» Search forum


Index :: BOINCstats general :: The way "Average floating point operations per second" is calculated @ BS...


Pages: [1]
[B^S] BOINC-SG
 
BAM!ID: 45646
Joined: 2008-02-09
Posts: 25
Credits: 1,472,797
World-rank: 174,591

2008-03-23 17:40:23

Hi,

I just found out that "Average floating point operations per second" is calculated by RAC here which seems wrong to me cause the more a project grants the more Flops there are - which cant be realistic.

If I follow this calculation, my AMD X2 5200+ does 7 GigaFLOPS @ WCG - but @ Milkyway, it does 30 GigaFLOPS.

I just remember the news from January 31st, which said that according to BOINCStats more than 1 PetaFLOP has been calculated per second...

Well, what would be a more realisitic number then?

Cheers, Shai
ASRock K10N78FullHD-hSLI Rev.3, AMD Phenom 9100e@4x 1.8GHz@1.0V max. 77W @ BOINC



1st of August 2007: Yes, I\'m a daddy!
Sid2
 
Forum moderator - BOINCstats SOFA member
BAM!ID: 28578
Joined: 2007-06-13
Posts: 7336
Credits: 593,088,993
World-rank: 3,385

2008-10-08 18:13:38



I just found out that "Average floating point operations per second" is calculated by RAC here which seems wrong to me cause the more a project grants the more Flops there are - which cant be realistic.

If I follow this calculation, my AMD X2 5200+ does 7 GigaFLOPS @ WCG - but @ Milkyway, it does 30 GigaFLOPS.




Shai:


The problem isn't with the stats here, it starts with the projects:


A Cobblestone of computation is supposedly 0.86 trillion floating-point operations. . .

. . . BUT with optimized aps, different processors, different projects, different operating systems [esp. 64-bit] the reality bears only casual resemblance to the theory.

The programming gurus have a acronym for it: GIGO [Garbage In, Garbage Out]



Pages: [1]

Index :: BOINCstats general :: The way "Average floating point operations per second" is calculated @ BS...
Reason: