-
Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.
The latest Far Cry 1.2 Benchmarks: (SM 2.0b ATI support)
- Thread starter Izzy
- Start date
Link to the article: http://www.xbitlabs.com/articles/video/display/farcry20b.html
Oh and anyone stupid enough to buy a 9800XT now deserves whats coming to them
Oh and anyone stupid enough to buy a 9800XT now deserves whats coming to them
seismologist
Banned
6800 series looks like it's owning everybody. Plus doesn't it have more features too?
Late Stage Marxist
Member
Everyone's always saying to get a 9800 pro but I don't even see it on the benchmarks. What's with?
Well this just confirms that the 6800 GT is the best valued card of this current generation. You'd be a fool to get the ultra considering GT can overclock to ultra speed plus uses only one molex, I think the only reason to wait is maybe for the PCI Express version and being able to SLI ala 3dfx style. The only regret of buying the card is that I didn't jump sooner to take advantage of the BB or Buy.com preorder
Buggy Loop
Gold Member
The benchmarks shows it, last gen it was ATI's 9800 over Nvidia's FX. This gen its 6800 > x800.
Its called competition, you know, something we hadnt seen since voodoo 2/3 vs tnt2
Its called competition, you know, something we hadnt seen since voodoo 2/3 vs tnt2
Outdoor Miner
Member
Um, wow. Talk about selective benchies. Jesus, did anyone bother to check out other benchies from this same report? I guess it's hard when a link to site isn't even provided.
Look i can too!
At high resolutions, w/ lots o' stuff going on, and w/ AA and AF enabled (a.k.a How you TEST and STRESS a card), the X800 series is still king.
Oh, look more!!!
There's no doubt Nvidia got it's stuff together this generation. And it definetly is the cardmaker to go w/ if Doom3 is your main reason for purchasing a new card. But it's getting tiring combating all the Nvidiot spin, bullshit, misleading benchies, etc. Jesus, that part of Nvidia is FULL BLOWN this gen.
Edit- WTF is the point of testing these "next-generation" cards if you're going to test at 1024x768 w/ no AA or AF? That is what Xbitlabs "pure speed" setting is.
Look i can too!
At high resolutions, w/ lots o' stuff going on, and w/ AA and AF enabled (a.k.a How you TEST and STRESS a card), the X800 series is still king.
Oh, look more!!!
There's no doubt Nvidia got it's stuff together this generation. And it definetly is the cardmaker to go w/ if Doom3 is your main reason for purchasing a new card. But it's getting tiring combating all the Nvidiot spin, bullshit, misleading benchies, etc. Jesus, that part of Nvidia is FULL BLOWN this gen.
Edit- WTF is the point of testing these "next-generation" cards if you're going to test at 1024x768 w/ no AA or AF? That is what Xbitlabs "pure speed" setting is.
Outdoor Miner
Member
I will say though, ATI's got nothing on Nvidia in the $399 market, and the GT seems to be a much better value then it's Ultra counterpart. There are rumors though that ATI's cooking up a 16-pipe variant of the X800 XT that'll be underclocked, meant to replace the PRO as direct competitor to the GT.
Time will tell.
DJ Demon J
Banned
So has anyone got their hands on the X800 XT yet? The only version released that I know of is the Xtasy brand (sold out in many places).
DJ Demon J
Banned
Actually, GameVE.com is taking orders and says it ships on August 6th:
http://www.gameve.com/gve/Store/ProductDetails.aspx?sku=VC-VT-007
Anyway, I'm dropping my 9800 Pro 256. Anyone interested in buying it? $240 (original box, software, manual and card, all 1 year old and in perfect condition). Shipping included in cost.
http://www.gameve.com/gve/Store/ProductDetails.aspx?sku=VC-VT-007
Anyway, I'm dropping my 9800 Pro 256. Anyone interested in buying it? $240 (original box, software, manual and card, all 1 year old and in perfect condition). Shipping included in cost.
Wow. Thank you so much for actually looking through the article and presenting that, Sal Paradise. I was really started to think I made the wrong decision ordering an X800XT, as I've been seeing a lot of spin for Nvidia lately after the Doom 3 article at HardOCP, despite the X800XT killing the 6800U earlier on in pretty much every other game benchmarked on the very same site. Those benchmarks with the full FSAA and AF make me very, very happy. I was wondering why those weren't posted in the first place, since those are what really matter, as it's what is actually showing how much the video card can be pushed, as opposed to those "pure speed" benches which are all but completely reliant on the CPU. I mean that's what they're made for in the first place.
That's all I wanted to hear.
That's all I wanted to hear.
Shompola said:
So I assume you are never going to use FSAA/AF ever? The X800Pro/XT has been doing better performance wise than 6800GT/Ultra in a lot of games(especially with AA/AF on) and suddenly it became a really bad buy when it doesn't do as well in FarCry with no AA/AF enabled or 5-15 (depending on AA/AF)fps behind 6800 at already good framerate in Doom 3?
Looks like Nvidia won this round, which is great for us. We don't want one company to win all the time or they're won't be competition.
Last generation, Ati beat Nvidia silly. While Nvidia isn't as far ahead as ati was last time, at least they are ahead.
I guess I shouldn't be too suprised. The X800s are just 9800s tweaked while 6800s are brand spanking new.
Last generation, Ati beat Nvidia silly. While Nvidia isn't as far ahead as ati was last time, at least they are ahead.
I guess I shouldn't be too suprised. The X800s are just 9800s tweaked while 6800s are brand spanking new.
the only problem I currently have with ATI is that I never know if the x800's are being benched with optimizations off. I know that while playing I wouldn't notice it and in that respect ATI wins. I just want to know which card will perform the best with the most stuff and no optimizations. Anyway, as said above with all the bells and whistles ATI wins again, but Nvidia really did their shit together and the GT is a great buy.
I wish they would list the average retail price besides each card...it makes the comparison much easier. I googled each card name in there, and saw 5 different prices for each
It's all very interesting though...and I plan to wait until HL2 is out before making a decision.
Thanks for sharing the info guys (and for showing both sides of the issue)!
It's all very interesting though...and I plan to wait until HL2 is out before making a decision.
Thanks for sharing the info guys (and for showing both sides of the issue)!
seismologist
Banned
Nvidia is still winning in my eyes. Also The 6800 series is more future proof with SM3.0 support.
mashoutposse
Ante Up
The X800 cards utilize AF and texture filtering optimizations that cannot be disabled.
seismologist
Banned
So as long as I keep AA&AF off my 6800GT will be faster than a x800xt?
Are AA&AF really needed? The HardOCP Doom 3 benchmarks seem to indicate that it's not.
Are AA&AF really needed? The HardOCP Doom 3 benchmarks seem to indicate that it's not.
Teknopathetic
Member
AF's cool, but past 4x AF it's really just a waste. Also, AA at high resolutions is just as useless.
Cmdr. Solrak
Member
Sal Paradise Jr said:Um, wow. Talk about selective benchies. Jesus, did anyone bother to check out other benchies from this same report? I guess it's hard when a link to site isn't even provided.
Look i can too!
At high resolutions, w/ lots o' stuff going on, and w/ AA and AF enabled (a.k.a How you TEST and STRESS a card), the X800 series is still king.
Oh, look more!!!
There's no doubt Nvidia got it's stuff together this generation. And it definetly is the cardmaker to go w/ if Doom3 is your main reason for purchasing a new card. But it's getting tiring combating all the Nvidiot spin, bullshit, misleading benchies, etc. Jesus, that part of Nvidia is FULL BLOWN this gen.
Edit- WTF is the point of testing these "next-generation" cards if you're going to test at 1024x768 w/ no AA or AF? That is what Xbitlabs "pure speed" setting is.
Doom 3 and FarCry are different story all together. Doom 3 is designed as so it doesn't need AA so badly. FarCry do benifit from some AA. But, for me, AA is just a waste of GPU, and I prefer cranking the resolution up and putting some AF (8 is a great setting).seismologist said:
To think of it, its the default setting in Doom3 for HighQ. (0AA 8AF).
Outdoor Miner
Member
seismologist said:
Um, what about Nvidia's cards are more future proof? SM 3.0? The ONLY worthwhile feature that is going to be used extensively from that set is geometry instancing. It's a neat little trick that can reduce workloads for the renderer by a high percentage on complex scenes.
ATI cards have this feature on the R3xx and up. What part of better geometry processing and better pixel shader processing don't people understand???
It states right there in this report (wouldn't be the first one) the ATI cards are going to be better suited for future apps. Unless of course a game is hand-tailored (literally) for a Nvidia card.
So now AA/AF aren't really important? Gotcha. *checks off list*
seismologist
Banned
Sal Paradise Jr said:
Well it's a trade off. According to these benches, even on the ATI card, enabling AA/AF cuts your framerate in half.
Hard to say if it's worth it. Especially in newer games where your struggling to stay above 30fps.
Outdoor Miner
Member
Well, i don't think anyone can argue that there's always going to be a trade off when talking about picture quality/IQ versus speed. It's always trying to find the perfect balance. And those needs are going to be different for every inidividual based on their preference.
Some people prefer to play games at 1024 with 4xaa/8xaf. Others prefer gaming at 1600x1200 with maybe just a touch of AF.
The point is you can't accurately measure a cards performance and future potential going the way Xbitlabs goes about in their "Pure Speed" mode. It's something i've always had an issue with. The same goes for their "DX9" benchmarks that is really just the leaked HL2 code.
It's hard to tell which sites are being objective anymore. All i know is currently I trust no one other than Beyond3d. They have the most knowledgeable forum concerning graphics on the web. Great site, and they are always clear on how they go about testing.
Some people prefer to play games at 1024 with 4xaa/8xaf. Others prefer gaming at 1600x1200 with maybe just a touch of AF.
The point is you can't accurately measure a cards performance and future potential going the way Xbitlabs goes about in their "Pure Speed" mode. It's something i've always had an issue with. The same goes for their "DX9"
benchmarks that is really just the leaked HL2 code.It's hard to tell which sites are being objective anymore. All i know is currently I trust no one other than Beyond3d. They have the most knowledgeable forum concerning graphics on the web. Great site, and they are always clear on how they go about testing.
Phoenix
Member
I really wish they would start doing their benchmarks based on price point. I could give a rats ass about how fast a $500-$700 card performs because there is a 0% chance that I am dropping that kind of money on a video card - and I have the money to do it, I simply refuse to spend that kind of money on something with that type of depreciation. That has got to be the worst return on investment in the universe!
Hey Sal,
Thanks for divulging the secret of the "Pure speed" setting. When these benches came up before I asked what the Pure speed setting was and nobody responded. At that point I knew there was something fishy going on. For the $400 price point the 6800GT seems like the card to get, and for the top-end the x800XT is the way to go. Sounds good to me.
Thanks for divulging the secret of the "Pure speed" setting. When these benches came up before I asked what the Pure speed setting was and nobody responded. At that point I knew there was something fishy going on. For the $400 price point the 6800GT seems like the card to get, and for the top-end the x800XT is the way to go. Sounds good to me.
Sal Paradise Jr said:
That's definitely not true. Let me help you out with a description of some of the more worthwile SM 3.0 features:
-One of the more major upgrades in Shader Model 3.0 is the addition of Vertex Texture Lookups. What this allows is features like Displacement Mapping. If there is going to be any major difference in image quality comparing Shader Model 3.0 to 2.0 it is going to be with the use of Displacement Mapping. Bump mapping which is currently used now to give the appearance of height in textures is just an illusion. There is no physical difference in the texture, meaning if you look at the texture from the side or dead on you will see that it is still flat, only from far away does bump mapping work. Even then it isnt the best option since the texture is physically still flat light and shadows do not reflect correctly. The answer is Displacement Mapping which physically adds surface detail by manipulating the height of the texture. Displacement Mapping can even go as far as to create the model itself. Displacement Mapping may be a huge boon to adding realism in games. If developers pick up on this technology and we see it implemented in games, this right here could be the deciding feature that shows the most difference between a game rendered in Shader Model 3.0 and a game rendered in Shader Model 2.0.
-Dynamic Branching is also a new feature that Shader Model 3.0 has that Shader Model 2.0 does not. Dynamic Branching gives the ability for programmers to control the actual flow of the program, starting and stopping code where they see fit instead of a straight execution from the beginning line of the shader program to the end. What this would allow is faster shader performance.
Hope that helps, Sal.