Memtest86+ Q

[Twiggz]

Does this test not only test my timings but also test the HTT ???
Im curious cus i've passed test 5 for like 8 hours which i hear is good but im wondering if This test or other tests are needed to detect the actual HTT that you are increasing once you've decided your -timings- x-x-x-x are stable. Also how long before -we- decide that or timings are stable ? is 8 hours the definitive ? or the minimum ?

[Xtreme Barton]

some say 24 hours. but i personally think if you are using your pc and you cant tell the difference than its fine.. run games and such no problem multiple task and yuo see nothing than its ok.. well at least to me it is..


as far as your htt test your asking about.

just consider that you will need to test your memory and cpu when adjusting that.
there is no specific test just for that. it affects the speed of your memory and cpu which will be the ones to give out errors not the htt.

[Twiggz]

Got ya , so ill just test Memtest86 since im pretty sure my memory will become unstable before my proc.

[Xtreme Barton]

Got ya , so ill just test Memtest86 since im pretty sure my memory will become unstable before my proc.

well not necissarily

what you should try to do is this..

set back your memory with a divider.. then find your max cpu clock, and test it for stability. than slowly raise your memory and tighten/loosen timings and test for errors.

long process to find max oc but a good starting point.

[RedDragonXXX]

well not necissarily

what you should try to do is this..

set back your memory with a divider.. then find your max cpu clock, and test it for stability. than slowly raise your memory and tighten/loosen timings and test for errors.

long process to find max oc but a good starting point.

That's not necessary.

If you want to see how far you can push your RAM just drop down your multiplier and raise the HTT/FSB up slowly. Using dividers when pushing RAM is not advised (others will disagree) cause I found that you can get the highest OC with 1:1. This may not be true for all chips but I do it just in anyways.

Everything you need to know about Memtest you can find here (http://www.ocforums.com/showthread.php?t=409152) and I also added another solid program that will test your RAM through Windows and it works like a charm. Don't get me wrong, dividers are just fine but avoid using them when your are testing your RAM limits.

[Xtreme Barton]

That's not necessary.

If you want to see how far you can push your RAM just drop down your multiplier and raise the HTT/FSB up slowly.

Everything you need to know about Memtest you can find here (http://www.ocforums.com/showthread.php?t=409152) and I also added another solid program that will test your RAM through Windows and it works like a charm.


well i wasnt getting at the only way ...

either or ..

lower cpu and find ram first or lower ram and find cpu first

[voodoothenoob]

In the long run its best to test both cpu and ram seperately for max oc

that way you have some baselines

otherwise you will be hitting and missing more often imo

[Twiggz]

Ok that helps but i hope my winnie can hold more than 105 mhz oc :) so im gonna push memory more i've got it as far as 16 mhz and havnt had to raise any voltage yet is that good ? normal ?, Also is Prime95 still good cpu tester ?

[RedDragonXXX]

Ok that helps but i hope my winnie can hold more than 105 mhz oc :) so im gonna push memory more i've got it as far as 16 mhz and havnt had to raise any voltage yet is that good ? normal ?, Also is Prime95 still good cpu tester ?

16MHz is not bad especially for your sticks. If you hit the wall don't be affraid to use divider to push the CPU farther and yes Prime95 is still good CPU Tester/Burn-In program, but also do SuperPI 32MB test for stability testing.

[Twiggz]

went to divider cpu/11 for now may use another one later so far timings are working good.

[Xtreme Barton]

run bandwidth and cpu benchmarks to help determine your increase ..