"Newer" Driver Heaven HL2 VST benchmark..!?!?!

B

Blad3

Gawd
Joined
Dec 15, 2003
Messages
537
http://www.driverheaven.net/showthread.php?t=53712&page=4&pp=15
http://www.driverheaven.net/showpost.php?p=423994&postcount=51

http://www.driverheaven.net/V3/retest.JPG

er...? That's the 6800Ultra OCed, with 2Gbs, opts on for nvidia and with the 65 driver's
.....vs the x800xt no oc with 1Gb of ram 4.8's :S

What is going on here? Is it a problem with his 6800Ultra??? Or are they totally bias like people say, even though this seems to bias nvidia in terms of the setup...?

EDIT # 1:

HL2 VST results from different sites. I can say now that Driver heaven isn't bias imo:

1600x1200 4xaa, 8xaf, full detail:

Site............. .CPU...X800XT..6800U..6800gt..X800pro

Driverheaven ..FX-53....72........50......-............-
Gamers-depot .P4-3.4...69........55.....50..........51
Firingsquad .....3800+....80........63.....57..........66
Techreport .....3800+....70........56.....51..........51
PCper ........ P4E3.4.....70.........50........................
AnandTech....FX-53.......70.........56.......................
Gamespot.....3000+.......68.........58.........................
Xbitlabs........3400+.......77........64.7.......................
 
Probably worth noting that all optimisations were on for the 65.62 and i tested with vsync disabled in the driver and app and also aa/af enabled in both to check the results were ok no matter what method was used. Finally, my original findings have been replecated by 2 organisations outside DH.
Click to expand...

Heh and where are these magical benchmarkers that got the same results he did. Obviously none of them showed thier faces anywhere :rolleyes:

Just shows ya how much crap spews out of DH :eek:

They probably don't know that when set on highest settings nV's opts are off..... or they know it but choose to ignore it because of thier bias
 
Blad3 said:
True, where are these other site's he talk's about?
Click to expand...


magic ones from the land of Wizard of OZ :D

hehe maybe a independent tester "cough" Gabe Newell
 
With a 6800GT and 65.62 I get 61.11fps at 4xAA/16xAF 1600x1200 in the VST. :rolleyes: Great job there DH.

with 66.00 I get 63.98fps in the same test and settings. :rolleyes: :rolleyes:

edit: A 6800U @ stock should be within 6-7fps of the x800XT PE @ stock at 1600x1200 4xAA/8xAF using the 65.62 drivers, not 20fps as he's showing in that graph.

edit2: don't bother questioning him because he'll flat out ban you and delete the post.
 
pxc said:
don't bother questioning him because he'll flat out ban you and delete the post.
Click to expand...


The funny thing is he is banning people that were just asking for more informations because they weren't getting the same results as he was, they weren't calling him a liar. Its obvious he has something to hide :rolleyes:
 
Maybe he does work for ATi or something, liek rancor for nvidia :p EXcept you dont bias results :)
 
DriverHeaven said:
Seems that other than that Gamers Depot and Firingsquad have roughly the same results as me. So hopefully that resolves things for people calling us bias.

I've asked Valve if 3dc is being used.
Click to expand...

They are the other sites - FiringSquad and Gamers Depot....
 
Blad3 said:
Maybe he does work for ATi or something, liek rancor for nvidia :p EXcept you dont bias results :)
Click to expand...


LOL maybe some slush money, oh yeah I don't work for nV anymore either :), but I do chat with a few friends at nV every now and then :D
 
Blad3 said:
They are the other sites - FiringSquad and Gamers Depot....
Click to expand...


They weren't anywhere near, they have a 28 fps difference from the 6800 ultra and the ati xt pe. none of the other sites had that much of a difference. I think the most was around 15 fps.
 
Newer because they haven't been shown on here, they are his updated one's...
 
rancor said:
LOL maybe some slush money, oh yeah I don't work for nV anymore either :), but I do chat with a few friends at nV every now and then :D
Click to expand...

you're like an nvidia *worker*/drone let alone a basic, lowly "!!!!!!" :p
 
lol and rickcain you know you like this thread, don't try to play down your f@nboyism by trying to get it locked :p :)
 
I think I had slower performance when I specified what level of AA and AF I wanted the stress test to use in the Nvidia CP. I can't remember what the difference was, but I think it was quite a large drop from setting the Nvidia CP to application controlled and just using the application settings. I wonder if that may be his issue since he did say he made sure it was enabled both in the Nvidia CP and in the game itself.
 
Ouch, that's terrible. What kind of motherboard are you using? Are you using Forceware 61.77 or beta drivers? Oh, and what OS and version of DirectX are you running?
 
an XP reinstall should fixed it but I cant find the damn XP CD, it costs a lot too.

HL2 VST results from different sites. I can say now that Driver heaven isn't bias imo:

Site............. .CPU...X800XT..6800U..6800gt..X800pro

Driverheaven ..FX-53....72........50......-............-
Gamers-depot .P4-3.4...69........55.....50..........51
Firingsquad .....3800+....80........63.....57..........66
Techreport .....3800+....70........56.....51..........51
 
Blad3 said:
an XP reinstall should fixed it but I cant find the damn XP CD, it costs a lot too.

HL2 VST results from different sites. I can say now that Driver heaven isn't bias imo:

Site............. .CPU...X800XT..6800U..6800gt..X800pro

Driverheaven ..FX-53....72........50......-............-
Gamers-depot .P4-3.4...69........55.....50..........51
Firingsquad .....3800+....80........63.....57..........66
Techreport .....3800+....70........56.....51..........51
Click to expand...

Anand's tech reports, gamer's depot, Firing squad, Vr-zones, and Tech report are all very close, Also Anand tech's they stated that AA's opts were turned off for nV's cards (someone wrote in and asked them) The average right now is around 15 fps difference. Driver haven is still at 22 thats a big difference thats 15% for the others to 35% for DH. All of the other sites have nV's average higher and ATi xt pe's lower......

Try not to use aa and af, because the engine performance doesn't test that. Thats all drivers
 
Blad3 said:
an XP reinstall should fixed it but I cant find the damn XP CD, it costs a lot too.

HL2 VST results from different sites. I can say now that Driver heaven isn't bias imo:

Site............. .CPU...X800XT..6800U..6800gt..X800pro

Driverheaven ..FX-53....72........50......-............-
Gamers-depot .P4-3.4...69........55.....50..........51
Firingsquad .....3800+....80........63.....57..........66
Techreport .....3800+....70........56.....51..........51
Click to expand...
Most of those tested with 61.77 (except DH), so DH's scores are even worse.
 
The reason they seem biased is because out of all the systems that the other sites tested with, DH should have had the best results possible for the 6800's. The problem is, they got the worst results of anyone, by an average of 10-15% less than the other sites. However, their X800 results managed to stay on par with the results from the other sites. To me that means they are doing something wrong, or something is broken. I won't come right out and say they are getting bad results on purpose, because they could be having issues just like you.

Believe me, my results were worse than theirs when I first ran the benchmark, but after a few driver updates and enabling optimizations, my performance is comparable to the Ultra scores you just listed. I'm running a GT at stock speeds.
 
OK, but I'm not sure if they are the orginal DH results or the newer ones...
 
Tobyus said:
The reason they seem biased is because out of all the systems that the other sites tested with, DH should have had the best results possible for the 6800's. The problem is, they got the worst results of anyone, by an average of 10-15% less than the other sites. However, their X800 results managed to stay on par with the results from the other sites. To me that means they are doing something wrong, or something is broken. I won't come right out and say they are getting bad results on purpose, because they could be having issues just like you.

Believe me, my results were worse than theirs when I first ran the benchmark, but after a few driver updates and enabling optimizations, my performance is comparable to the Ultra scores you just listed. I'm running a GT at stock speeds.
Click to expand...


I just don't see how someone can botch it up so bad where no else gets those results and they can't go back and test thier test method :eek:
 
Yeah, I don't think we are going to get a good set of benchmarks out of that site for this test unless someone can convince them to use a different setup. They know that their results are bad compared to other people, so why not keep changing variables until they find the problem. If they found the actual issue they could even be helpful and let everyone know what the problem was and how to fix it. They could be heroes! :D
 
Maybe the extra RAM doesnt make a difference, same with the OC in the VST? * not much of a difference?

There are still 5 other site's showing results which are similar....just but a few less fps.
 
Uhh, the 6800 series has always been weak when AA and AF are turned on, ATi was always the king of that hill. Turn them off, and ATi will only be edging out nVidia.
 
I've heard that the VTS isn't even a great comparison because it doesn't represent what the game will really do to cards. I could be wrong...

As for this, I could care less. Valve is in ATi's corner and its not a surprise that it runs faster on an ATi card.
 
When are those asshats going to release the VST to normal non stupid enough to buy the filth that is CSCZ people? I want to have at it.
 
Ok....

1) ALL of these benchmarks leave ops off for Nvidia when it should be ON,considering Ati's optimizations are always on (since drivers do not offer to turn them off).
2) HL2 goes gold this monday, so we will have some real numbers to kick around soon.
3) Anyone who doesnt think that this game was specifically tailored to benefit ATi is blinder than a bat.
4) From a videocard owner perspective, I want to see how my card scores on that game given all of the above, but from a gamer's perspective, I couldn't care less. Among the competitive gamer community there are 2 groups of people: CS players and non CS players. I fit into the second group. I prefer my competition to offer strategy, movement, timing.I play UT2k4, Quake 3,and am looking forward to Quake4,which will be based on the d3 engine so obviously OpenGL.
 
I guess it's safe to say the GT only really beats the Pro in one game........ That being D3 of course. Not bad for a 12 pipe card. :D
 
R1ckCa1n said:
I guess it's safe to say the GT only really beats the Pro in one game........ That being D3 of course. Not bad for a 12 pipe card. :D
Click to expand...


lol only in the mind of DH :D
 
R1ckCa1n said:
I guess it's safe to say the GT only really beats the Pro in one game........ That being D3 of course. Not bad for a 12 pipe card. :D
Click to expand...

but pretty shitty for a $400 card...
 
Optimummind said:
lol, so true. In case anyone wants to know what that biased asshole looks like, here is a pic:
Click to expand...

Could be biased but don't need to call him an asshole
 
R1ckCa1n said:
I guess it's safe to say the GT only really beats the Pro in one game........ That being D3 of course. Not bad for a 12 pipe card. :D
Click to expand...

Actually, from every comparison i have seen, the GT is ahead of the Pro in basically every bench EXCEPT the HL2 VST. And the Pro only EQUALS the GT in this test. So, as far as bang for the buck goes, the GT basically wins everytime. This is before overclocking is even brought into the equation, which favors the GT because it has the full 16 pipes of the Ultra. Not arguing or flaming, just giving a few facts. Both cards are nice, but i obviously felt the GT was the better deal. (See sig.)

As far as this whole debate goes over the HL2 VST benches, i couldn't care less. Am i the only person who looked at all the benchmarks and said, "Alright! It looks like i will be able to run HL2 at 1600 X 1200!" ? I didn't even notice the other benchmark scores. Because, why should i care, as long as my card can run it at the top resolution with good frame-rates? It doesn't really matter to me if someone else gets 8 fps more than me. I would only be upset if i had to lower my resolution or turn off some effects in order to play it smoothly. That doesn't appear to be the case, so yippee! Bring on the gaming goodness! :)
 
You must log in or register to reply here.
Back
Top