Half Life 2 Benchmarks (From Valve)

Dave Baumann

Gamerscore Wh...
Moderator
Legend
hl21.gif

hl22.gif

hl23.gif

hl24.gif


* Note: "Radeon 9600" is actually Radeon 9600 PRO
** Further Notes: Resolution is 1024x768, no extra feature utilisation (AA/AF). Demo used is all the E3 demos, with the results averaged.
 
I really want to see AA/AF numbers, as well as what system these performance numbers were taken from. Any idea when an "in-depth" review will be released Dave?

p.s.: Is it me, or are those numbers of the Ti4600 vs the 5900 in DX8 mode really damning. Only 15% faster?? :oops:
 
I'm astonished the latest high-end hardware barely gets playable framerate with maximum settings at 1024x768 without anisotropic filtering or antialiasing.
 
Yet another dx9 game that shows us exactly how right on the money the early dx9 synthetic benchmarks were.

I can't imagine the number of people who bought the 5900 that wouldn't have if they had only known how accurate those early dx9 synthetics were. Hard for them to believe they were accurate with all the people denouncing the dx9 synthetics though.

$400 later the truth rears it's ugly head. I hope certain websites lose a lot of their readership after the disservice they did to consumers. I also hope they realize by this point just how much they were duped by the Nvidia pr machine.
 
Now if only the 'synthetic benchmark in question' would have stuck to their guns, instead of folding. One of the biggest titles of all time for the PC and the numbers don't lie, and no BS kiss and make up apology letters and 'optimizations' :devilish:

I for one would much rather see IHVs work on improving this benchmark, since it will help gameplay....and they don't allow 'cheating'...or Half-life 2 approved drivers ;)
 
This benchmark is irrelevant. Resistance is futile.

Someone is going to say very soon, 'perhaps ATI tweaked their drivers for this benchmark/game.' Or words to those effect.

Either way this is pretty damning evidence that HL2 really owns the competition on ATI DX9 gfx cards. I speculate one of the major reasons why is that when HL2 was being developed it was ATI who had the tech ready at a time when Valve began optimising their engine for the highend parts and assumed NVIDIA would do something similar to the architecture of the Radeon 9700 Pro. Evidentally they did not and in this game at least NVIDIA suffers heavy losses.

Or it is simply because ATI have much better technology with regards to the new fangled DX9 thingymibobs.

Edit: Forgot to state the obvious in red.
 
Should I be enjoying this as much as I am...it just feels so WRONG somehow... :LOL:

This should FINALLY put an end to the bloody arguments at least. 8)
 
http://www.beyond3d.com/forum/viewtopic.php?p=155677&highlight=trend#155677

Chalnoth said:
already posted in regards to this.

I'll post again:
In order for a general trend to be apparent, it must be visible in at least three games that are primarily designed for the PC that make use of DX9-level shaders (OpenGL or Direct3D, but since I expect differences between the API's, it would be better if all three used the same API).

So...do we have a trend here yet? 8)

Major....MAJOR coup for ATI.

Guess what benchmark everyone and their Uncle will be using in this fall's new product launches?

Another related prediction of mine: any bets on whether or not a 64 bit 9600 will beat the crap out of any 5200?

And as bad as this is for nVidia....this can be catastrophic for 3DMark. Futuremark had better do something miraculous with their "later this month" relase of their assesment of drivers / optimization guideline findings.

Looks like the "Bias Meter" in my sig is pretty spot on...
 
I'm going to mirror what has already been stated, simply because it should be reiterated.

Those (Kyle, you know who you are) who insisted that synthetic benchmarks were basically worthless because they do not have any bearing any "real" games should stand up and admit the error in their ways.

So far, there has been indisputable proof that ATI's DX9 hardware is vastly superior to anything from nVidia, in terms of doing what it was actually designed to do. Take every single benchmark that exists thus far, remove the ridiculous optimizations that nVidia has been working very hard on, and it becomes even more so.

Anyhow, all numbers point to ATI having the ultimate DX9 gaming hardware. As was pointed out by somebody above, there are a _lot_ of people who have/were duped by the PR machine at nVidia.

Truth be told, I really wish the guys @ Valve hadn't wasted any time whatsoever in optimizing the NV3x path at all. I would have much rather they spent more time on the game and/or released it sooner. If their hardware sucks too badly to run, put the ball back in nVidia's corner and let them do whatever they have to do in order to get their performance up to scratch.
 
Anyone else feel like they are seeing a really strange episode of the X-Files?.....

The truth is out there......

And Valve is gona set it free!

;)
 
Tahir said:
This benchmark is irrelevant. Resistance is futile.

Someone is going to say very soon, 'perhaps ATI tweaked their drivers for this benchmark/game.' Or words to those effect.

I can see it now:

"Don't use real games....use synthetic benchmarks!" :rolleyes:

(I hope you don't mind my borrowing of your color coding, Tahir!)
 
Typedef Enum said:
Those (Kyle, you know who you are) who insisted that synthetic benchmarks were basically worthless because they do not have any bearing any "real" games should stand up and admit the error in their ways.

If Kyle actually admits he was wrong, I might actually visit his site again. However, I get the feeling that he won't admit to anything...and just blame everyone else (principally nVidia) for misleading the public. He will adopt ATI as his "new" best friend, and spin some story to make it look like he knew all along this was going to happen, and that he was instrumental in uncovering nVidia's plot.
 
Joe DeFuria said:
If Kyle actually admits he was wrong, I might actually visit his site again.
Ditto, but I doubt it'll happen.

But if nothing else, [H]'s reaction to this should be interesting to say the least... :LOL:

(I'm DEFINATELY enjoying this too much!)
 
Seeing the different sights benchmarking HL2 and their interpretation of things will be interesting indeed.

On a side note (kind of humerous, kind of what-if?), back when 3dfx was struggling, near the end, they moved heavily into marketing via slogans and name recognition--everyone remember the "3dfx" stickers that were being put on game boxes? It's like 3dfx recognized they were behind and knew it would be a while before that was corrected, so they put lots of publicity and energy into marketing.

Hmm. What is Nvidia doing these days: "The way it's meant to be played?"

It appears that with the 5800 they had a big misstep. Now it seems like their entire DX9 line-up may be a major screw-up--if that turns out to be the case, I think I'll look at marketing blitz's in a different way (first 3dfx, now NVidia, etc.).
 
digitalwanderer said:
But if nothing else, [H]'s reaction to this should be interesting to say the least... :LOL:

Yeah, in that "watching a train wreck" interesting kind of way. ;)

(I'm DEFINATELY enjoying this too much!)

(I'd be lying if I didn't admit to getting some distrubing enjoyment out of seeing bullshit PR coming back to haunt the bullshitter...."Use real games!" heh....)
 
Back
Top