• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

3DMark2K5 1.1 patch/hotfix out

Mister E said:
My score went up 12 points to 5724. Whatcha got to say to that Mr. I'm pissed cause my GT is getting owned?

DoomIII was not the say all end all in the graphics war. The sooner you realize that Id and Nvidia pulled one over on you guys the better.

I'm SO GLAD I didn't fall for the hype and buy a 'Built For D3' card. :rolleyes:
Click to expand...

I hope you arent comparing a XT PE to a GT. XT PE = $500+, GT = $399. Wow, the gameplay must be so improved on 3d05 now that youve broke 5700. Tell us mortals what its like :rolleyes:
 
Graphics cards aren't the only devices that get benchmarked believe it or not. Most testing is done synthetically. How many network providers do you think would get away with live testing on their production networks? How about hard drive transfer tests.. updated chipset/device drivers would never effect how the device performs under testing.... Seems like a rather reactionary argument to me.
 
Summoner said:
I hope you arent comparing a XT PE to a GT. XT PE = $500+, GT = $399. Wow, the gameplay must be so improved on 3d05 now that youve broke 5700. Tell us mortals what its like :rolleyes:
Click to expand...


It's better and more exciting than your GT (a.k.a. crippled 6800U) will EVER understand! :D
 
Brent_Justice said:
Notice I emphasized "Synthetic" benchmark.

A true Synthetic benchmark that is supposed to show how powerful your video card is should not be swayed by driver updates or application updates that changes the score up or down.

A game is different from a "synthetic" test.
Click to expand...
hmm, I don't quite follow you there. The driver is part of the card, without it the card wouldn't work. A good driver could make the hardware work more efficient and by that increase the score of a benchmark. Could you please explain to me why you think a driver shouln't affect the score of a synthetic benchmark?
 
Mister E said:
So are you going to do any investigation into this Brent? Any kind of quality/driver comparison? Any kind of patch test? Or are you just gonna sit there and bash Futuremark because you don't like it?
Click to expand...

I have better things to do then to spend time investigating 3DMark05, things that actually mean something to gamers.

I don't bash Futuremark, and I never said I didn't like 3DMark05. In fact you will find posts by me in the last few days explaining how much I do enjoy it as a 3D demo. The graphics are very pretty, and I would really like games in the future to have graphics of that level, it really is a very solid 3D demonstration of 3D graphics.

But an evaluator of game performance produced by a video card it is not.
 
Brent_Justice said:
I have better things to do then to spend time investigating 3DMark05, things that actually mean something to gamers.

I don't bash Futuremark, and I never said I didn't like 3DMark05. In fact you will find posts by me in the last few days explaining how much I do enjoy it as a 3D demo. The graphics are very pretty, and I would really like games in the future to have graphics of that level, it really is a very solid 3D demonstration of 3D graphics.

But an evaluator of game performance produced by a video card it is not.
Click to expand...

The part that amazes me is that you are blanketing this statement over and over.

3DMark05 "is" an evaluator of graphics related game performance. It may not match your ideal, but it does still perform it's function.
 
Brent_Justice said:
I have better things to do then to spend time investigating 3DMark05, things that actually mean something to gamers.

I don't bash Futuremark, and I never said I didn't like 3DMark05. In fact you will find posts by me in the last few days explaining how much I do enjoy it as a 3D demo. The graphics are very pretty, and I would really like games in the future to have graphics of that level, it really is a very solid 3D demonstration of 3D graphics.

But an evaluator of game performance produced by a video card it is not.
Click to expand...
Maybe in the next version they can put Brent in the credits like Hilbert and Lars :p
 
Brent_Justice said:
I disagree.
Click to expand...


Ditto, there are a hunderd different ways to write code for a game, no one game or benchmark can show you relative performance over a wide gamit of products.

3dmark05 uses a generic build shaders at load time, no other game use this or will use it because hand optimized shaders perform better and are better suited for different types of games.
 
Brent did you miss my post? It was posted just before you answered Mister E so maybe you didn't see it. Im really interested in why you don't think a driver should affect the score of an synthetic benchmark (see my post above).
 
Brent_Justice said:
I have better things to do then to spend time investigating 3DMark05, things that actually mean something to gamers.

I don't bash Futuremark, and I never said I didn't like 3DMark05. In fact you will find posts by me in the last few days explaining how much I do enjoy it as a 3D demo. The graphics are very pretty, and I would really like games in the future to have graphics of that level, it really is a very solid 3D demonstration of 3D graphics.

But an evaluator of game performance produced by a video card it is not.
Click to expand...
3Dmark 03 showed us how the R300 wasn't good at stenciled shadows, and guess what?
It wasn't(see doom3).
It also shows the 5XXX FX series slow as hell, which they are(see source engine benchmarks).
It all stems from nvidia calling it crap, then you guys followed(kyle being nv's right hand man, not so much anymore) and now you have to open a door to your egos past :p
 
Moloch said:
3Dmark 03 showed us how the R300 wasn't good at stenciled shadows, and guess what?
It wasn't(see doom3).
It also shows the 5XXX FX series slow as hell, which they are(see source engine benchmarks).
It all stems from nvidia calling it crap, then you guys followed(kyle being nv's right hand man, not so much anymore) and now you have to open a door to your egos past :p
Click to expand...


It does but then driver tweaks covered those problems up and this is why driver updates shouldn't really effect a systhetic benchmark but they do don't they.
 
Yes, it's like an un-named company releasing scores with a yet to be released set of cards; only to be beat by a single card solution. :rolleyes: This app must be important to someone.
 
rancor said:
It does but then driver tweaks covered those problems up and this is why driver updates shouldn't really effect a systhetic benchmark but they do don't they.
Click to expand...
If it effects a game, it should effect the benchmark, if its going to be covered up by driver hacks and such in a game, isn't what that what 3dmarks should be?
Emulating games in the future.
 
rancor said:
It does but then driver tweaks covered those problems up and this is why driver updates shouldn't really effect a systhetic benchmark but they do don't they.
Click to expand...

The only way a synthetic benchmark would be able to accomplish what you and Brent are asking for is to remove the OS and driver from the picture. The bench would have to run purely on HW and execute raw instructions. While that would be another perhaps more "pure" way of testing it wouldn't be any better as games do not run in this environment anymore.
 
Moloch said:
If it effects a game, it should effect the benchmark, if its going to be covered up by driver hacks and such in a game, isn't what that what 3dmarks should be?
Emulating games in the future.
Click to expand...


If its being covered up it won't, in games its a bit different, since its more variable in games, even a driver tweak will not be effective in a situations. Unless its a complete change of shader or what not.
 
rancor said:
It does but then driver tweaks covered those problems up and this is why driver updates shouldn't really effect a systhetic benchmark but they do don't they.
Click to expand...

Then how do you explain why each revision of NV drivers tested in this thread HERE show improvements in a "systhetic" benchmark?
 
a patch already? it came out when? wednesday?

i'd rather wait for something good and polished than rushed and broken.
 
rancor said:
If its being covered up it won't, in games its a bit different, since its more variable in games, even a driver tweak will not be effective in a situations. Unless its a complete change of shader or what not.
Click to expand...
Erm a huh :rolleyes:
Ati seems to do well with their application detection hacks :D
 
Moloch said:
Erm a huh :rolleyes:
Ati seems to do well with their application detection hacks :D
Click to expand...

Not until the next set of drivers will we see true NV type ATI drivers. :p
 
hamm3rhead said:
The only way a synthetic benchmark would be able to accomplish what you and Brent are asking for is to remove the OS and driver from the picture. The bench would have to run purely on HW and execute raw instructions. While that would be another perhaps more "pure" way of testing it wouldn't be any better as games do not run in this environment anymore.
Click to expand...
Exactly my thought!
 
R1ckCa1n said:
Not until the next set of drivers will we see true NV type ATI drivers. :p
Click to expand...


Either way ATi won't be able to shed the fact that nV is outperforming them without aa and af, and if nV starts doing shader replacements too, well there will be alot more differences on the horizin. So far there hasn't been any shader replacements for what we have done with our game on the Far Cry engine (we have changed a the shader subsystems) if thier was any shader replacement going on in Far Cry they would have been rendered useless.

It is almost impossible to do shader replacement without differences, there are a million ways to write a shader, but only one way that will give optimal performance and optimial visual representation, regardless of the architecture it runs on since its API dependent.
 
Starts doing?
Nvidia is doing shader replacements in doom3, even carmack said so, who plays without AA and AF, and why do they have a 400 doller graphics card?
The Pro and XT-PE handle heavy loads better...
 
rancor said:
Either way ATi won't be able to shed the fact that nV is outperforming them without aa and af, and if nV starts doing shader replacements too, well there will be alot more differences on the horizin. So far there hasn't been any shader replacements for what we have done with our game on the Far Cry engine (we have changed a the shader subsystems) if thier was any shader replacement going on in Far Cry they would have been rendered useless.
Click to expand...

It's not "IF" nV starts doing sharder replacements...... What do you think nV does with their detection of over 20 games right now in their drivers? lol
 
Moloch said:
Starts doing?
Nvidia is doing shader replacements in doom3, even carmack said so, who plays without AA and AF, and why do they have a 400 doller graphics card?
The Pro adn XT-PE handle heavy loads better...
Click to expand...


He never stated it, it was assumed
 
R1ckCa1n said:
It's not "IF" nV starts doing sharder replacements...... What do you think nV does with their detection of over 20 games right now in their drivers? lol
Click to expand...


Benchmarkers don't use the control panel game settings do they?
 
Moloch said:
You dont have to assume nvidia is doing replacements,I guarantee they are.
Click to expand...


Your garuntee is not good enough, there has to be proof behind a statement like that.
 
rancor said:
Your garuntee is not good enough, there has to be proof behind a statement like that.
Click to expand...
Shut up dude lol, what makes you think they aren't?
 
rancor said:
Benchmarkers don't use the control panel game settings do they?
Click to expand...

Nope they don't... But like you mentioned earlier, drivers shouldn't effect scores.

66.72: 5,119
66.70: 5,129
66.51: 5,032
66.32: 4,622
66.31: 4,649
66.29: 5,003
 
R1ckCa1n said:
Nope they don't... But like you mentioned earlier, drivers shouldn't effect scores.

66.72: 5,119
66.70: 5,129
66.51: 5,032
66.32: 4,622
66.31: 4,649
66.29: 5,003
Click to expand...

Sorry, that's not real unless you do the same on doom3 :p
 
R1ckCa1n said:
Nope they don't... But like you mentioned earlier, drivers shouldn't effect scores.

66.72: 5,119
66.70: 5,129
66.51: 5,032
66.32: 4,622
66.31: 4,649
66.29: 5,003
Click to expand...


Exactly, I'm not saying ATi's only getting performance increases. But it comes down to is 3dmark05 doesn't have any bearing on futrue games.
 
kcthebrewer said:
That has to be one of the funniest posts I have seen in a long time :)
Click to expand...
You realize it used to be true right, but somewhere along the way he realized he was being puppeted by nvidia.
Right around the time nvidia starting making noise about 3dmark not being a good benchmark, coincidently kyle writes a big article about the same thing.
Hmm..
 
rancor said:
Exactly, I'm not saying ATi's only getting performance increases. But it comes down to is 3dmark05 doesn't have any bearing on futrue games.
Click to expand...


nice edit dude !
 
You must log in or register to reply here.
Back
Top