I think a number of you are getting a little carried away, however FYI, courtesy of jb:

http://www.chaoticdreams.org/ce/jb/ReflectAll.jpg
http://www.chaoticdreams.org/ce/jb/ReflectWorld.jpg

Well this is intresting, the more in depth performance review will be very nice.

Dont know about everyone else but Xbitlabs is my main source for computerhardware reveiws and it has been ever since i first visited this site.

Off-topic:
Hardocp in my view "may differ form your obviously" is Nvidia biased and give doom3 way to much credit. For me doom3 was the biggest gaming dissapointment in 2004 so far. Also I wonder how many games we will see with the Doom3 engine except Quake 4.

HL2 and Farcry Engine seems better to me, Stalker Engine seems ok too but since it isnt released yet im not sure.

/Nevyn

Reading this review I know what I BUY.
I was looking at the 5900 review (2years) ago, and watching sites saying best buy, best performance, beyond DX9.0, .....
And today see where this card is.

A 2 year old Ati card (9800) is as good as a brand new nvidia card (6600) so must say that nvidia did catch up, but I am not taking any risks like my friends did with their 5200, 5600, 5900 cards.

Regarding to FP32, why we don’t see any image quality difference, yet?
Is it because DX9 is FP24 only compliant?
I thought that when DX9.0 games start appearing we was going too see some differences.

Sped: "Hardocp used an overclocked nvidia card, and didnt enable the high quality water shaders"

How do you know this? Quoting the Hardocp article:
"All the game settings were set at their highest values with “Reflect World” selected."

I'm not taking sides here, I just think the considerable gap between Nv and ATI cards in the article is a little strange. Most benchmarks see ATI on top but not by THAT much.

This is ridiculous. I am running everything on the highest level at 1280X1024 with a 6800gt and guarantee you that my frames do not drop to 43 fps ever! Not even once.

3. I'd really love to see a couple benchmarks with older/slower CPUs.

5. And it would be nice if we could see how last generation cards handle the game... e.g. (5200, 5600, 9200, 9600) Try it on medium settings for an example, you don't need 'pure' mode (highest settings) to make it playable.

Agreed with post 3 and 5. Most Hardware sites are so caught up with fast CPUs/video cards, they forget to realise that the majority of gamers don't have fast machines or video cards.

How about a future article that involves various lower end CPUs and video cards to give us an idea how well (acceptable) does HL2 run on the "average Jane/Joe" system? That is a far more helpful article.

Here are the other links I found.

http://www.pcper.com/article.php?aid=93&type=expert
My favourite review. Basically a tie, expect in the water levels where ATI won.

http://www.anandtech.com/video/showdoc.aspx?i=2278
Another detailed one. Page 10 has the direct comparison showing things are pretty much a tie, other than the super expensive cards.

http://www.thetechlounge.com/article.php?directory=hl2_per ormance
Basically another tie. This was interesting as it showed the 256meg cards killing the 128meg cards as the resolution went up. They used Catalyst 4.11 and ForceWare 66.93 drivers.

http://www.firingsquad.com/hardware/half-life_2_performanc /
At lower resolutions on levels without water nvidia wins (barely), while at higher resolutions ATI wins. Of course not many people will play at 800x600 so nvidia's win is moot. More proof in water ATI has a clear advantage.

http://www.hardocp.com/article.html?art=Njky
High end performance is similar, but ATI is ahead.

http://www.amdzone.com/modules.php?op=modload&name= ections&file=index&req=viewarticle&artid=79&p ge=1
Kind of a lousy article. The X600 and 6600 perform similar...

And of course in this article ATI dominates the canel level, but in others the lead is smaller, often around 10%. 10% is close enough that I'd call it a virtual tie, even though it is a clear win for ATI since it isn't that big a lead. Other than the water levels that is...

Out of everything I've read the cards are close enough you basically can't tell a difference, except at really high resolutions. I don't think I'd play at those resolutions myself as the even though ATI has an advantage it isn't enough to actually make the frame rate high enough. Once the resolution is lowered to get a good frame rate then it basically a tie (other than water levels).

So they're a tie. ATI wins at water levels. nVidia wins at Doom3. You can't go wrong with either brand. IMO there is no winner this generation.

There is card bias - whether you choose to agree to it or not. The reviewers have their heads in ATi clouds and it's blatantly obvious in the review. I believe the stats - It's the jabbering about how the "venerable" 9800XT "won't lose to" the 6600GT that's disgusting. Despite the fact that the 6600GT performed comparably well enough to the 9800XT throughout the test overall (who the hell plays these games in 1600x...? at 24 fps?)
It's the logical argument of:
http://www.newegg.com/app/ViewProductDesc.asp?description= 4-119-130&depa=1
vs
http://www.newegg.com/app/ViewProductDesc.asp?description= 4-150-078&depa=1
That make the decision so much more vague than the reviewers cared to give attention to. I'll be frank, that are the cheapest model available on newegg; the next "venerable" (piece of crap) is $50 the link given. A 100 dollar difference, up to $150 (75% more from a 6600GT) if you plan on recieving the frames shown in this test. (Do NOT make claims the 9800xt used was the ELSA in the link... please, for the sake of the impressionable reader)
Get Real: A sensible pc-user doesn't buy a

I think these benchmarks are just another example of hardware reviews taking in-game settings for granted. I don't agree with maxing out settings and then benchmarking on that basis as if the results are the letter of the law. Most gamers adjust in-game settings according to their card because some do better than others at certain things. For example I get a huge fps boost in far cry with my 9800 pro by lowering the shadow and lighting details whereas maxing others out at very high has virtually no impact.

I appreciate the time it would take to experiment with the settings but for me this shouldn't be overlooked when you're supposedly reviewing performace full stop rather than "max quality" performance. It looks like the 6800 is struggling with "reflect all" enabled. Why not test out "reflect world" as well, at least on a limited scale, and consider why ati does it better, and if the iq loss is really that big a hit for the sake of performance? I'm sorry, but without doing this a benchmark is and always will be biased.

Don't just conclude with "nvidia blows because it can't do reflect all," but come to an intelligent conclusion as to why a card that overall performs more or less equally to the x800 is hit so hard. There are no enemies here.

OMFG....
This is probably the most bias review I have ever read. (Not just HL2 reviews) I myself own both 6800 Ultra and X800 XT (not PE) and I can tell you that while it's true that X800XT perofrms better the Ultra, the diff is not all that great. in d1_canals_09 running 1280x1024 4AA, 4AF. my XT did 70fps avg and the Ultra gave me 66fps avg. They're both great cards. Oh, not that it matters, my CPU is A64 3400+ on a DFI nForce 3 with 1G of Corsair 2-2-2-5 Ram.

I am running a GeForce 2, wonder if it will work on HL2...

just saying a good review and well done for taking flack i took note of the overclock with D3 and the punters around were wailing NV play it 1million times better etc dont like it when its other way round mind

Great update!

a quand la demo jouable de hl2

Don't know what kind of FPS I was getting, but I know this: I finished HL2 a couple of weeks ago. Had absolutely no slowdowns except loading times were quite long between segments. Having bought Doom 3 a while earlier, and not able to play, due to choppiness, needless to say, I was VERY happy.
Memory: 512
CPU: Pentium 4 - 2.0
Graphics Card: Asus v8200 (GeForce 3 - 64meg)
Not exactly cutting edge, but adequate for this game anyway.
Settings for the game were:
Model Detail: Medium
Texture Detail: Medium
Water Detail: Simple Reflection
Shadow Detail: High
Filtering Mode: Triliniar
Shader Detail: High
Antialiasing Mode: Disabled
Vertical Sync: Disabled
Direct X: 9.0c
Resolution: 800 x 600

hi

kikimora

I'am from Romania!!Can i ask you something??: In your country are more perfomance 4computer componence???sombedy have INTEL P5????if you wont do anser me>djcoco@personal.ro

I wish they put in CS:Source too. That games requires much more if you are going to play with bots that are hosted on your computer.