Wprime WR, 1.147s 32 M

[steponz]

I was able to acheive this time without any cheats.

 

Not sure what to believe... Really make me think that wprime should be removed as a bench.

 

I can replicate and can take video if needed to show the result and what happens.

 

My SS drops 10 to 15 to degrees when I run this and the SS makes a whoosh sound.

Very interesting... and haven't seen this happen even when I have had my gulftown up at 5.1ghz with 1.8v vcore..

[knopflerbruce]

If bugged scores are enough to remove a benchmark, we need to kick the whole FM series from HWBot:p

 

If the bug is obvious, it's no problem.

[steponz]

Honestly the wprime versions need to be splitup into the different versions on the board. Kind of like how FutureMark stuff is... A benchmarking program should have much more logging in it then what is out there today...

[Splave]

because you can replicate a bug doesnt make it valid

 

this is def not a keepable score, why do you think you are so much faster than everyone else with higher clocks?

[knopflerbruce]

There was also posted a replicable superpi 32m score that was WAY out of line. So... we must remove anything except PiFast, CPUZ and AM3:p

[Massman]

I hope you see that your run is just way out of line ...

[steponz]

I see its out of line.. I totally understand that.

 

I was just wondering if there was a way to see if it was invalid.

 

I would run another test and it would tell me that the last was successful...

 

There should be a better way of validation.. I guess we will just have to wait.

[Oj0]

I was just wondering if there was a way to see if it was invalid.

 

Is it being ten times quicker than it should be not enough?

[steponz]

Never..... a boy can dream ya know....lol....