Sign in with
Sign up | Sign in
Your question
Solved

Stability issues...need help

Tags:
  • RAM
  • CPUs
  • Overclocking
Last response: in Overclocking
Share
a b à CPUs
November 6, 2009 1:20:51 PM

thanks in advance to anyone who can help out. I'm a little stumped on the cause of my instability.

here's my system:
CPU.....................QX9770 - 3.6GHz (9x400) - 1.1975v
Motherboard.......ASUS Maximus Extreme
RAM.....................2x1GB Cellshock DDR3-1800, 2x2GB Patriot Viper DDR3-1600 - 8-8-8-24 - 1.80v
Videocard............Powercolor 4870 1GB
watercooled CPU, NB & SB chips...even under stress tests nothing gets above 30C

I have had the full 6GB of RAM in for the past 12 months clocked to 1600MHz and CAS of 8-8-8-24, the CPU has been overclocked to 3.6GHz (9x400). for the most part it has been stable (maybe 1 BSOD per month). Now all of a sudden I have had 3 or 4 BSOD per day...

here's what I've done so far:
- I removed the Cellshock 2GB of RAM and just populated slots 1 & 3 with the Patriot Viper 2gb modules (4gb total)
- I relaxed the timing to 9-9-9-24
- I upped the RAM voltage to 1.90v (recommended for the Patriot RAM at DDR3-1600 speeds)
- I upped the CPU voltage to 1.25v
- backed the CPU to stock speed of 3.2GHz (8x400)

I ran prime95 and as I relaxed timings and it gets better, but most of the time still fails. Nothing seemed to work until I left the RAM frequency to "auto" and it scaled back to DDR3-1067 (even leaving the CPU at 3.6GHz). However...I ran the RAM on memtest86 4.00+ for 2 hours at DDR3-1600 7-7-7-20 speeds and did not have one error. I also tested another time at DDR3-1600 9-9-9-24 speeds for 1 hour without an error. So I think the RAM can handle the speeds. So I'm starting to suspect the North Bridge. I have had the voltage set to both maually at 1.42v and auto (~1.44 in everest). Should I bump up the NB voltage to ~1.5v?

When I run prime95 and select the CPU intensive test it passes...it only fails when I run the stress test that also stresses the memory.

sorry for the long post...and thanks in advance




More about : stability issues

Best solution

a b K Overclocking
November 6, 2009 3:37:21 PM

firstly i would up the nb and cpu voltage, link the mem speed to the fsb and then try again.
Share
a b à CPUs
November 6, 2009 4:03:23 PM

thanks...what nb voltage would you recommend? I googled the motherboard and find answes all over the map. from 1.4v - 1.61v. When I link the mem speed to the fsb...that's when I get BSOD. I now think that my nb voltage is to low (I'm at work jnow, so I'll change and test when I get home tonight). I just wanted confirmation that I was on the right track and that the nb could cause problems like this.
m
0
l
Related resources
a b à CPUs
November 7, 2009 9:42:17 PM

I found the problem...the Patriot Viper DDR3-1600 2x2GB ram chips were bad...I put in just the 2x1GB Cellshock DDR3-1800 chips and now I'm running fine at 1600MHz with 7-7-7-7-20 timing. So since I'm running windows7 x64...I have to haev more than 2GB of RAM...so what RAM is good for this board? Corsair dominators?
m
0
l
a b à CPUs
a b K Overclocking
November 8, 2009 3:01:55 AM

japps2 said:
I found the problem...the Patriot Viper DDR3-1600 2x2GB ram chips were bad...I put in just the 2x1GB Cellshock DDR3-1800 chips and now I'm running fine at 1600MHz with 7-7-7-7-20 timing. So since I'm running windows7 x64...I have to haev more than 2GB of RAM...so what RAM is good for this board? Corsair dominators?


Bet if you put only the Patriot sticks in, they would be fine too. Mixing different ram = trouble.
m
0
l
a b à CPUs
November 8, 2009 3:11:26 AM

Originally I did mix RAM, but lately the problems I had was with just the patriot stick...just the 4GB...in slots 1 & 3...in slots 2 & 4...at DDR3-1067, DDR3-1333 & DDR3-1600 with a lot of different CAS values and countless different voltages...NB voltages. nothing seemed completely stable with the Patriots. I'll pick up some Corsair dominator stick from newegg this week.
m
0
l
a b à CPUs
February 26, 2010 5:13:08 PM

Best answer selected by japps2.
m
0
l
!