lol, no wonder, i was indeed shocked to see very superb performance with low latency using pc5300 card.

yeah, very good performance, but you need better cooler for your card then..

orthos stable?

water cooled....?

8cm fan ontop n took out all the sticker..reason: so tat every D9GMH chipset running on same temp..more stable.

Superpi 32mg pass..neber try orthos.

if im not mistaken larr, superpi wont stress ram or cpu to 100%,

orthos does.. and you can continue stressing for longer period of time..

best to check stability is orthos..

IINM lar.. kalu silap jgn marah..

AFAIK, the best way to test for stability is memtest. However, you have to run memtest from bootdisk rather than in windows as the problem with running memtest in windows is that it cannot test all the memory as windows uses a large part of the available ram....

I disagree, memtest DOS is not consistent. U may pass all the tests there, but you cant boot into windows at times. My best test is to use Prime Blend Test, where alot of rams are tested

I disagree too!!! There are a lot of factors that is important for windows to boot. The whole rig has to be stable, I guess.

Since memtest only test the RAMs, it is not testing the whole system and thus may still result in a non-booting windows if other parts of the system are not stable.

I feel that the Blend Test is to test the whole rig's stability......

Edit : This is what I usually do although I'm not too sure whether it is the correct way.

OC ram - test in memtest for stability

OC CPU - test with orthos

OC both CPU & ram - test with blend test for whole system stability....



This post has been edited by kmarc: Aug 22 2007, 03:17 PM

Well yes off course it depends on other factors. Let's assume we are talking bout rams overclocking since we are on this topic anyways, and leave out other factors. It has been said over a few times that Memtest isn't a proper test tool for ram testing purposes. Blend testing is sufficient as it taxes the rams quite a lot

i ran 2 superPI.. faster result if anything not stable..
orthos sometimes must wait 5 minutes to catch errors..

but even u able to pass the superPI test, when u run orthos, sometimes there will be error as well...

to me, 24 hours orthos should justified the stability of the system at that particular.
I don't believe few minutes SuperPI equal to 24 hours orthos torturing test.
By then, can consider OCCT as well.

I know what you mean. However, the problem with blend testing is that it will not stress the RAM continuously, like what it does with the CPU when you set it to small FFTs. Perhaps a custom setup where the RAM is stressed continuously?

---

Added on August 22, 2007, 5:03 pmFrom my experience, superPI is not even close to being a stability test program. I can usually pass superPI but still fail memtest or orthos at a particular OC.....

This post has been edited by kmarc: Aug 22 2007, 05:03 PM

ofcourse.. just save the time instead running orthos at the first place.. so can tweak the ram right faster..

24 hours is too long for me , normally i just run orthos for around 8-9 hours then i consider it is stable already...
ya, OCCT is quite good for the stability testing, at least its layout look nicer than orthos.

funny, last time memtest failed but orthos belnd test for 12 hours no prob... superPI 32mb oso ok...

hi guy i recently bought a pair of crucial 10'th anneversary ty12864aa663, anyone using right now? and how far this ram can go,timing,vdim...

check google la...last i saw was 1200mhz 5-5-5-15 @ 2.3v

can go so high?? checking google... cant see any... u have link?

hi guy i recently bought a pair of crucial 10'th anneversary ty12864aa663, anyone using right now? and how far this ram can go,timing,vdim...

PM coolice or superfly