Waco Posted May 15, 2011 Posted May 15, 2011 Usually is a key word. Just state your advise and move on would be my best advise and not troll. TY Troll? This coming from the guy that didn't read the first post at all, posted advice about something the OP had already done, and didn't realize his advice wasn't even applicable for this case (since Prime clearly didn't catch the problem)? Right. Anywho - I noticed that some of you don't like IBT/Linpack because it's not a sustained load. If you run it with larger memory sizes like you're supposed to it's far more stressful for a lot longer time than Prime95 is (which runs very small tests thousands of times per test). As for your BSODs - it does sound like a memory problem or something. An overnight test of memtest would be more than sufficient to rule the memory itself out. I'm not sure what the safe limits are for VTT on the new SB CPUs are though... Share this post Link to post Share on other sites More sharing options...
momosmokey Posted May 15, 2011 Posted May 15, 2011 Troll? This coming from the guy that didn't read the first post at all, posted advice about something the OP had already done, and didn't realize his advice wasn't even applicable for this case (since Prime clearly didn't catch the problem)? Right. Anywho - I noticed that some of you don't like IBT/Linpack because it's not a sustained load. If you run it with larger memory sizes like you're supposed to it's far more stressful for a lot longer time than Prime95 is (which runs very small tests thousands of times per test). As for your BSODs - it does sound like a memory problem or something. An overnight test of memtest would be more than sufficient to rule the memory itself out. I'm not sure what the safe limits are for VTT on the new SB CPUs are though... It's not SandyBridge, if that's what you meant my "SB." It's last years 1156 I3's. From what I've read, the max safe voltage for the 1156 i3's is 1.4V for VTT--right now I'm at 1.270V...I think. My sister didn't know what memtest86 was and tried to use the computer, long story short, she restarted the computer several times in an attempt to get it "working." Needless to say, it restarted memtest86 and I lost that nights work. I'll run it again tonight and let you guys/gals know. Share this post Link to post Share on other sites More sharing options...
Drdeath Posted May 16, 2011 Posted May 16, 2011 Troll? This coming from the guy that didn't read the first post at all, posted advice about something the OP had already done, and didn't realize his advice wasn't even applicable for this case (since Prime clearly didn't catch the problem)? Right. Anywho - I noticed that some of you don't like IBT/Linpack because it's not a sustained load. If you run it with larger memory sizes like you're supposed to it's far more stressful for a lot longer time than Prime95 is (which runs very small tests thousands of times per test). As for your BSODs - it does sound like a memory problem or something. An overnight test of memtest would be more than sufficient to rule the memory itself out. I'm not sure what the safe limits are for VTT on the new SB CPUs are though... IntelBurn does not stress the cores 100% for very long. Has nothing to do with memory. It is not a good stress test for quads and 6 cores period. Sorry to break the news on IntelBurn. Prime95 remains the strongest stress test. I have been stressing for many years to pass IntelBurn/linpack to fail Prime95. Prime95 will stress the Northbridge much harder also. There is no stress test that stresses a chip 100% even if it says 100% on the screen. Just sayin. Also,would be nice to stay out of it Waco. You tryin to be a bully? Won't work..... Thats why many stay away from forums when peeps troll around............. Anywho back to topic. Appreciate it. Share this post Link to post Share on other sites More sharing options...
momosmokey Posted May 16, 2011 Posted May 16, 2011 (edited) IntelBurn does not stress the cores 100% for very long. Has nothing to do with memory. It is not a good stress test for quads and 6 cores period. Sorry to break the news on IntelBurn. Prime95 remains the strongest stress test. I have been stressing for many years to pass IntelBurn/linpack to fail Prime95. Prime95 will stress the Northbridge much harder also. There is no stress test that stresses a chip 100% even if it says 100% on the screen. Just sayin. Also,would be nice to stay out of it Waco. You tryin to be a bully? Won't work..... Thats why many stay away from forums when peeps troll around............. Anywho back to topic. Appreciate it. OK, ran memtest 86 for 9+ hours; something like 21-22 runs, no errors--would you guys consider that stable memory? Further suggestions? Edited May 16, 2011 by xxmastermushxx Share this post Link to post Share on other sites More sharing options...
medbor Posted May 16, 2011 Posted May 16, 2011 IntelBurn does not stress the cores 100% for very long. Has nothing to do with memory. It is not a good stress test for quads and 6 cores period. Sorry to break the news on IntelBurn. Prime95 remains the strongest stress test. I have been stressing for many years to pass IntelBurn/linpack to fail Prime95. Prime95 will stress the Northbridge much harder also. There is no stress test that stresses a chip 100% even if it says 100% on the screen. Just sayin. Also,would be nice to stay out of it Waco. You tryin to be a bully? Won't work..... Thats why many stay away from forums when peeps troll around............. Anywho back to topic. Appreciate it. Then how come it crashes in linpack but not Prime95? You just suck... all tests are good (if any test fail you are not stable, even if it passes 5 weeks of prime95) OK, ran memtest 86 for 9+ hours; something like 21-22 runs, no errors--would you guys consider that stable memory? Further suggestions? Then memory seems fine in my opinion. Do you have any lower memory multipliers? Try the lowest one, since then we can be (almost) sure that it is memory related. Share this post Link to post Share on other sites More sharing options...
BluePanda Posted May 17, 2011 Posted May 17, 2011 IntelBurn does not stress the cores 100% for very long. Has nothing to do with memory. It is not a good stress test for quads and 6 cores period. Sorry to break the news on IntelBurn. Prime95 remains the strongest stress test. I have been stressing for many years to pass IntelBurn/linpack to fail Prime95. Prime95 will stress the Northbridge much harder also. There is no stress test that stresses a chip 100% even if it says 100% on the screen. Just sayin. Also,would be nice to stay out of it Waco. You tryin to be a bully? Won't work..... Thats why many stay away from forums when peeps troll around............. Anywho back to topic. Appreciate it. FYI- linpack is embarassingly parallel... OP- if you're able to pass all these tests its got to be stable by now . Has it crashed doing anything else? Share this post Link to post Share on other sites More sharing options...
sticknstone Posted May 17, 2011 Posted May 17, 2011 CPU-Z says it's 1.32V when idle and 1.296 when maxed. This looks like power saving to me. I just started oc my cpu so I am no expert, but did you go to control panel, system and security, power options and select high performance under more plans. Share this post Link to post Share on other sites More sharing options...
momosmokey Posted May 17, 2011 Posted May 17, 2011 (edited) FYI- linpack is embarassingly parallel... OP- if you're able to pass all these tests its got to be stable by now . Has it crashed doing anything else? It never seems to crash in any test other than LinPacks; however, it did crash in Prime95 after 19-20 Hrs...possibly--my sister was watching some TV show online while Prime95 was running when it black screened/froze. It never crashes in games, at least not from hardware related issues I don't think. It appears pretty darn stable from what I've observed. If it's only LinPacks do you think there's a bug in the software or something? I'll try LinX and get back to you guys/gals. Update: OK, I tried LinX. I let it run and left it, when I came back the computer had rebooted. I dunno, I'll try stock settings to make sure it's actually the OC. I'll keep you posted... Update 2: Stock settings are fine for LinX/LinPacks. I ran for 7 hours and it didn't crash. So are the culprits of crashing, in my case: CPU, RAM, and North-bridge? What do you guys/gals recommended? Edited May 17, 2011 by xxmastermushxx Share this post Link to post Share on other sites More sharing options...
momosmokey Posted May 17, 2011 Posted May 17, 2011 (edited) This looks like power saving to me. I just started oc my cpu so I am no expert, but did you go to control panel, system and security, power options and select high performance under more plans. I believe it's actually called "Vdrop" where the CPU's voltage drops slightly when maxed out. It's something that needs to be compensated for when OCing. Good suggestion none-the-less. Edited May 17, 2011 by xxmastermushxx Share this post Link to post Share on other sites More sharing options...
ccokeman Posted May 17, 2011 Posted May 17, 2011 Share this post Link to post Share on other sites More sharing options...
iskout Posted May 17, 2011 Posted May 17, 2011 Share this post Link to post Share on other sites More sharing options...
jammin Posted May 17, 2011 Posted May 17, 2011 My suggestion would be - back off the CPU OC a little until it does pass everything (seeing as it seems fine at stock), then slowly work your way up again to see where it becomes unstable. Boring but methodical wins most of the time I find (but it is boring.. booo). It's either that, or trying to fiddle with voltages some more.. change, test.. change, test.. oh the fun. Share this post Link to post Share on other sites More sharing options...
Recommended Posts
Please sign in to comment
You will be able to leave a comment after signing in
Sign In Now