***The OcUK Graphics Card Performance & Review Thread***

[rjk]

you are missing the point

its a feature of the card

its like me saying that im goint to run the 4870x2 with the fan turned off

its there so why not use it

i would say physx adds around 15% anyway

way i see it is that it is info for buyers - why not account for all the features

 

[Locky]

Just for comparison

Core i7 920 @ 4ghz and 280 GTX Tri SLI.

I get just over 35k in 3dmark Vantage.


1000000000000000 Fps in Crysis, Cod4, Cod5, L4D, and every other game I play on it

 

[rjk]

exactly - but that is because your case is so amazing! eh Locky?

[for the record, Locky's tri sli never has to work that hard because he turns all the detail down so he can see enemies better

- a tip he got from his #1 hero Johnathan “Fatal1ty” Wendel]

 

[Locky]

for the record I have never spoken to him, and hes just a nub quake 3 player and rubbish at everything else

not my fault you vanish from games cus your pc BSOD

 

[layte]

you are missing the point

its a feature of the card

its like me saying that im goint to run the 4870x2 with the fan turned off

its there so why not use it

i would say physx adds around 15% anyway

way i see it is that it is info for buyers - why not account for all the features

No, I am just pointing out that Futuremark do not allow results using non sanctioned features to be published by companies signing up for the professional version of the benchmark. This could even be a violation of the license (not sure but from reading the T&C's it seems likely), but seeing as blatant NDA leaks are the business of the day on this forum I doubt that fact even entered anyone's mind.

 

[Ben M]

you are missing the point

its a feature of the card

its like me saying that im goint to run the 4870x2 with the fan turned off

its there so why not use it

i would say physx adds around 15% anyway

way i see it is that it is info for buyers - why not account for all the features

But with physx not being used by many games, it's a more rounded, more representative benchmark if you turn off physx, surely?

 

[w3bbo]

None of the 3DMark benchmarks give an accurate description of what performance will be in a game. ATI cards do well in 05 and 06, beating the NV equivelents easily but this can be the opposite in games. Anyone who relies solely on 3DMark to purchase their GPU is a fool. It's a benchmark and should be used as such.

Futuremark allowed PhysX innitially, as it was owned by Aegia. Now NV own Aegia and have optimised it's GPU's to calculate PhysX, ATI cry foul and as their top tier manufacturer sponsors Vantage, Futuremark say no fair and don't allow physX to be taken into consideration. Much to the glee of anyone who owns an ATI card lol.

 

[layte]

Futuremark say no fair and don't allow physX to be taken into consideration. Much to the glee of anyone who owns an ATI card lol.

Futuremark were fine with PhysX up until the point Nvidia made their entire graphics benchmark pointless. Hence it being banned from official benchmarks. Much to the disappointment of Nvidia owners everywhere who were suddenly feeling threatened by ATI no longer being a laughing stock.

 

[w3bbo]

Made it pointless by utilising one of it's features? You can hardly blame Nvidia for that lol. Imo it's just sour grapes as Sapphire didn't want their cards to look inferior, especially after no doubt paying a tidy sum to have their emblem on the speedboat.

Since when was Vantage only a graphics benchmark? If that were the case then why does it display graphics and CPU scores at the end?

I couldn't give a damn either way as I said before, it's only a benchmark, nothing to get upset about.

 

[dante6491]

So why not throw in a Dx10.1 benchmark to FutureMark. Oh wait...

 

[Helios1234]

I think the physx dispute is tied wth the fact that dx10.1 implimentation was removed from vantage to prevent the clear advantage ati would have had over nvidia, and yet physx stayed giving nvidia the advantage.

Edit: hehe^^

 

[layte]

Made it pointless by utilising one of it's features? You can hardly blame Nvidia for that lol.

Since when was Vantage only a graphics benchmark?

If that were the case then why does it display graphics and CPU scores at the end?

I thought you were smart enough to realise that the Vantage PhysX results do not give a real indication of either when being able to make use the full capabilities of a GPU. Hence it's use being prohibited in ORB and published commercial benchmarks.



(taking this even more OT, I would put good money that the next 3DMark physics test will also load up the GPU with plenty of graphics work to avoid this situation)

 

[Alpherah]

So why not throw in a Dx10.1 benchmark to FutureMark. Oh wait...

Quoted for truth unfortunatly..

 

[w3bbo]

I thought you were smart enough to realise that the Vantage PhysX results do not give a real indication of either when being able to make use the full capabilities of a GPU. Hence it's use being prohibited in ORB and published commercial benchmarks.

I guess I was mistaken.


(taking this even more OT, I would put good money that the next 3DMark physics test will also load up the GPU with plenty of graphics work to avoid this situation)

No need for the sly derogatory remarks. Clearly this is a very sensitive subject for you.

It's a BENCHMARK. I don't know how clearer I can say this. It does not give a true representation of what to expect in 'real' games regardless of whether PPU is used. Much the same as 06 and 05.

So why not throw in a Dx10.1 benchmark to FutureMark. Oh wait...

I agree, it SHOULD be used. IF 10.1 was incorporated then I certainly wouldn't have issue with that. If it's something that advantages one card because the other is not capable then so be it. To say otherwise would be plain hypocritical. Hence why I think PhysX and much the same Havoc should be used to assess a cards capabilities. If a feature of a card can increase the score (reflecting it's capabilieis in that benchmark) then it's all fair imo - isn't that why we run benchmarks anyway? Or is it simply a case of willy waving?

 

[Rroff]

Would be interesting to see what difference DX10.1 really makes with this generation... as IIRC it was more of a performance advantage for R600 series cores and doesn't really apply so much with the newer ATI cards.

 

[londonbairn]

Fair enough, but it should be an even playing ground, we're comparing GPUs here, PhysX is not utlised in any game worth mentioning and neither is DX10.1.

I think it's going to be more beneficial to try more games out than 3DMark....

 

[dante6491]

Fair enough, but it should be an even playing ground, we're comparing GPUs here, PhysX is not utlised in any game worth mentioning and neither is DX10.1.
.

I thought assassins creed was quite good... I also notice that Far Cry2 and Stalker clear sky take advantage of DX10.1. Whilst not particularly popular here there is a significant fanbase of each.

 

[londonbairn]

I thought assassins creed was quite good... I also notice that Far Cry2 and Stalker clear sky take advantage of DX10.1. Whilst not particularly popular here there is a significant fanbase of each.

Ah do they? Might have to try them out, do I need patches or anything? Must admit didn't really like FC2 and AC I found boring (but looked pretty gorgeous at times with full everything). The PC games of recent I have loved and enjoyed have been Crysis, Crysis Warhead, Fallout 3, all the major valve games & Stalker 2 (like the story). Was disappointed with Crysotasis (grossly overhyped, probably by Nvidia I'd imagine ), Mirror's Edge (boring after 5 mins, sold on the bay straight away more or less

If it makes you feel better about my opinion of you Rroff, if PhysX was going to add a dimension to games I enjoy, I'd have absolutely no problems in relegating the X2 to the second machine and getting myself 2 NV cards

 

[dante6491]

Assassins creed is just the game minus 1.02 patch since it was taken out with that patch. Far cry 2 is from the off I believe, just install and away. Stalker clear sky requires 1.5.06 patch or so I read.

 

[Rroff]

Fair enough, but it should be an even playing ground, we're comparing GPUs here, PhysX is not utlised in any game worth mentioning and neither is DX10.1.

I think it's going to be more beneficial to try more games out than 3DMark....

Hardware physics is part of the benchmark tho... nothing to prevent ATI implementing it... and surely if their stream processors arch is really so good they can take back the crown