Why is the AMD 5800K still showing a CPU Mark of 4898? Please Explain...

[Armcandy]

Why is the AMD 5800K still showing a CPU Mark of 4898? I received a CPU Mark of 7840. Seriously, to get a score that low, I would have to be Running four VMs all concurrently running:
A. Norton Scan
2. BF3
3. A Netflix Movie
4. Disk Defrag
5. Pass Mark

Question, what is Pass Mark posting for the final result? The Median, Mean or the Mode of all 5800K's benched? Also, does Pass Mark provide a list of all 5800K's benched from lowest to highest? Can PassMark determine if multiple programs are running during a system benchmark?

I have a theory, Intel is so terrified of this CPU cutting into their mid-range i5 - i7 market-share, they purposely build a bunch of really slow AMD 5800K decoys and run/post the lowest CPU Pass Mark scores to this site. How would PassMark know if a company or hired contractor was purposely doing something this malicious?

I mean really, what I built was nothing special but this thing is Evil Fast:

1 x SAMSUNG 830 Series MZ-7PC256B/WW 2.5" 256GB SATA III (SSD)
1 x ASUS DRW-24B1ST/BLK/B/AS Black SATA 24X DVD Burner
1 x Microsoft Windows 7 Home Premium SP1 64-bit
1 x RAIDMAX HYBRID 2 RX-530SS 530W Modular Power Supply
1 x Antec Nine Hundred
1 x ASRock FM2A85X Extreme6
1 x AMD A10-5800K Trinity 3.8GHz (4.2GHz Turbo)
1 x G.SKILL Sniper Series 8GB (2 x 4GB) 240-Pin DDR3 SDRAM DDR3 1866

This CPU is really amazing!

AMD A10-4655M APU	3209	381	NA	NA
AMD A10-5700 APU	4775	210	38.20	$124.99
AMD A10-5800K APU	4898	197	37.68	$129.98

[David (PassMark)]

I think you are just confusing the scores from PerformanceTest V7 and PerformanceTest V8.

The charts use average non overclocked scores from PerformanceTest V8, and the 5800K consistently scores around 4850 in PT8. Standard deviation is 277. So ~68% of the 5880K results are between 4573 and 5127.

In the old PT7 software the 5800K scored significantly higher (around the 7000 numbers you are seeing).

The difference is due mostly to 3 factors,
1) There is a performance effecting bug in the Llano CPUs.
2) Our PT7 test was too focused on doing division and didn't use a broad enough set of instructions. This was improved in PT8. PT7 thus favoured this class of AMD chips.
3) We added a single threaded test in PT8. Which (relatively speaking) boost the overall score of single and dual core CPUs and hurts CPUs with more cores.

While you are entitled to your opinion that your own machine is amazing, the fact of the matter is that while the A10-5800K is a solid, value for money CPU, there are lots of better performing, more efficient CPUs available (if you have the budget). There is no global conspiracy. But it has been our experience that people don't like being told that their new computer isn't as good as they imagined it was.

[a7751218]

Thanks for the sensible reply David, I was also wondering regarding the difference in scores.