Ok so i just built a new pc and sometimes it just crashes,sometimes im gaming others im not,sometimes the GPU fan revs up some

[alucardnz]

HELP. GPU keeps crashing.

Ok so i just built a new pc and sometimes it just crashes,sometimes im gaming others im not,sometimes the GPU fan revs up sometimes it doesnt.And i have to reboot.

case NZXT H440W NE

motherboard Asus Maximus VIII Formula

CPU Intel Core i7 6700K 4.0GHz

RAM G.Skill Ripjaws 4 Blue DDR4 PC20000MHz 32g

water cooler NZXT Kraken X61

GPU Gigabyte GeForce GTX 1080 Founders Edition

power supply EVGA SuperNOVA G1 650W

hard drives,500G SSD,SEAGATE 8T,SEAGATE 3T,SEAGATE 1T.

Any help would be greatly appreciated.

 

[Ripthruster]

Do you mean your GPU fans rev up before crashing? What kind of temps are you getting on GPU and CPU? Also, is your ram on your mobo's compatible list?

 

[MeanMachine41]

Hi alucardnz

You have an unstable system. Due to an OC or hardware malfunction.
At first glance it may be the GPU, however you should conduct a stress test.
Download HWMonitor and AIDA64. The results can be interpreted here.
Any issue or confusion just ask.
You DIMMs can be tested with Memtest86+ from a USB boot.

 

[alucardnz]

Ok so downloaded them and it crashes when i try to stress test it and i got 46 degrees under load on my GPU.

I sawped it out for my old GPU and hasn't crashed (yet)

 

[MeanMachine41]

Ok so downloaded them and it crashes when i try to stress test it and i got 46 degrees under load on my GPU.

I sawped it out for my old GPU and hasn't crashed (yet)

Which test did you conduct with AIDA64, when it crashed. Your temps are OK at 46C. Have you been OCing your card or still at stock settings.

Can you post a desktop pic of HWMonitor during test using CPU and FFU and nothing else. I want to see if any voltage variation under load. Take the shot after ten mins or before temps reach 70-75c.
Best to have both HWMonitor and AIDA64 along side each other on you desktop.

 

[alucardnz]

No OC just stock settings.

 

[MeanMachine41]

Well your CPU core temperatures are peaking at 65c so that's OK for a 12min test.

Now do the GPU test and post the results.

What I don't understand is why HWMonitor is not showing Voltage on your 12v, 5v, and 3v rails.
Can you change settings to reflect this.?

 

[hang-the-9]

Ok so downloaded them and it crashes when i try to stress test it and i got 46 degrees under load on my GPU.

I sawped it out for my old GPU and hasn't crashed (yet)

If it's stable with another card, maybe bad card, maybe bad motherboard, maybe bad power supply that can't run the new card. Is the old card same wattage use as the 1080?

Check for a BIOS and chipset update for the motherboard.

 

[alucardnz]

Sorry couldn't see any settings in HWMonitor.

 

[MeanMachine41]

AIDA64 has a pull down tab for voltage readout. try that.
My HWMonitor is ver 1.29.0 and I can read my rail voltages. Whats yours.?

Your GPU test is OK in AIDA however there does not seem to be much load.
Run EVGA precision x16 stress test for confirmation and report the results.

 

[alucardnz]

Utilizations,GPU was around 80% then come down

 

[MeanMachine41]

Your card is functioning well and can be OC if you wish. HWMonitor indicates all 12v, 5v and 3v volt rails are within acceptable limits. Temps on TMPIN0 reached 80C and that would either crash the test or your system.

You may need better cooling. Run the test again with your side panel off. Place a desktop fan pointing directly at the GPU and run the test again. If you have an improvement, then install more fans in your case to assist in disposing the heat or a water block is required.