Originally posted by Hanners
This is my whole point - This isn't a gaming benchmark.
I know a few people who have had perfectly stable overclocked systems for months on end, then one day they'll do something different with their machine and bang! Blue screen heaven. Down clock a bit, and everything is fine again.
This is exactly what you're seeing, this benchmark obviously just pushes your machine just that little bit too far. That doesn't make PCMark some terrible, evil benchmark.
Originally posted by bkswaney
I'm scoring 5200 with my setup to.
Tho I did have pcmark crash 2 times trying to run it.
What's up with that? I run every test I can throw at it on being stable
and it is stable.
I guess it does not like overclocks very well.
EDIT: WHat I cannot figure out is why I'm not crashing running the test.
But when I try and use the online compare it locks me up.
I'm not the only one with an Intel setup that is having their system overclocked
, yet we can both run other stability programs like (Sandra, SysMark or Prime95) without problems at our normal o/c speeds.
You can't say. "Flawed"?
How do you know it is so perfect? How do you know it isn't plagued with bugs?
In your own words:
"nVidia takes the crown by a massive margin here, and gives some rather interesting results. As expected, the extra bandwidth and fillrate of the 5900 over the 9800 Pro gives it the edge in most tests (although ATi's card performs better in the transparant window test). However, the score in two of the tests relating to graphics memory show an absolutely immense difference in favour of nVidia, to the tune of around 3,000 FPS!! I'm unsure of whether these scores are accurate and reflective of the hardware or down to a driver issue on ATi's part, but nevertheless those results give a huge boost in the AMD systems favour, and as one of these tests is also present in the main PCMark test to calculate an overall score, it explains the large advantage the AMD rig shows in the final scoring."
You are straight out contradicting yourself.