Some strange results I've come across

[komadyret]

I've been looking over the results for a cpu I have on my shelf that I may bench, and came across this anomality. CPU category and cpu-z doesn't match up, and same cpu-z version (1.45) reports different stats/names for the cpu for different benches done by this user.

May this be an unlocked cpu (look at the multi in cpu-z verification) that confuses cpu-z?

 

Hope a mod can look at these and make an educated opinion and answer me what is going on here.

 

Sorry for the long link:

http://www.hwbot.org/searchResults.do?direction=&applicationId=0&teamId=1046&userName=pepek&gpu=&numberOfVideocards=0&cpu=Athlon+XP+2600%2B+Thoroughbred+%28210%29&chipset=&model=&manufacturer=&minScore=&maxScore=&gpuId=0&cpuId=227&chipsetId=0&modelId=0&manufacturerId=0&offset=0&displayAdvanced=false&countryId=0&dateFrom=&dateUntil=&minGpuCoreFreq=&maxGpuCoreFreq=&minGpuMemFreq=&maxGpuMemFreq=&minCpuFreq=&maxCpuFreq=&system=&minTotalPoints=

[CaNNon]

It's cpu-z that does that koma, It seems to set the chip type of a clock range for some amd's (maybe all but i can't say for sure) and their may be differences in whatever they have biased the ranges on from version to version.

I don't see any unlock done on the chip but it can also be done under the chip or at socket level.

Might be best if you compared core names and stepping windows.

[komadyret]

I did, and it all add up. Cores, steppings etc. But multis change over the various screenies, and the cpu-z was the worst with it'sx22 multi.... lol

 

Well. No big deal really. Results are reasonable and all is fine, except the cpu-z in each of the screenies