The official HWBOT Team Cup 2012 - SC6: "PCMark05 xCore" thread.

[knopflerbruce]

I think we should be consistent. We don't question text edit and TW scores that are crazy high, we just "assume" they're OK based on trust (it looks that way to me, at least). Vapor's word is just as good as Pro, CherV, Stevero's - or the rest of that bunch of crazy guys, so if we can trust them when they say their tweaks are legit (without proof), then we have to trust Vapor as well.

[Moose83]

Im testing it atm. 12.8 driver with SSDs onboard:)

[Moose83]

So, the results of VapoR are real! I did an short test, with no overclocking, all running standart

 

http://www.abload.de/image.php?img=unbenanntzhp1a.jpg

[VapoR.]

I test again, http://www.3dmark.com/pcm05/3168391 almost 10k VS! I think Samsung 830 i the best SSD, i tested Vertex 4 128GB and it has poor performance in PCM.

[I.M.O.G.]

So it can be done on two SSD's without any 3rd party non legit software... cool...

 

Yes but its a bug that tricks the software into thinking it ran faster than it did. It doesn't actually produce higher results in ANYTING, except PCM05, because it bugs out on a setting which has nothing to do with actual performance.

 

But then, nothing in PCM05 has anything to do with actual performance anymore, which is why only a handful of people bother running it anymore.

 

I shouldn't have said anything about the settings I mentioned, because that same handful of people will use it, despite it being a bug, and everyone will clamor over how great the scores are. Certain drivers and firmware have the same result, they bug out PCM05.

Edited August 31, 2012 by I.M.O.G.

[El Gappo]

Yes but its a bug that tricks the software into thinking it ran faster than it did. It doesn't actually produce higher results in ANYTING, except PCM05, because it bugs out on a setting which has nothing to do with actual performance.

 

But then, nothing in PCM05 has anything to do with actual performance anymore, which is why only a handful of people bother running it anymore.

 

THIS. Favorite bench is ruined

[VapoR.]

it help also in ATTO, 2MB size length, Transfer size >128KB. Write speed grow from 600 to 855MB/s. When i try use other benchmark (HDTune), when i select my RAid Volume i get BSOD. AS SDD cant start.

[Moose83]

It uses ram for caching, but no 3rd Party Programm needed.

Also is legal i think. Areca do same using cach for score.

[xxbassplayerxx]

PCM05 is dead with all of these new "tweaks".

[Strunkenbold]

Simply cap general & virus like hdd startup...

[VapoR.]

PCM05 is dead with all of these new "tweaks".

 

is dead long time ago.

[I.M.O.G.]

Simply cap general & virus like hdd startup...

 

No good. With the super tweaks artificial capping would be a joke. Transparent window, audio/video, memory latency, text edit, and I probably missed one in there... they can all be supertweaked.

[Leeghoofd]

Congratz to the tweakers !! You pOwned us all

[dhenzjhen]

Congratz to the tweakers !! You pOwned us all

 

This ^^

[VapoR.]

Why mtech use only 4 activ core cpu in x6 CPUCore Stage?

[Moose83]

Its allowed, using 6 core and deaktivate 2 cores.

I did the same with bulldozer, using 4 cores.

[Christian Ney]

As long as the CPU is x Core native you can compete in the x Core stage and disable how many Core you want. Unlocking Cores is not allowed for the x Core stages but allow the processor to compete into the x+1 or x+2 stage.

[I.M.O.G.]

Why mtech use only 4 activ core cpu in x6 CPUCore Stage?

 

No subtest can use more than 4 core. If you might clock higher or more stable, 4 cores are best.

[TaPaKaH]

No subtest can use more than 4 core. If you might clock higher or more stable, 4 cores are best.

If so, why can't 4+ core categories be merged?

[I.M.O.G.]

If so, why can't 4+ core categories be merged?

 

They should be in my opinion.

 

However I assume its more complicated than that, because multigpu categories should also be merged. Scores go down with more gpus in various 3d tests.

 

EDIT: To be exact...

 

Covering the 3d tests divided by GPU count:

 

Doesn't scale beyond 1x GPUs and shouldn't have globals in 2x-4x: 3d01, 3d05, aquamark, 3d06

 

Doesn't scale with anything beyond 2x GPUs and shouldn't have globals in 3x-4x: 3d03

 

Doesn't scale with anything beyond 3x GPUs and shouldn't have globals in 4x: Unigine Heaven

 

Scales with 4x and should have globals in 1x-4x: Vantage, 3d11

 

Covering the 3d tests divided by core count:

 

Doesn't scale with anything beyond 4x cores and shouldn't have globals in anything higher: PCM05

 

Doesn't scale (really) with anything beyond 12x cores and shouldn't have globals in anything higher: wprime32m, wprime1024m

 

Scales on everything: UCBench2011 (still doesn't award any globals)

Edited September 2, 2012 by I.M.O.G.

[Massman]

Yes but its a bug that tricks the software into thinking it ran faster than it did. It doesn't actually produce higher results in ANYTING, except PCM05, because it bugs out on a setting which has nothing to do with actual performance.

 

Does the firmware/driver trick the benchmark or is it just boosting the one thing the benchmark is designed to test for?

[Moose83]

The RAID expert Software use ram for caching. This is the same like areca cards do.

[Massman]

Doesn't the Areca have onboard cache?

 

So it's something like MFT or ramcache?

[knopflerbruce]

Areca has onboard cache, usually as a stick of RAM.

 

I don't see any difference, really. The Areca cache and the controller communicate on the same PCB, same as this controller - the only real difference is that the cache of the onboard raid controller is also used by other apps, as that cache also happens to be the system memory.

[El Gappo]

Seems to be no difference between 1 ssd and several acards, no scalling ( or very minimal ) with ram frequency or ddr type either. Would call it a bug but still.. No different to any other 'tweaks'.

 

ATTO would go bonkers if it were normal ram cache.

Edited September 3, 2012 by El Gappo