upgraded to a 1000W Corsair 80 plat psu an same error.

[Olly1923]

Hi,

I'm not saying that it is but could it be a faulty motherboard, seems weird that it is only happening a once every few days.

I would also check the hard drive that you are installing the operating system on, also change the SATA port that you have connected it to and the cable.

 

[MCCOPRA]

9 minutes ago, Olly1923 said:

Hi,

I'm not saying that it is but could it be a faulty motherboard, seems weird that it is only happening a once every few days.

I would also check the hard drive that you are installing the operating system on, also change the SATA port that you have connected it to and the cable.

 

yes my next action will prob be a new motherboard (Thinking about  X99A XPOWER GAMING TITANIUM)

also the system is on M.2 PCI-E Samsung 950 Pro drive formatted it 4 times with different w10 and 7 current systems reseated it couple times.

[tlink]

try unplugging frontpanel connectors and io etc and see if it still happens, also check if the front panel headers on the mobo are clean and not connected in any way. if it still happens my bet is that somethings either wrong with the motherboard or some component is doing really weird stuff with voltages. you can try checking the window logs but i doubt you'll find anything there since it just flat out stops the machine. no blue screen means no kernel crash which highly probably is hardware error that doesn't go trough the operating system. have you tried running on a single gpu or just a different gpu alltogether? what are your temps?

[Olly1923]

Could it possibly be the motherboards support for the SSD, I know that I have had this problem before but on other systems.

[MCCOPRA]

Just now, tlink said:

try unplugging frontpanel connectors and io etc and see if it still happens, also check if the front panel headers on the mobo are clean and not connected in any way. if it still happens my bet is that somethings either wrong with the motherboard or some component is doing really weird stuff with voltages. you can try checking the window logs but i doubt you'll find anything there since it just flat out stops the machine. no blue screen means no kernel crash which highly probably is hardware error that doesn't go trough the operating system. have you tried running on a single gpu or just a different gpu alltogether? what are your temps?

The Case is new Clean Corsair 750D - so front panel connections are clean and not damaged.

 

Tried it with gtx1070 - 1070 Sli - single 1080 and 1080 sli same.

 

temps are for mobo max 50  - cpu max 80 - gpus max 80

 

i was suspecting hd audio onboard but even does same with xonar dgx card and onboard disable.

[Olly1923]

Can you try using the system with a gpu that does not need power form the supply just to see if that stops it ?

[tlink]

4 minutes ago, MCCOPRA said:

The Case is new Clean Corsair 750D - so front panel connections are clean and not damaged.

 

Tried it with gtx1070 - 1070 Sli - single 1080 and 1080 sli same.

 

temps are for mobo max 50  - cpu max 80 - gpus max 80

 

i was suspecting hd audio onboard but even does same with xonar dgx card and onboard disable.

well all thats left in my opinion is trying a different motherboard. honestly can't think of anything else unless its extremely fringe.

[MCCOPRA]

9 minutes ago, Olly1923 said:

Could it possibly be the motherboards support for the SSD, I know that I have had this problem before but on other systems.

the Samsung 950 m.2 is in mobos supported list so might not be a problem and also i can even try pci mode or the pch sata express lines on mobo so run the m.2

[Olly1923]

Just now, tlink said:

well all thats left in my opinion is trying a different motherboard. honestly can't think of anything else unless its extremely fringe.

Unless his motherboard just likes going to sleep for a bit every 2 days hahaha

[MCCOPRA]

1 minute ago, Olly1923 said:

Can you try using the system with a gpu that does not need power form the supply just to see if that stops it ?

i have a R7 240 lying around can try with it.

[Olly1923]

Just now, MCCOPRA said:

i have a R7 240 lying around can try with it.

Try that, just so see if it does it. Make sure you uninstall all of nvidias drivers with a tool so that they cant interfere.

And then get back to us.  

[MCCOPRA]

3 minutes ago, tlink said:

well all thats left in my opinion is trying a different motherboard. honestly can't think of anything else unless its extremely fringe.

i might get  X99A XPOWER GAMING TITANIUM next week if tomorrow i can sell the other system i built.

[MCCOPRA]

Another Theory of mine is that with overclock and dual cards.

 

is there a voltage protection mechanism in the board if the gpu s tries to pull more watt from system or even the overclock of the cpu might get the vrms to reset point since i run aio they might not get enough airflow.

 

also since my lovely cpu is capable of 4.6ghz with 1.335 can can handle up to 4.8 1.425v it may be a time for X99A XPOWER GAMING TITANIUM board since i can still use the MSI HB L BRIDGE

[Olly1923]

What power supply if it is a corsair one you should be able to monitor voltages and wattages going to the cards and it could give you (Us) more of a inside view. 

[MCCOPRA]

2 hours ago, Olly1923 said:

What power supply if it is a corsair one you should be able to monitor voltages and wattages going to the cards and it could give you (Us) more of a inside view. 

its the RM1000i with link software the voltages looks fine all around but starting from today i will log all cpu gpu and psu activities till it crashes 

[MCCOPRA]

Log Day 1 - System Passed the 3d Mark Firestrike Ultra Stress test 3 times back to back

after re-flashing bios and not touching bios. gpus are not overclocked and using their own overclock official bios.

 

so far fine. will keep you updated for 7 days and see.

[plus]

weird problems might be caused by faulty memory.

did memtest86 find anything unusual?

[MCCOPRA]

1 hour ago, plus said:

weird problems might be caused by faulty memory.

did memtest86 find anything unusual?

tested with memtest86+ - tested with windows memory diagnostics - no error - all pass

tried with 1 - 2 - 4 and 8 sticks same error.

 

now running flashed bios - default settings. looks like day 1 safe will see in 7 days

[MCCOPRA]

Log Day 2 - Started the pc run the 3d mark firestrike ultra stress test left it for run - come back after 1hr and ta da black screen with computer running but frozen.

 

so what is the problem Get a new mobo ?

 

get new ryzen 1700 + new mobo ?

 

i am confused now.

[MCCOPRA]

Log Day 2 update - Some friend suggested maybe 4k monitor or the display port cable is faulty tested

 

monitor on different system no crash - changed the displayport cable with a proper new gold plated one and bam still crashed full load on firestrike ultra at stock bios and settings.

 

any ideas ?

[MCCOPRA]

Log Day 3 - I can play most games fine but when i load system to 100% i crash the system with no blue screen.

 

changed the display port cable. and still get got couple screen freezes - no bluescreens.

 

What to do now - new mobo or new system could it be cpu or m.2 ?

[Jumper118]

I am going to say it's a power issue and because you have tried another psu already the motherboard would be the next thing that the power goes through. The other things is try each gpu 1 by 1