heres my specs:
asus M5A97 PLUS mobo
2x4GB GSkill sniper F314900CL9D-8GBSR 1.5V 9 10 9 10 28
AMD FX 8350
evga 780 ftw vcard
Thermaltake tough power 750W (the $80-90 one)
2 sata 500 gig Hard drives
1 sata DVD burner
windows 7, sp1 all important updates. (fresh install)
Latest bios, chipset, dx, gpu, all drivers up to date.
EDIT: The motherboard, cpu and ram are the new parts. this is why my system is unstable, I got these parts yesterday.
current bios settings
all power features disabled, turbo disabled, spread spectrums disabled
HPC is enabled (dont know if this helps or not?)
memory set to 1866 and I only set 9,10,9,28, it auto sets to 1T.
cpu 1.35
cpu/nb 1.32
ram 1.525
nb 1.125
nb frequency 2400
HT 2400 (advertised as 2600 but 2400 is highest selectable)
there two LLC options: cpu and cpu/nb they are on auto. I only have enable,disable or auto
With these settings this seems to be the most stable I can get this system but eventually (might be 2 minutes or 2 hours) I am going to get a blue screen and I get all kinds, IRQ, PAGE FAULT, SERVICE something. I checked them out with 'whocrashed' and they say they are all driver related....
I thought it might be a power issue because one time the system was looping blue screens until I powered completely off and then booted, but during that process I noticed my dvd burner wasn't powering on. So I disconnected my extra hard drive and dvd burner to free up some power, but still blue screens eventually.
When I run memtest with one stick of ram in. It's ok, but if I change the cpu settings in memtest to 'parallel all cpus' I get errors in test 3, thousands of errors.... so that's confusing. windows memory tests out fine with single stick and dual channel.
I use cpuid hardware monitor and prime95 to stress my system and temperatures are fine, 55C~ underload. I have a nice aftermarket cpu cooler, cant remember the name but temps with stock cooler on stock 4.0mgz settings, the cpu would throttle itself down at 76C/63C from 4.0 to 1.4.. Stock cooler was one of the nicest I have seen from amd for a while so I wanted to try it out, but it was garbage. I used stock heatsink, then my heatsink, then turned the cooler around and same results each time = too hot. ofcourse I didn't let it run at high temps, just let it max out one time like (what the hell this is stock settings, with stock cooler, cant even run at 100%?
I tryed using my old ram ddr3 4gigs 2x2 1333 in dual channel and it was fine, the only thing I havnt tryed was running memtest with this ram and selecting parallel cpu and seeing if i get errors as well but I had no blue screens for the hours I was using this ram so i didnt run any memtests...
Anyway does anyone have any suggestions or see something I am not seeing?
What would be the next change to make, think the cpu/nb could be ok at 1.35? I can't think of anything else to do here....
asus M5A97 PLUS mobo
2x4GB GSkill sniper F314900CL9D-8GBSR 1.5V 9 10 9 10 28
AMD FX 8350
evga 780 ftw vcard
Thermaltake tough power 750W (the $80-90 one)
2 sata 500 gig Hard drives
1 sata DVD burner
windows 7, sp1 all important updates. (fresh install)
Latest bios, chipset, dx, gpu, all drivers up to date.
EDIT: The motherboard, cpu and ram are the new parts. this is why my system is unstable, I got these parts yesterday.
current bios settings
all power features disabled, turbo disabled, spread spectrums disabled
HPC is enabled (dont know if this helps or not?)
memory set to 1866 and I only set 9,10,9,28, it auto sets to 1T.
cpu 1.35
cpu/nb 1.32
ram 1.525
nb 1.125
nb frequency 2400
HT 2400 (advertised as 2600 but 2400 is highest selectable)
there two LLC options: cpu and cpu/nb they are on auto. I only have enable,disable or auto
With these settings this seems to be the most stable I can get this system but eventually (might be 2 minutes or 2 hours) I am going to get a blue screen and I get all kinds, IRQ, PAGE FAULT, SERVICE something. I checked them out with 'whocrashed' and they say they are all driver related....
I thought it might be a power issue because one time the system was looping blue screens until I powered completely off and then booted, but during that process I noticed my dvd burner wasn't powering on. So I disconnected my extra hard drive and dvd burner to free up some power, but still blue screens eventually.
When I run memtest with one stick of ram in. It's ok, but if I change the cpu settings in memtest to 'parallel all cpus' I get errors in test 3, thousands of errors.... so that's confusing. windows memory tests out fine with single stick and dual channel.
I use cpuid hardware monitor and prime95 to stress my system and temperatures are fine, 55C~ underload. I have a nice aftermarket cpu cooler, cant remember the name but temps with stock cooler on stock 4.0mgz settings, the cpu would throttle itself down at 76C/63C from 4.0 to 1.4.. Stock cooler was one of the nicest I have seen from amd for a while so I wanted to try it out, but it was garbage. I used stock heatsink, then my heatsink, then turned the cooler around and same results each time = too hot. ofcourse I didn't let it run at high temps, just let it max out one time like (what the hell this is stock settings, with stock cooler, cant even run at 100%?
I tryed using my old ram ddr3 4gigs 2x2 1333 in dual channel and it was fine, the only thing I havnt tryed was running memtest with this ram and selecting parallel cpu and seeing if i get errors as well but I had no blue screens for the hours I was using this ram so i didnt run any memtests...
Anyway does anyone have any suggestions or see something I am not seeing?
What would be the next change to make, think the cpu/nb could be ok at 1.35? I can't think of anything else to do here....
Comment