prime95 failed at 45 min, FSB345, E6300/P5B dlx

Wondering the best way to approach where the problem is...

Using the voltage recommendations in wusy's guide, got to FSB345 before failing prime95 in 45 minutes.

Tried to back off any of the voltage settings back at FSB340 had prime95 failing within a minute or so.

At these voltage settings, I gave FSB of 401 a try just for the hell of it, and it let me get to Windows fine, failed prime95 in about a minute.

Zalman 9500 cooler
corsair xms pro
P5B deluxe

core temps are 38 running prime95, system temp of 33. core temps go to 43 with prime95/intel tat at the same time.

vcore dropped 0.400 when prime95/intel TAT running at the same time, then stays stable. Prime95 by itself dropped vcore 0.200 then stayed stable.

I'm going to run a conservative overclock anyway, but I'm just curious what the problem is. Was expecting to be stable to at least a FSB of 380 running the recommended voltages and the hardware purchased.

Thanks for any insight!
11 answers Last reply
More about prime95 failed fsb345 e6300
  1. the p5b and p5b dlx have serious vdrop issues with the cpu, mch, memory, etc

    your going to have to do some resistor soldering on the board to get the voltages stable and everything working properly. I am not sure how the people who reviewed the p5b/p5b dlx boards got them to work without vmods. I have had to do them. Did them on my dlx and on a friends vanilla p5b.

    info can be found here->

    good luck
  2. Soldering on a brand-new motherboard? I'll save that for a few years from now...

    Didn't come across any of this while searching for a solid motherboard, as you said, all reviews/comments for the most part had the board being fine.

    Prime95 just failed at 56 minutes at FSB340, a booo! Down to FSB333 I go...
  3. ok i cant go beyond 420 with mine but 340? there is some problem with your board man
  4. Knowing my luck probably, hehe.

    At FSB401 right now, going to keep it here for a while and stress it with Intel's TAT, keep an eye on temps, and do my installs over the course of the day and see how it goes. Read up some more on the issues and go from there later.

    Thanks again folks!

    PS Hmmm, reduced mem timing from 5-5-5-12 (which it was rated for) to 5-5-5-15. Prim95 which failed in under a minute at this FSB speed before (and lower ones) is now 15 minutes in without issue. Interesting...
  5. im telling you its a vdrop issue

    with micron d9 ram kept clunking out above 333mhz until i did the vdimm mod & then i did vdrop and vcpu 0 got it 90% stable at 400x7

    finally did vmch and vfsb, 100% stable @ 7x470 with some tweaking

    5 resistors will solve an aweful lot of trouble
  6. The fact that some of these tests like prime or orthos or whatever , fails after a few minutes , doesnt mean you cant play games without problems ofcourse.So , its good if you pass those tests but not a disaster if you dont.
  7. With one step lower voltages than the wusy guide other than the RAM, seems more stable at FSB400 than ~FSB360 with higher voltages, hehe.

    Thanks for th advice folks! Seeing the benefits of dual-core right now; got a batch RAW process running in the background and I can actually still browse!
  8. best to get it prime stable because then the computer is stable

    might be able to play games if not prime stable, but it may lock up at the worst possible times.

    do some resistor mods, thats my advise having dealt with those boards
  9. Im not sure if it was specific to your board but I have been reading alot of people having instability between 400 and 405 FSb. once they jumped to 405 and went higher everything was stable. Anyone know anything else about this?
  10. Got orthos to fail with vdimm of 1.9V at FSB333, a boo!

    Going to pull back timings even more on RAM and see...
  11. Success! RAM timings were the issue, and not the Big 4 either.

    There's a sticky about it on the corsair forums:
    If you have an ASUS P5B-X and have errors in Memtest!
    ‘Write to Precharge delay’ in the BIOS (same page as for latencies) from default 10 to 15.

    Running a conservative overclock on FSB400. 2.1vdimm, 1.3625 vcore (tested down to 1.3V was stable, but passmark showed a drop in performance), 1.3V FSB, 1.25V NB, 1.5V SB.

    Core temp of 41C running Orthos, 25C idle.
Ask a new question

Read More