Repetitive test results not very consistent

[Strakele]

I have a new Core i7 machine and today I found your site and software and decided to give it a go to see some effects of my overclocking efforts.

Most everything is great... my CPU and video card scores are slightly above the averages in the charts on this site.

However, my hard drive and CD/DVD drive scores vary quite a bit when retesting. I understand that each time there will be a bit of difference, but for example, a CD Mark score that I recorded when testing all components at their stock speed was 530.8. Just now, I ran the same test, on the same drive, with the same disk (a Call of Duty game that happened to be laying around) and got a score of 1257.8, nearly 58% better.

Also, my IDE CD/DVD drive seems to get better scores than my SATA CD/DVD drive. Thought that would be the other way around.

In the first test I did, my hard drive (WDC 640 gig) scored right around the average in your charts. In subsequent tests, its score varies between 570 and 670.

Anything that I can do to make these tests more consistent?

Core i7 920
Asus P6T Deluxe V2
12GB G.Skill RAM
GTX 280 OCX
WDC 640GB HDD
Samsung IDE DVD burner
HP SATA DVD burner

[David (PassMark)]

CD drives aren't very quick, compared to hard drives. So the interface (SATA / IDE) doesn't matter very much from a performance point of view. CD drives also spin up and spin down, have different data rates depending on where the files are on the disk & how badly the disc vibrates,etc...

Hard drive performance is similarly effected by a multitude of different factors.

If you want a longer (more accurate) test, have a look at the advanced test options.

[Strakele]

I completely understand that there are tons of factors playing into the scores reported. My question just had to do with the fact that I could run a test, say Tests -> Disk -> All, let it finish, then immediately run the exact same test again and get substantially different scores.

[David (PassMark)]

I understand your issue, but the answer is still the same. Our software does the same thing each time you run it, other external factors change, and they can change from 1 minute to the next.

Long tests can smooth out some of these factors. But most people don't want a disk speed test that runs for 30min.

[Strakele]

Ok thanks, I guess that makes sense. Would overclocking other hardware normally have any effect on the hard drive score? My highest scores seemed to come when I had CPU, memory, and/or graphics card overclocked.

[David (PassMark)]

Overclocking the CPU should only have a minor impact on the disk speed. But the faster the hard drive, the more the CPU & RAM come into play.