Sign in with
Sign up | Sign in
Your question
Solved

More issues

Last response: in Systems
Share
February 25, 2010 6:57:15 PM

This is a continuation of this thread: Prime 95 Results

I got the warranty replacement RAM in. When I tried to run Prime95 the PC froze (no mouse movement, etc). I rebooted and loaded windows a second time and it froze when trying to browse a mapped network drive (no more than a couple minutes after reboot).

I'm on my 3rd pass of MemTest as we speak without errors. My question is: Could this be caused by setting the voltage to 1.66 in BIOS (the RAM requires 1.65)? I forgot to write down temp readings, but I'm fairly certain that none of them went above 70 degrees C when the freeze occurred (could they have shot up before the screen refreshed? I don't know.)

Anyway, just curious where I should start my search when I get home from work. I'm thinking about dropping the voltage to 1.64 to see if that alleviates the errors. From there, I'm not sure where to go.

The system ran fine with the old RAM (no freezing, rebooting, etc), it just wouldn't pass Prime95; cores were failing after some time. MemTest found a bad stick, so I RMAed the set. Now, with supposed good RAM, I'm getting freezing in the OS.

Any suggestions appreciated. Thanks guys!

Josh

More about : issues

February 25, 2010 8:33:39 PM

MemTest ran for 4 hours. Completed 4 passes with no errors. Dropped the voltage to 1.64 and went back into Win 7. Ran Prime95, and then after 2-3 minutes the PC locked again. The hottest core hit 71 degrees C.

Any ideas?
m
0
l
a b B Homebuilt system
February 25, 2010 8:42:35 PM

Did you also set the RAM speed and timings correctly in the BIOS? If not, you should.

Which exact RAM kit did you get as a replacement? Is it the same as you had before? I wouldn't be surprised if the replacement RAM isn't any good either. That's the kind of luck I've had with OCZ RAM. I've had much better luck with G.Skill. I've bought quite a few G.Skill RAM kits in the past couple years and I've only had one faulty stick out of all those.
m
0
l
Related resources
February 25, 2010 8:51:37 PM

shortstuff_mt said:
Did you also set the RAM speed and timings correctly in the BIOS? If not, you should.

Which exact RAM kit did you get as a replacement? Is it the same as you had before? I wouldn't be surprised if the replacement RAM isn't any good either. That's the kind of luck I've had with OCZ RAM. I've had much better luck with G.Skill. I've bought quite a few G.Skill RAM kits in the past couple years and I've only had one faulty stick out of all those.


Yes, all the speeds and timings, etc, are set correctly. It's the same RAM as before. 4 passes with no errors in MemTest is usually a pretty good indication if there RAM's bad or not. I bought this RAM specifically because it was on the motherboards approved RAM list.

Any idea what else I can try. I'm going to have a hard time getting them to RMA another kit if I don't have proof. I wish I had the money to just buy another kit to test, but I can't put that kind of money upfront and then afford to be able to pay the restocking fee if it still freezes.

Thanks for the reply.
m
0
l

Best solution

a b B Homebuilt system
February 25, 2010 9:00:47 PM

I would let Memtest86+ run overnight for at least 8 hours. I've personally seen a set of RAM pass 5 passes without any errors and then start giving errors.

Are you running at stock speed? Those temps are pretty high for a stock speed CPU with the ARCTIC COOLING Freezer 7 Pro Rev.2. Are you 100% sure all four push-pins are fully seated? What is your idle temp? Did you install the HSF while the motherboard was outside the case so you could make sure all the pins were fully seated?
Share
a b B Homebuilt system
February 25, 2010 10:14:10 PM

Those temps are pretty warm. According to a few manufacturers, you're supposed to install the HSF with the board mounted, but I've found it much more effective to do it with it out. Lost 8C at load that way :) 

m
0
l
February 26, 2010 12:10:47 AM

I plan on letting it run overnight. Yes, everything's running at stock speed. Everything should be pushed down fully, as I put it on with the motherboard outside the case. If the RAM doesn't fail, then I'm at a loss.
m
0
l
a b B Homebuilt system
February 26, 2010 12:14:16 AM

I'm not convinced the cooler is installed properly with those temps. What software are you using to read the temps?
m
0
l
February 26, 2010 12:23:18 AM

shortstuff_mt said:
I'm not convinced the cooler is installed properly with those temps. What software are you using to read the temps?


Real temp. See, the problem is, I never had problems with freezing before this RAM came in. The other RAM failed MemTest and also caused issues with Prime95, but I never had an issue with installing, booting, or things running well. Everything was fine.

Now, with good RAM that's passed MemTest to this point, I'm having issues, which just seems weird. Why would it have not gotten overheated and froze with the old RAM, but does with the new stuff?
m
0
l
February 26, 2010 9:37:30 AM

Alright, so just an update. MemTest locked up after 2 hours and 45 minutes, although it showed no errors.

I held the power button in to turn it off, then when I tried to power it on, everything turned on and the fans ran, etc, but I have no display. Tried it two or three times with the same results. I'm going to reseat everything and go through the boot problems post.

I really think the RAM is bad now though. I had none of these issues when using the other stuff.

Edit: So, after 2-3 minutes, I turned it on again and it came up without reseating anything. I've got it running MemTest again, since I really can't do anything with it today. (Normally I'd take it to work with me and work on it in my free time, since I work at a local computer place, but I have 14 laptops and 4 PCs to set up, 2-3 PCs to repair, and 15-20 servers that I have to monitor remotely. Not going to be any free time today, but I'll try and hit the boards several times just to see if anyone has any other suggestions.)
m
0
l
a b B Homebuilt system
February 26, 2010 4:38:10 PM

I had an experience like that once with my personal machine. It had a failing stick of Crucial Ballistix in it, and if I left it at 1.8V, it would lock up Memtest. If I jacked the RAM to 2.1 or so, it would complete the passes, just completely full of errors.
m
0
l
February 26, 2010 4:51:37 PM

jack_attack said:
I had an experience like that once with my personal machine. It had a failing stick of Crucial Ballistix in it, and if I left it at 1.8V, it would lock up Memtest. If I jacked the RAM to 2.1 or so, it would complete the passes, just completely full of errors.


I'm going to go ahead and pick up some Artic Silver 5 and redo the HSF installation, just to be safe, but I really don't think thermal is the issue, because I was able to run over 2 hours of Prime95 with one of the good sticks of RAM from the last kit that I had, and I had no lockup type issues.

I'll also go ahead and reseat everything, just to be sure, but if anyone wants to give me the steps they think I should take, I'm willing to try it.

Thanks for the replies, guys.

Edit: I figured I'd update the status of MemTest now, too. I checked when I went home from lunch and it had been running for over 6 hours without errors (or locking up). I went ahead and left it running, and will update when I get home from work.
m
0
l
March 4, 2010 11:18:21 PM

Best answer selected by MaizeNBlue2.
m
0
l
!