• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

FarCry SM2.0b versus SM3.0 - a comprehensive 1.2 patch analysis

T

tranCendenZ

2[H]4U
Joined
Jun 6, 2004
Messages
3,844
In English, no less :)

http://www.xbitlabs.com/articles/video/display/farcry20b.html

Some snippets:
"FarCry Performance Revisited: ATI Strikes Back with Shader Model 2.0b"

"The latest 1.2 patch for the currently most-discussed game not only enabled Shader Model 3.0 on NVIDIA’s GeForce 6 hardware, but also brought Shader Model 2.0b for ATI’s latest RADEON visual processing units in order to improve performance and image quality. X-bit labs has run an exhaustive selection of FarCry benchmarks to find out whether long pixel shaders along with geometry instancing helped ATI to bring the performance up."

"While there are talks that NVIDIA encourages game developers to use 16-bit arithmetic precision instead of 32-bit precision (or 24-bit precision, like ATI) in order to improve performance, but possibly degrade image quality, we see no issues with NVIDIA’s quality under brief investigation."

"Even though ATI considerably improved performance of the RADEON X800-series graphics cards in FarCry’s “Pure Mode”, NVIDIA’s GeForce 6800 Ultra still can claim performance leadership here, as in quite a lot of cases speed delivered by the GeForce 6800 Ultra is a bit higher compared to the rivalling RADEON X800 XT.

The same situation is effective for a bit less expensive graphics options – the GeForce 6800 GT and the RADEON X800 PRO – 16 pixel pipelines definitely help the former to beat the latter in quite a lot of cases."
 
No, no, no, no, no!!!! The results are all wrong!!!!

They obviously didn't follow ATi's benchmarking guide. ;)
 
pxc said:
No, no, no, no, no!!!! The results are all wrong!!!!

They obviously didn't follow ATi's benchmarking guide. ;)
Click to expand...

Yeh none of the Driverheaven "enable trilinear optimizations for ATI but disable them for Nvidia" crap ;)
 
Why are you pro-NV people so excited about this review? The X800's get a nice boost from 2.0b, and are faster a lot of the time.The X800's will probably do better than the 6800's when the AA/AF results are in. The 6800's usually do better with no AA/AF, but who is not going to use it?

One thing is does show, is how really craptastic the 5950U is in Farcry.
 
fallguy said:
Why are you pro-NV people so excited about this review? The X800's get a nice boost from 2.0b, and are faster a lot of the time.The X800's will probably do better than the 6800's when the AA/AF results are in. The 6800's usually do better with no AA/AF, but who is not going to use it?

One thing is does show, is how really craptastic the 5950U is in Farcry.
Click to expand...

Correction, the 6800 only loses when AF is turned on and we all know why that is.
 
its funny, i remember when the x800 series first came out and it was considered the winner of this round. funny how things change.
 
Ardrid said:
Correction, the 6800 only loses when AF is turned on and we all know why that is.
Click to expand...

BTW, word is that Nvidia is working on a better Brilinear/Optimized Trilinear filter for even faster filtering performance, apparently will be released in 62.xx drivers.
 
Ardrid said:
Correction, the 6800 only loses when AF is turned on and we all know why that is.
Click to expand...

In my testing also with AA, but not as much as compared to AF.

Yes I know why ATi does better with AF. I also know it doesnt impact image quality.

Im all for NV doing the same type of thing, as long as image quality is the same.
 
tranCendenZ said:
BTW, word is that Nvidia is working on a better Brilinear/Optimized Trilinear filter for even faster filtering performance, apparently will be released in 62.xx drivers.
Click to expand...

As long as it remains an option that would be kinda cool...

I just started looking through this review and the GT is out-performing the X800XT in several places, even more impressive is when the XT loses badly in average fps...I realize this will change with AA and AF...but you have to admit it's pretty impressive what the GT can do with 350Mhz...

But then the X800s spank up on the 6800s in geometry instancing...which xbit says shows them to have better "future-proofing"...but they say that right after they say that future games will probably give the X800s more than they can chew since they are already hitting the max doing 3 lights in one pass (where nvidia is doing 4)...

I dunno...I don't think this article did much more than to show that the X800s still do have some surprises left in them with geometry instancing...and a brief confirmation that FP16 doesn't hurt image quality...

Nice to get this down in writing after a couple of days of rumors...
 
^eMpTy^ said:
As long as it remains an option that would be kinda cool...
Click to expand...

Yup NV will retain full trilinear, this new better brilinear filter will replace the current one.
 
tranCendenZ said:
Yup NV will retain full trilinear, this new better brilinear filter will replace the current one.
Click to expand...

Performance boosts are always welcome.
 
Certainly interesting. I may take a new look at the ATI line.
 
tranCendenZ said:
BTW, word is that Nvidia is working on a better Brilinear/Optimized Trilinear filter for even faster filtering performance, apparently will be released in 62.xx drivers.
Click to expand...

I'd rather see ATi just put a check box in...but if Nvidia pulls this off...ATi will lose a major one of its few advantages...
 
fallguy said:
In my testing also with AA, but not as much as compared to AF.

Yes I know why ATi does better with AF. I also know it doesnt impact image quality.

Im all for NV doing the same type of thing, as long as image quality is the same.
Click to expand...

I think you need to do some more testing then. The 6800's have higher AA performance then the X800's.

You can see how well the 6800's scale with AA enabled in this new review at Hexus.

http://www.hexus.net/content/reviews/review.php?dXJsX3Jldmlld19JRD04MDQ=

Then you can see how the 6800GT performs against the X800Pro with just 4xAA enabled here:

http://www.hothardware.com/viewarticle.cfm?articleid=550&cid=2

The X800's lose almost no performance from AF because of their excellent brilinear filtering tricks which they refuse to give the option to turn off. ATI has claimed FULL trilinear filtering all the way up till review sites like Tech Report and Tomshardware posted articles proving otherwise.

We also see almost no differences between NVIDIA’s and ATI’s image quality, except the “shadow issue” we discussed previously and some other minor things we are working on at this moment. While there are talks that NVIDIA encourages game developers to use 16-bit arithmetic precision instead of 32-bit precision (or 24-bit precision, like ATI) in order to improve performance, but possibly degrade image quality, we see no issues with NVIDIA’s quality under brief investigation.
Click to expand...

ATI and their anti FP16/FP32 campaign can just go fly a kite. :D

They also are still using the older 61.45 drivers in that review. They should of used the new WHQL 61.76's and just manually enabled SM 3.0. Far Cry has seen alot of performance boosts on the 6800's with the 61.72's, 61.76's, and especially the 62.01's.
 
tranCendenZ said:
BTW, word is that Nvidia is working on a better Brilinear/Optimized Trilinear filter for even faster filtering performance, apparently will be released in 62.xx drivers.
Click to expand...

BTW...where did you hear this?
 
^eMpTy^ said:
BTW...where did you hear this?
Click to expand...

The 62.01's already gave a huge Far Cry performance boost. Check out the 62.01 thread on nV News in the ForeceWare section. That was the first thing people said they got. A big Far Cry boost.

I wish they would of used that in the review instead of the 61.45's, especially since their using beta 4.8 cats.

Of course most reviewers dont keep up with things like that alot of the time.

X-bit Labs said:
For some reason, Shader Model 3.0 could not be enabled on NVIDIA’s latest official 61.76 drivers, which is why used a beta driver provided earlier by NVIDIA.
Click to expand...

See what i mean? lol
 
burningrave101 said:
The 62.01's already gave a huge Far Cry performance boost. Check out the 62.01 thread on nV News in the ForeceWare section. That was the first thing people said they got. A big Far Cry boost.

I wish they would of used that in the review instead of the 61.45's, especially since their using beta 4.8 cats.
Click to expand...

I got ip banned from the nvnews forums the first time I ever posted there...
 
Far Cry is the first game to use SM 3.0 and there are over a dozen others that have announced support for it. Far Cry doesn't begin to take advantage of what SM 3.0 can do. You can just look at the specs for SM 2.0b vs SM 3.0 and see there are vast differences.
 
I wonder how many people noticed that in there pics there, the ATI 2.0 path and 2.0b path when compared directly to Nvidia's 3.0 path, that there is a bunch of trees missing in the background, its only when the 2.0b path is engauged with the sprites distance ratio = 100 command that they show up.

Makes me wonder what Nvidias 2.0 path looks like. (going in game to find out)

Edit, 2.0 looks the same on ati and nvidia i guess, they musta set the sprites on both ati and nvidia cards for the review

Allthough I did notice that the waterline is completely missing from all the ATI shots.

yeah i know im nitpicking sue me :p
 
You must log in or register to reply here.
Back
Top