Half Life 2 Benchmarks @ HardOCP.com

[FrgMstr]

And sorry for skipping out on the thread for so long, I had to fly out to San Fran for the day yesterday.

 

[^eMpTy^]

OK.
You use, 9800 pros (video game cards) for rendering?
Uhm, and you don't see a problem here?
Should not QUADRO or FIREGL or 3DLABS be in your boxes?
and you expect ati to have written a driver for an UNFINISHED os?
Does nvidia do this? Did nvidia just come out with "Fedora3 drivers"?
Part ot the big problem with supporting *nix is there are so damn many of them!
http://www.nvidia.com/object/linux_display_ia32_1.0-6629.html
I see no mention of Fedora3
yeah.. the driver there should probably work..just like the ati ones SHOULD work.
But damn.. asking any hardware vendor to write for unfinished operating systems is like asking me to pick a motherboard for a 8 ghz pentium 6 cpu.

Here's the thing, nvidia just works on linux...ati, doesn't...you can argue it any way you want, nvidia linux support is light years beyond ati...

I don't think it matters if they are pro cards or not, the guys that use it are making direct use of the hardware to do computational type stuff...personally I don't even understand it...

 

[FrgMstr]

What we're you doin there mr?

I could tell you, but then I would have to kill you.

 

[^eMpTy^]

from another thread:

But for most of those 15-20 minutes you're limited by, and thereby testing, the CPU rather than the various GPUs, and that's not the intention of a GPU benchmark is it? As a consequence, the GPU candidates cannot be ranked based on your study. You claim that you are " facing off NVIDIA’s and ATI’s latest AGP video cards in Half Life 2", but your choice of method fails to do that. What you have done is not a ranking/screening test, but a performance test, which in itself is very interesting.

One of the ground rules for experimental screening of candidates is to eliminate any parameters constraining or accelerating the phenomenon to be studied beyond the measurable range. For example, when ranking the corrosion resistance of different metallic alloys one would attempt to chose testing conditions that give severe corrosion on the worst alloy, some corrosion on the intermediates and no corrosion on the best alloy (rather than no corrosion on all alloys or massive corrosion on all alloys).

I think the above principle applies for CPU ranking tests as well, but not necessarily for CPU performance tests.

if we wanted graphical worst case scenarios we would run the VST...but people don't care how fast the VST goes, they want to know about the gaming experience...

 

[FrgMstr]

if we wanted graphical worst case scenarios we would run the VST...but people don't care how fast the VST goes, they want to know about the gaming experience...

True dat.

 

[TheRapture]

I'll just assume it has something to doto the san fran sterotype

Oh oh...smells like someone stepped into a pile of...shit.

 

[Jonsey]

I'll just assume it has something to doto the san fran sterotype

Anyone want to bet what Moloch's new custom title will be?

 

[Moloch]

Oh oh...smells like someone stepped into a pile of...shit.

actually I did a few days ago.. wtf is up with people not cleaning up their dogs shit?

 

[darksynth]

Sadly, i must say that i am quite disappointed in this. No rad 9800? There are millions of us out there that are not ready to upgrade yet and have this card. this makes this artical useful for those who have already bought, but it gives me absolutely no refrence point of where i am and where i might go with an upgrade. Useless.

I hope you guys include older gen. cards, even if it is a preview. Just 1 is needed.

 

[CastleBravo]

Hee hee, this is all very entertaining.

I'd suggest that if making the benchmark more like actual gameplay handicaps your product, it is your product that needs to be reconsidered, not the benchmark. Especially when the CPU is an FX-53 and the benchmark is running at 1600x1200 4xAA 8xAF, and you still lose because of being "CPU limited." With an FX-53? At huge resolution and lots of AA and AF? Ooh, persuasive.

What CPU do you need to beat a 6800 GT with an x800 Pro? An FX-57? A P4 4.0EE? Oops, those don't exist yet.

Funny how benches you lose are always the fault of some subterfuge, while every bench you win is legit and held up as proof of how great your product is... even though the former outnumbers the latter for the x800 Pro by about 10:1. That's one hell of a conspiracy against your product. I'd blame the Illuminati.

Back to your regularly scheduled hype, already in progress...

 

[pxc]

I guess there is no way to speak logically to some people that are ruled by blind fanaticism.

a mirror might shock you.

 

[Visable-assassin]

Only?
How about any heavy DX9 game?
Far cry anyone?
The problem is there isn't enough DX9 games, and even far cry I've heard has mostly SM1 shaders.

I guess I am one of the few that played FarCry and it didnt get me all hot and bothered and just uninstalled it and its now sitting on a shelf. I dunno but for whatever reason the "ohhhh ahhh" factor stayed for about the first level. To me FarCry doesnt mean a damn thing. Is it pretty? sure. Does it rule the way I should purchase anything? No.

 

[FrgMstr]

a mirror might shock you.

OK, both of you, that shit is over right here right now. Let's all get back to being adults. It was going along so well.

 

[FrgMstr]

Sadly, i must say that i am quite disappointed in this. No rad 9800? There are millions of us out there that are not ready to upgrade yet and have this card. this makes this artical useful for those who have already bought, but it gives me absolutely no refrence point of where i am and where i might go with an upgrade. Useless.

I hope you guys include older gen. cards, even if it is a preview. Just 1 is needed.

A 9800 series will roughly give you 6600 like performance. I think we did mention that in the conclusion, but data was not given due to time constraints.

 

[SocketA]

"Here's the thing, nvidia just works on linux...ati, doesn't...you can argue it any way you want, nvidia linux support is light years beyond ati..."--Empty--



If you know the support from Linux sucks? And by the way Nvidia's support is far from being admirable. Why don't you just dual-boot? That's what big partitioned hardrives are for. --Sorry, off topic--

 

[Moloch]

I guess I am one of the few that played FarCry and it didnt get me all hot and bothered and just uninstalled it and its now sitting on a shelf. I dunno but for whatever reason the "ohhhh ahhh" factor stayed for about the first level. To me FarCry doesnt mean a damn thing. Is it pretty? sure. Does it rule the way I should purchase anything? No.

I played it for a few levels, but I dont like fps games, so I uninstalled it, but I know of people who like fps and like it alot, but I hear the multiplayer is lacking.

 

[cjm18]

Anyone using the beta nvidia drivers and notice a difference( bad or good)? I have a geforce 6600gt agp. Im going to play hl 2 pretty soon.

 

[^eMpTy^]

"Here's the thing, nvidia just works on linux...ati, doesn't...you can argue it any way you want, nvidia linux support is light years beyond ati..."--Empty--

If you know the support from Linux sucks? And by the way Nvidia's support is far from being admirable. Why don't you just dual-boot? That's what big partitioned hardrives are for. --Sorry, off topic--

umm...or I could use an nvidia card and not dual boot? hmm...let me think about this for a minute...

 

[FrgMstr]

PS: Half-Life 2 runs fine on my old P3E 650mhz on a 440bx mobo with a 32mb GF2 MX/400 and AWE32 soundcard and with 768mb pc100 cas2 ram

That is awesome. What quality settings are giving you good playability?

 

[CATALYST MAKER]

Hee hee, this is all very entertaining.

I'd suggest that if making the benchmark more like actual gameplay handicaps your product, it is your product that needs to be reconsidered, not the benchmark. Especially when the CPU is an FX-53 and the benchmark is running at 1600x1200 4xAA 8xAF, and you still lose because of being "CPU limited." With an FX-53? At huge resolution and lots of AA and AF? Ooh, persuasive.

What CPU do you need to beat a 6800 GT with an x800 Pro? An FX-57? A P4 4.0EE? Oops, those don't exist yet.

Funny how benches you lose are always the fault of some subterfuge, while every bench you win is legit and held up as proof of how great your product is... even though the former outnumbers the latter for the x800 Pro by about 10:1. That's one hell of a conspiracy against your product. I'd blame the Illuminati.

Back to your regularly scheduled hype, already in progress...

I assume this post is directed at me? Shame...

I guess I best be moving along

 

[pxc]

These two reviews have compared the 9800pro to the 6600gt.

http://www.anandtech.com/video/showdoc.aspx?i=2277&p=3

this review has HL2 benchmarks, not just CS:S/VST benchmarks: http://www.anandtech.com/video/showdoc.aspx?i=2278&p=3

no AA/AF
AT_canals_08, 6600GT wins by 10-12fps @ 1024x768-1600x1200
AT_coast_05, 9800 Pro wins by 0-3fps @ 1024x768-1600x1200
AT_coast_12, 6600GT wins by 6-13fps @ 1024x768-1600x1200
AT_prison_05, 6600GT wins by 9-13fps @ 1024x768-1600x1200
AT_c17_12, 9800 Pro wins by 4fps @ 1024x768, 6600GT wins by 1-3fps @ 1280x1024 and 1600x1200

Anand didn't run 9800 Pro benchmarks with AA/AF on.

 

[TheLonelySwedish]

I wanted to extend a quick thanks to kyle and brett for not throwing us some bullshit benches. DH's stuff is embarassing, in my opinion, and it diminishes their credibility. Keep up the good work, and if you could dig up a 9500 pro for some benches that would be awesome.

 

[coz]

I guess I best be moving along

Whoah!! Don't leave because of the asshole minority, stay for the appreciative majority!! Most guys here (myself included) love to hear what a real expert has to say. Ok, you work for ATi and some guys won't trust you - personally I think you're too smart to lie about this stuff so I do trust you!! We're up to our necks in people who *think* they know what they're talking about (myself included!) but very few who *know* what they're talking about.

So don't leave, or else you'll be chalking up another victory for the phanboys.

Pretty please?

 

[TheRapture]

Whoah!! Don't leave because of the asshole minority, ....We're up to our necks in people who *think* they know what they're talking about (myself included!) but very few who *know* what they're talking about.

So don't leave, or else you'll be chalking up another victory for the phanboys.

Pretty please?

I second that motion....give the idiots a cream squirting man banana for good measure and get back to business...

 

[batty]

Eh, CastleBrave brought up a good point. I really don't buy this CPU limited whining on a FX-53 at max rez/settings.

 

[pahncrd]

Whoah!! Don't leave because of the asshole minority, stay for the appreciative majority!! Most guys here (myself included) love to hear what a real expert has to say. Ok, you work for ATi and some guys won't trust you - personally I think you're too smart to lie about this stuff so I do trust you!! We're up to our necks in people who *think* they know what they're talking about (myself included!) but very few who *know* what they're talking about.

So don't leave, or else you'll be chalking up another victory for the phanboys.

Pretty please?

I would prefer an unbiased expert, personally.

 

[coz]

Eh, CastleBrave brought up a good point. I really don't buy this CPU limited whining on a FX-53 at max rez/settings.

His question may have been a valid one but I don't agree with way he went about asking it. Guys like Terry deserve a bit of respect, they don't deserve to be treated like bloody phanboys. He brings considerable expertise to the discussion and provides a valuable link between the community and ATi. He may be here 'selling' ATi to us (that's part of his job) but he's way too smart to feed us BS and his comments don't deserve to be dismissed as 'hype' based upon notions of bias.

 

[CastleBravo]

Whoah!! Don't leave because of the asshole minority, stay for the appreciative majority!! Most guys here (myself included) love to hear what a real expert has to say. Ok, you work for ATi and some guys won't trust you - personally I think you're too smart to lie about this stuff so I do trust you!! We're up to our necks in people who *think* they know what they're talking about (myself included!) but very few who *know* what they're talking about.

So don't leave, or else you'll be chalking up another victory for the phanboys.

Pretty please?

Well, thanks for giving me the "asshole" tag for pointing out the obvious. My feelings are SOOOO hurt, man.

I'm pretty sure he's just talking about this thread, I don't think he's such a wuss that he'd leave THE ENTIRE FORUM just because somebody subjected him to a rather mild mocking.

I have yet to hear anything remotely approaching an attempt to refute my points, incidentally. Go figure.

 

[coz]

I would prefer an unbiased expert, personally.

Well, all product experts work for someone, do we accuse them ALL of bias and get rid of them? If an NV engineer was here would you tell him you're not interested in his biased opinions? I prefer to listen to experts and then make my own mind up.

 

[NEVERLIFT]

His question may have been a valid one but I don't agree with way he went about asking it. Guys like Terry deserve a bit of respect, they don't deserve to be treated like bloody phanboys. He brings considerable expertise to the discussion and provides a valuable link between the community and ATi. He may be here 'selling' ATi to us (that's part of his job) but he's way too smart to feed us BS and his comments don't deserve to be dismissed as 'hype' based upon notions of bias.

I have been around the Rage3D forums since the original Radeon 7200 and own a AIW Radeon 7200 Damn good card!
And ole Terry is ok in my book but he is biased so take what he says with a grain of salt
I have seen alot good posts by him and the fact he takes time to post on alot the forums and seems to enjoy is good! We need more people in the industry like him getting involved in the forums.

PS: Terry you need something to match the 6600GT in its price range ok.

 

[coz]

Well, thanks for giving me the "asshole" tag for pointing out the obvious. My feelings are SOOOO hurt, man.

Heh, I wasn't referring to anyone in particular so consider your feelings un-hurt.

Like I said, your question is a valid one. But I didn't like the way you presented it using a mix of sarcasm and derision. If we treat guys like Terry with a little respect then they may stick around to answer our questions instead of leaving because they think we're a bunch of squabbling kids.

 

[pahncrd]

Well, all product experts work for someone, do we accuse them ALL of bias and get rid of them? If an NV engineer was here would you tell him you're not interested in his biased opinions? I prefer to listen to experts and then make my own mind up.

Certainly.
Besides, it is not what he says, but how he says it.

I really don't care which card is the fastest by a few FPS, what matters to me are the features I have already mentioned. I will certainly buy whichever card has them.
It's really idiotic to restrict yourself to one company, unless that company has gone out of its way for you before.

I must say that ATi has never implemented a feature that I expressed interest. If you look on Rage3D, you will see large threads dedicated to SSAA and stereo 3d. What features do we get though? How about some smartshader, and temporal AA . I must admit 3dc looks tempting, but I don't think it will make much of a difference this gen.

I forgot to mention that the MAC drivers had SSAA and the linux drivers had Stereo 3d. Lame

 

[coz]

I must say that ATi has never implemented a feature that I expressed interest.

Well I expressed a LOT of interest in the 9700Pro's feature set (the full DX9 spec etc.) when it was released and that's why I bought one and ran it for 2 very happy gaming years!!

 

[Moloch]

Well I expressed a LOT of interest in the 9700Pro's feature set (the full DX9 spec etc.) when it was released and that's why I bought one and ran it for 2 very happy gaming years!!

QFT, SM3 is SOO MUCh better than SM2B

 

[bizzy420]

moloch sm3 is ALOT better?
i guess it helped alot in farcry huh....sm2 on ati's still smoking nvidias sm3

 

[spruce]

I hate to bring this up, but I'm going to. I'm wondering if the time demo in doom 3 is the same sort of benchmark that the short, more specific(ati) hl2 benchmark/demo is (favoring one card much more than the other)?

just asking because I really dont know

 

[fallguy]

IIRC..

it was before the source code was 'stolen' and an entire 'rewrite' was done.. so .. those claims could pertain to the 'original' half life 2..

Then they seem to be more than correct. Going by Anand's 5900XT's part of the review. Its horribly slow, compared to any ATI card, including the 9600XT.

I dont see why CrimandEvil thinks Game is "talking out of his @ss", when he said ATi is about 40% faster.

I would prefer an unbiased expert, personally.

Having an expect, bias or not, is better than someone whos bias and doesnt have a clue what they are talking about.

Dont go, CM. :/

 

[CATALYST MAKER]

Dont go, CM. :/

No, I think its best I leave this thread. My job is to make CATALYST drivers as good as possible and to bring useful features like smartgart, vpu recover, overdrive, etc...

I wasn't implying I would leave these forums altogther, but I will stick to the CAT related forums, where I know for sure I can help people (either with problems, or explanations of why we do the things we do). This thread is great though, keep the spirit alive guys. I just realize that anything I post will be taken as biased (which if you know me you will really realize that I'm not). I only spoke up because I was asked to by some members of this forum.

Post any questions or problems in the ATI/CAT forums, I will be sure to pop in there when I get a chance.

 

[NEVERLIFT]

Hope my post did not insult you dear Sir, it was meant to be a logical contribution rather than offensive or biased...

I can assure you that I've read every word in the article, actually that's how I learned that the HL2 performance was CPU limited in most cases. I also understand that some data points taken at the highest resolution and quality are not CPU limited. Looking forward to your tests with midrange/older cards, for which I anticipate most of the avg fps data will strongly reflect GPU limitations.

However (I cannot let it go yet), the full title of the article is: "Half Life 2 Benchmarks : Preliminary framerate benchmarks facing off NVIDIA’s and ATI’s latest AGP video cards in Half Life 2 using our own custom made timedemos compared apples to apples."

"Benchmarking" or not, I will argue that the above title advertises some kind of competition between ATI and Nvidia, using the phrase "facing off" which I interpret to be some kind of confrontation between two parts. And this is not a bad thing, competition is one of the main driving forces for development and implementation of new technology.

Or maybe I have not perceived the correct meaning of the title (English is not my native language), and the title should be interpreted like "ATI/Nvidia cards (on one side) facing off the challenge of running HL2 (on the other side)"?

You gonna need a bigger test line dear sir to bait this one.

And yes there is a face off between ATi and nVidia... there always has been and always will be. Thats why games like Doom3 and Half-Life2 have different rendering paths and code for each... I guess Half-Life2 needs a logo popping up when it starts, saying ATi the way it should be played (hehehe)


EDIT:
I keep hearing talk of games coming out that wont even work on both cards. Like you buy the game and it only works for nVidia or either ATi... did not one of them MedalGear Solid games already come out and not work at all with ATi? Been a while, but I do recall this.

 

[NEVERLIFT]

No, I think its best I leave this thread. My job is to make CATALYST drivers as good as possible and to bring useful features like smartgart, vpu recover, overdrive, etc...

I wasn't implying I would leave these forums altogther, but I will stick to the CAT related forums, where I know for sure I can help people (either with problems, or explanations of why we do the things we do). This thread is great though, keep the spirit alive guys. I just realize that anything I post will be taken as biased (which if you know me you will really realize that I'm not). I only spoke up because I was asked to by some members of this forum.

Post any questions or problems in the ATI/CAT forums, I will be sure to pop in there when I get a chance.

Sorry to burst your bubble Terry but everyone I know that has a ATi card and me included turn off vpu recover(its buggy and not stable). And we hate dumbgart(we use our bios and tend to know better what settings should be used... and there are alot posts on the Rage3D forums about how bad smartgart is)
And PowerStrip, Rage3D Tweak and RivaTuner overclock videocards and unlock alot more stuff
PS: Alot my friends are still even using the Cat 3.10's cause the newer drivers are not stable for them.(this is with 9800pro's).