The ultimate question.

[rancor]

WTF I showed you where XT beats a GT. Show me where the 6800GT beats a XT? You have no evidence no proof nothing. You just keep saying your 61.47 or whatever. I don't care it's probably a game specific optimization like quack. That is why ATI will be doing the same thing hence CCC.

But yeah what i've been saying is that this Vivo goes beyon XT speeds easy more likely XTPE speeds so ATI wins. BOO HOO LOL

gt at ultra speeds...... take alook at the main topic again. Thats what I was talking about. who in the right mind would put a gt against and xt pe or even an xt for that matter, its 100 bucks less.

they all don't go to xt pe speeds, only about 50% of them you have to be very careful on the type of vivo you get to get to the xt pe speeds.

Still no numbers?

Plus so far there has been no indication of game specific shader replacement on nV's part, like they did for the fx line. So you really can't bring that up without proof.

 

[Marvelous]

Obviously your brain can't cope with the true facts. Lof of people are getting vivo's and flashed them getting XTPE speeds. I say a Ultra might hang with XT but not XTPE. You in the other hand have a GT and you can't cope with that your card is inferior to the X800Vivo flashed to XT. Oh poor baby. What are you going to do now?

It's obvious almost the whole vid card forum here has eaten up Nvidia's PR hype and gotten a GT. That's how it is every year. Sooner or later you will realize that it's okay to lose once in a while you can'. There's hope for you yet. And these Vivo's are swarming and selling like hot cakes. When they do this board will be filled with x800 owners swarming on you.

 

[rancor]

Obviously your brain can't cope with the true facts. Lof of people are getting vivo's and flashed them getting XTPE speeds. I say a Ultra might hang with XT but not XTPE. You in the other hand have a GT and you can't cope with that your card is inferior to the X800Vivo flashed to XT. Oh poor baby. What are you going to do now?

It's obvious almost the whole vid card forum here has eaten up Nvidia's PR hype and gotten a GT. That's how it is every year. Sooner or later you will realize that it's okay to lose once in a while you can'. There's hope for you yet. And these Vivo's are swarming and selling like hot cakes. When they do this board will be filled with x800 owners swarming on you.

Dude man you can't even read, you never answed the poll, the poll was a gt and ultra speeds. You were talking about the gt at regular clocks this whole time lol, your insite is so out in left field you could set up camp.

What true facts come on show them to me.

 

[Marvelous]

Doesn't matter XT hurts a Ultra. I provided proof. What prood did you have except your big mouth.

Your pretty lame you keep telling me to provide with proof but you can't even come up with 1 benchmark where a GT to ultra can beat a XT.

The guy who posted this poll is misinformed most people are get more than 500mhz core easy.

 

[rancor]

Doesn't matter XT hurts a Ultra. I provided proof. What prood did you have except your big mouth.

proof of what a the first sm 3.0 drivers released by nV, which btw was thier second driver release also.

Man you are sooooo blind, first you talk about about something the poll wasn't even about now you're trying to back track, keep tracking.

Where are the numbers Marv? about the fillrates, shader performance, etc?

 

[Marvelous]

you can't even come up with 1 benchmark where a GT to ultra can beat a XT.

Don't you feel stupid? I mean you have no proof nothing. your magic drivers that supposedly GT getting close to a XT is crap.

 

[rancor]

you can't even come up with 1 benchmark where a GT to ultra can beat a XT.

Don't you feel stupid? I mean you have no proof nothing. your magic drivers that supposedly GT getting close to a XT is crap.

I don't feel stupid, this poll says it all. Since you are the only one that didn't read the poll properly to begin with you were giving out the wrong information all this time. A gt at ultra speeds is an Ultra, you don't need to look any further. Still no numbers?

 

[Marvelous]

Denial first human response when they feel stupid.

Doesn't matter XT hurts a Ultra

 

[rancor]

Doesn't matter XT hurts a Ultra

its a waste of time to debate with a person that doesn't read properly

most benchmarks with 61.77 the ultra is 10 fps slower with max aa and af on, the newer aa and af opts where added in the 62 driver sets and up, which give a general boost of 10-20% overall in all games with aa and af on. Not to mention a small boost in the base frame rates with no aa and af aswell.

Also the 6800 ultra is almost always in the lead with no aa and af on with the 61.77s

and this is my last comment to you........

 

[Marvelous]

You haven't given me a single link? Why cause it's non existent?

 

[rancor]

You haven't given me a single link? Why cause it's non existent?

Alright this is the last respose you will get from me, here ya go, quite hard to find a link that used the newest drivers.

http://www.digit-life.com/articles2/digest3d/index.html#config


Interesting, can you hold your breath?

Hmm interesting to find the gt is matching the xt pe at times (they used an xt pe not an xt as you see the clock speeds for thier xt is 525/1150)

This is with the 65.76 dirvers from nV btw, and digit life is very good with thier reviews, one of the best, still like [H]'s reviews better since they go from playable testing.

And the 66 driver set gives more of a performance boost as well.

 

[tornadotsunamilife]

called the Reality engine (actaully its already being used to make a game the engine is done)

is that the one from Artificial Studios?

btw i recommend the 6800gt. but you don't have to take my word for it the poll, review sites and forums speak for themselves

 

[Marvelous]

Funny how that review they do not show real resolutions these cards should be playing. They should benchmark 1280x1024 4x AA and 16x aniso or 1600x1200 with 4xaa and aniso for that matter. See how XT closes the gap when AA and aniso are used at 1024x768. Well this shows that XT is better high resolutions with AA and aniso which proves my point.

Nvidia cards always had faster fps in lower resolutions. This has always been. And you see in Unreal 2 engine 6800ultra gets smacked by even a radeon x800pro in 1024x768 with AA and aniso so what does this mean. Nvidia's cards are crippled? Obviously if it gets smacked by a x800.

Also there's a level where Nvidia cards cannot render good at all. The level is Pier in Farcry

http://techreport.com/reviews/2004q3/radeon-x700xt/index.x?pg=11

Techreport is using your precious forceware 65.76 drivers. Digit life is one site and they are not even well known. Only thing they know how to benchmark flight sims.

 

[CleanSlate]

Funny how that review they do not show real resolutions these cards should be playing. They should benchmark 1280x1024 4x AA and 16x aniso or 1600x1200 with 4xaa and aniso for that matter. See how XT closes the gap when AA and aniso are used at 1024x768. Well this shows that XT is better high resolutions with AA and aniso which proves my point.

Nvidia cards always had faster fps in lower resolutions. This has always been. And you see in Unreal 2 engine 6800ultra gets smacked by even a radeon x800pro in 1024x768 with AA and aniso so what does this mean. Nvidia's cards are crippled? Obviously if it gets smacked by a x800.

Also there's a level where Nvidia cards cannot render good at all. The level is Pier in Farcry

http://techreport.com/reviews/2004q3/radeon-x700xt/index.x?pg=11

Techreport is using your precious forceware 65.76 drivers. Digit life is one site and they are not even well known. Only thing they know how to benchmark flight sims.

Dude you're so wrong. actually no I just wanted to comment on your sig... clean it up cause I can't tell wth it's trying to say..

Btw yes the higher the res and higher the aa/af... the more then x800 shines. Been that way with ATi cards all along. Read an [H] review once in a while if you don't know this ppls.

~Adam

 

[memphis_1220]

This is fun I want more!

 

[Marvelous]

I'm totally not saying a GT is a bad card. It's definitely up there with most bang for the buck in the high price range. But it just can't handle a XT in DirectX, AA or aniso. Most games are directx. When prices are about the same I would go for the XT no doubt. Any ding dong who buy these cards and turn down the resolution is lame.

There's only 2 games in that digitlife review that is Opengl (doom 3 and COD) and who cares about COD. This game is based on a quack engine which Carmack made 5 years ago.

I mean if we had half and half I would understand justifying the GT but there's less than 10% of OpenGL games created today. And most of them are based on quack engine which x800xt does fine no problem and it also does fine in Doom engine as well. If this was like 6 years ago when most of these games companies were using quake engines but it's not the case any more. Eveyone has a engine now. Carmack is all by himself and so is Nvidia with their openGL. If you can't see this your blind.

 

[tornadotsunamilife]

yeah CoD is crap and is not the best ww2 shooter to date

and if you haven't tried the expansion pack demo yet you would realise how it can bring systems to their knees still

 

[rancor]

Funny how that review they do not show real resolutions these cards should be playing. They should benchmark 1280x1024 4x AA and 16x aniso or 1600x1200 with 4xaa and aniso for that matter. See how XT closes the gap when AA and aniso are used at 1024x768. Well this shows that XT is better high resolutions with AA and aniso which proves my point.

Nvidia cards always had faster fps in lower resolutions. This has always been. And you see in Unreal 2 engine 6800ultra gets smacked by even a radeon x800pro in 1024x768 with AA and aniso so what does this mean. Nvidia's cards are crippled? Obviously if it gets smacked by a x800.

Also there's a level where Nvidia cards cannot render good at all. The level is Pier in Farcry

http://techreport.com/reviews/2004q3/radeon-x700xt/index.x?pg=11

Techreport is using your precious forceware 65.76 drivers. Digit life is one site and they are not even well known. Only thing they know how to benchmark flight sims.

Again what cards were they testing?


GeForce 5900 XT 128MB AGP with ForceWare 61.77 drivers
GeForce 6600 GT 128MB PCI-E with ForceWare 65.76 drivers
GeForce 6800 128MB with ForceWare 65.76 drivers
Radeon X600 XT 128MB PCI-E with CATALYST 4.10 beta drivers
Radeon 9800 XT 256MB AGP with CATALYST 4.10 beta drivers
Radeon X800 Pro 256MB AGP with CATALYST 4.10 beta drivers
Radeon X700 XT 128MB PCI-E with CATALYST 4.10 beta drivers

Were is 6800 gt and ultra?

And the 6600 gt, and the 6800 are almost always the fastest cards in those benchmarks, I don't even see the x800pro in thier dx benchmarks.

 

[rancor]

=tornadotsunamilife]is that the one from Artificial Studios?/QUOTE]


Yes it is shaping up to be a very nice engine.


Btw Marv Unreal 2 was a Dx8 engine with almost no shaders, so the burden is put on the vertex level, which ATi is stronger in.

 

[Tigerblade]

but you don't have to take my word for it the poll, review sites and forums speak for themselves

I agree with this statement....

Rancor....for someone that 'seems' to be very technically minded this sentance : 'I don't feel stupid, this poll says it all' is a crock of shit. Since when does a poll give a definative answer?? Poll's are not hard facts.Just because x% of people say the GT is better, doesnt mean they are right.

At the end of the day this arguement is pointless...u put identical set up's one with an ultra one with an XT PE and NO ONE here will be able to tell the difference, not in Doom 3 or Half Life 2 either....

 

[rancor]

I agree with this statement....

Rancor....for someone that 'seems' to be very technically minded this sentance : 'I don't feel stupid, this poll says it all' is a crock of shit. Since when does a poll give a definative answer?? Poll's are not hard facts.Just because x% of people say the GT is better, doesnt mean they are right.

At the end of the day this arguement is pointless...u put identical set up's one with an ultra one with an XT PE and NO ONE here will be able to tell the difference, not in Doom 3 or Half Life 2 either....

LOL, coming for who, yeah forgot about Far Cry and Pain Killer? whos on what sides? can't see the middle ground can ya?

As I said about a page back, not many benchmarkers have used the latests driver from nV since they weren't officially released by them. The poll does have alot of do with this, because alot of people here know about the driver improvements on nV's side, and how they effected the aa and af perfromance and the base performance, plus the ease of getting the gt to ultra speeds. And the risks of the vivo to xt mod and the clock rate issues.

We aren't talking about frame rate differences, that why I stated the ultra is just as good and equally competitive with the xt pe. Guess what Marv was the one that said the xt pe washes over every other card! NOT ME, I didn't take a side on that.

And according to digit life in Ogl the ultra crushes the XT pe, and in Dx its just a tad faster, which I still don't agree with, I still say they are equal, because its not that much of a difference.

 

[Tigerblade]

LOL, coming for who, yeah forgot about Far Cry and Pain Killer? whos on what sides? can't see the middle ground can ya?

I'm not arguing with u your the one that said the poll answers the question

I'm just getting sick of reading the same crap day after day about who's card is 3fps faster then the other.....at the end of the day, noone will notice the differences.....

The reason we bought such kick ass cards is to play the latest games at full everything....I'm sure we can all do that just fine, so why don't we just enjoy the games we can now play at full everything, instead of arguing who's e-penis is longer than the others?

One final word...can people stop starting these piss poor polls!! There is one here every god damn week on the same subject.....just use the feckin search....

Peace out

 

[rancor]

I'm not arguing with u your the one that said the poll answers the question

I'm just getting sick of reading the same crap day after day about who's card is 3fps faster then the other.....at the end of the day, noone will notice the differences.....

The reason we bought such kick ass cards is to play the latest games at full everything....I'm sure we can all do that just fine, so why don't we just enjoy the games we can now play at full everything, instead of arguing who's e-penis is longer than the others?

One final word...can people stop starting these piss poor polls!! There is one here every god damn week on the same subject.....just use the feckin search....

Peace out

LOL same here, both cards are so damn close, can't really say which one is faster a 5 fps difference at those speeds don't make a difference. Sorry about that, was all worked up for another debate lol

 

[jiminnyjellickerz]

OMG I MUST HAVE THE XTPE SO I HAVE 5000000009FPS INSTEAD OF 500000001FPS SO IT RUNS SMOOTHAR YEAH! MY CARD PWNZ YOURS ETC. I WILL JUSTIFY MY BUY INSTEAD OF ENJOYING MY GAMES!11111


*sad

sorry I couldn't resist

nobody will be able to tell the bloody difference as the fps are so stonkingly good with any of those cards for godsake

 

[creedAMD]

OMG I MUST HAVE THE XTPE SO I HAVE 5000000009FPS INSTEAD OF 500000001FPS SO IT RUNS SMOOTHAR YEAH! MY CARD PWNZ YOURS ETC. I WILL JUSTIFY MY BUY INSTEAD OF ENJOYING MY GAMES!11111


*sad

sorry I couldn't resist

nobody will be able to tell the bloody difference as the fps are so stonkingly good with any of those cards for godsake

that makes too much sense, no thread hijacking please

 

[Marvelous]

yeah CoD is crap and is not the best ww2 shooter to date

and if you haven't tried the expansion pack demo yet you would realise how it can bring systems to their knees still

Yeah like 3 years ago and I'm getting so tired of COD. Oh my god expansion pack what are we going to do now? There are massive amounts games based on DX? Who still makes OpenGL games? Can you name anybody? I mean nobody makes it except carmack. The trend is OPENGL will die out. DirectX will always get support.

I've already told you there are only couple of opengl games really worth playing and there is bag full of games based on DX that's worth playing.

Again what cards were they testing?


GeForce 5900 XT 128MB AGP with ForceWare 61.77 drivers
GeForce 6600 GT 128MB PCI-E with ForceWare 65.76 drivers
GeForce 6800 128MB with ForceWare 65.76 drivers
Radeon X600 XT 128MB PCI-E with CATALYST 4.10 beta drivers
Radeon 9800 XT 256MB AGP with CATALYST 4.10 beta drivers
Radeon X800 Pro 256MB AGP with CATALYST 4.10 beta drivers
Radeon X700 XT 128MB PCI-E with CATALYST 4.10 beta drivers

Were is 6800 gt and ultra?

And the 6600 gt, and the 6800 are almost always the fastest cards in those benchmarks, I don't even see the x800pro in thier dx benchmarks.

Yeah the whole point is even a POS x700 or a 9800XT gets really close to a 6800. I would really like some reviewer to benchmark this level with high end cards. But the trend will be same unless 6800GT and ultra are entirely different architecture. But it's not.

and how they effected the aa and af perfromance and the base performance, plus the ease of getting the gt to ultra speeds. And the risks of the vivo to xt mod and the clock rate issues.

LOL. I freakin gave you a link where all X800pro owners getting to mod an XT. There's no risk period. There's a rish it won't do XT-PE. But there's also no guarantee the GT will get to Ultra speeds.

 

[rancor]

Yeah like 3 years ago and I'm getting so tired of COD. Oh my god expansion pack what are we going to do now? There are massive amounts games based on DX? Who still makes OpenGL games? Can you name anybody? I mean nobody makes it except carmack. The trend is OPENGL will die out. DirectX will always get support.

I've already told you there are only couple of opengl games really worth playing and there is bag full of games based on DX that's worth playing.




Yeah the whole point is even a POS x700 or a 9800XT gets really close to a 6800. I would really like some reviewer to benchmark this level with high end cards. But the trend will be same unless 6800GT and ultra are entirely different architecture. But it's not.

LOL they are entirely different the 6600 is an off shoot of the gf 6 series with less transistors, about 70 million less, and the 6600 beats the x700xt in everything even with aa and af on

and why do you so a level in Far Cry thats geometically intensive and not really shader intensive, check the next page where there is a level that is more shader intensive,

http://techreport.com/reviews/2004q3/radeon-x700xt/index.x?pg=12

Look at the entire picture, as I stated before depending on geomteric complexity (ATi is better with vertex calculations, nV with pixel shader calculations)

BTW you're the only one calling the x700 a POS, since I've already mentioned the x700 should be better matched with newer drivers from ATi!

Why even bother talking about OGL, look at only the dx games, damn freakin close, the 6600 is almost always infront anyways.

 

[Marvelous]

LOL they are entirely different the 6600 is an off shoot of the gf 6 series with less transistors, about 70 million less, and the 6600 beats the x700xt in everything even with aa and af on

and why do you so a level in Far Cry thats geometically intensive and not really shader intensive, check the next page where there is a level that is more shader intensive,

http://techreport.com/reviews/2004q3/radeon-x700xt/index.x?pg=12

Look at the entire picture, as I stated before depending on geomteric complexity (ATi is better with vertex calculations, nV with pixel shader calculations)

BTW you're the only one calling the x700 a POS, since I've already mentioned the x700 should be better matched with newer drivers from ATi!

Compared to 6800 it is a POS. Obviously better playability goes out to x700 better then the 6600GT in certain conditions in Farcry. And if 1 game does this more games will follow.

 

[rancor]

Compared to 6800 it is a POS. Obviously better playability goes out to x700 better then the 6600GT in certain conditions in Farcry. And if 1 game does this more games will follow.

geomtric intesity isn't really going to much of a factor, not till the 512 mb cards show up, since they are both tied into one and the same, more games will use high level shaders for outdoors aswell, since shaders at the texture level use both a vertex and pixel shader component both are neccesary. Far Cry really didnt' use an intensive shader for outdoors for a reason, because it hurt too much on both cards, its gotta be one or the other (well until they rerelease the sm 1.2 patch). Plus thier engine renders about 150k polys max for thier outdoor scenes, They really can't go much further because of HDR, once that is in there it will effectively turn that 150k to 300k as it takes 2 passes, but HDR is more at the pixel level then the vertex level, so first pass hurts nV, and second pass hurts ATi when they come out with and sm 3.0 part if its the same architecture.

Also the 6800nu takes on the 9800xt and beats it even with aa and af on, partlly due to the 12 pipelines, and its got 128 mb vram less (and this is why you see the margin decrease with high aa and af) not that its slower by any means.


So you can't just generalize on future games, its pretty much equal!

 

[CleanSlate]

Compared to 6800 it is a POS. Obviously better playability goes out to x700 better then the 6600GT in certain conditions in Farcry. And if 1 game does this more games will follow.

You're comparing a pci-e mid-range solution to an agp enthusiast solution? Of course it'll get trampled.

~Adam

 

[Marvelous]

geomtric intesity isn't really going to much of a factor, not till the 512 mb cards show up, since they are both tied into one and the same, more games will use high level shaders for outdoors aswell, since shaders at the texture level use both a vertex and pixel shader component both are neccesary. Far Cry really didnt' use an intensive shader for outdoors for a reason, because it hurt too much on both cards, its gotta be one or the other (well until they rerelease the sm 1.2 patch). Plus thier engine renders about 150k polys max for thier outdoor scenes, They really can't go much further because of HDR, once that is in there it will effectively turn that 150k to 300k as it takes 2 passes, but HDR is more at the pixel level then the vertex level, so first pass hurts nV, and second pass hurts ATi when they come out with and sm 3.0 part if its the same architecture.

Also the 6800nu takes on the 9800xt and beats it even with aa and af on, partlly due to the 12 pipelines, and its got 128 mb vram less (and this is why you see the margin decrease with high aa and af) not that its slower by any means.


So you can't just generalize on future games, its pretty much equal!

You say it's not a factor but it is a factor now. Look at farcry I gave you link.

Cleanslate. I wasn't comparing 6800 to x700. Just in that particular scene. x700 gets really close to 6800 results in Farcry Pier level.

 

[rancor]

You say it's not a factor but it is a factor now. Look at farcry I gave you link.

Cleanslate. I wasn't comparing 6800 to x700. Just in that particular scene. x700 gets really close to 6800 results in Farcry Pier level.

Its equal a factor as other parts are too, so they equalize themselves. And if you can't understand that, enummerate why you think so because so far you have not showed any where you get your crazy ideas about shader performance, fillrates, etc. as nV being less all in all as one. You can't just point out one and not talk about the other because they go hand in hand. Especially true for future games. If you go through the numbers you will see the end results, are plain and clear, the cards are equal in performance.

Texture fillrates and pixel fillrates can do very little if you have less pixel shader performance as was seen with the 9800 vs the 5900 line.

Right now ATi has more texture and pixel fillrates but thier pixel performance is whats slowing them down. And for nV vertex shader performance is slowing them down. And this is what you are seeing.

And since vertex shader performance and pixel shader performance are both utilized, for now in differenent aspects of gaming, later on will be used at the same time, you can't talk about one and not the other.

And when I say its a non factor, intruth it is, because right now out door scenes only stress vertex levels, not pixel shader levels which indoors levels stress, and pixel shader levels will raise in the future games during outdoor scenes, not the vertex level. What will happen is during outdoors scenes you will see an equilization of frame rates.

oh yeah but what about the 9800xt its getting less then the x700 at lower resolutions should that be happening? As you can see as resolution increases, the x700xt is getting a bigger hit then the 6800 and the 9800xt and this is because of the 128 bit memory interface, it becomes the bottleneck. Before the bottleneck was geomtric calculations.

The x700xt is supposed to replace the 9800 pro which as a 256bit memory interface but doesn't have the shader performance of the x700xt. The 9800 pro was very close to the 9800xt just that it had a little slower clocks and 128 mb less ram.

The 6800 goes up against the 9800xt. The numbers are all relative.

Just for more instances of varing shader performance,

http://www.xbitlabs.com/articles/video/display/ati-radeon-x700-games_9.html

clearly shows you a 6800nu with 128mb vram keeping up with an x800 pro in a shader heavy level in far cry. And x8 aa and x16 af only hurt the 6800nu by 18 fps, while the x800pro lost 28 fps

now with the geometric intesive level, the 6800nu lost 33 fps and the x800pro lost 27.

The % loss for the different types of levels clearly state that nV has much higher pixel shader performance while the x800's have higher performance in vertex shader performance but the discrepency is not as great as the pixel shader difference.

 

[Marvelous]

nice one.

But you got to realize without fillrate or bandwidth these pixel, vertex shaders are useless.

In your little summary you've answered which is more powerful card.

XT=bigger fillrate, more bandwidth, better vertex shaders

GT=smaller fillrate, less bandwidth, better pixel shaders

Pixel shaders often require data from the vertex shader. And is driven by vertex shaders. So in theory GT will get peformance drop no matter what shaders are used.

 

[rancor]

nice one.

But you got to realize without fillrate or bandwidth these pixel, vertex shaders are useless.

In your little summary you've answered which is more powerful card.

XT=bigger fillrate, more bandwidth, better vertex shaders

GT=smaller fillrate, less bandwidth, better pixel shaders

Pixel shaders often require data from the vertex shader. And is driven by vertex shaders. So in theory GT will get peformance drop no matter what shaders are used.

LOL really again you don't know anything about that. I'm going to go through this step by step,

First of all, fillrates can be split up to two different catagories,

a)polygon/vertex fillrates
b)pixel fillrates.

Texture bandwidth has nothing to do with either of these, since textures are sent the vram to the gpu, the fillrates are the calculations neccessary to give you the effect on your monitor.

Now as you said the gt will always get hurt since it uses both types of shaders to draw to the screen, but that goes to ATi too, can't have a pretty enviroment without some sort of textures either procedural or hand drawn, both are stored in vram and both go through the same process just one is hand drawn and the other calculated at load time. Then go through a pixel shader for the final results, depending on that shader is what we were talking about.

Just as a side note, all fillrate numbers that reviewers show are PEAK THEORETICALLY FILL RATES, and the graphics cards sites too, they also have another called VERTICES PER SECOND, which pretty much is your polygon/vertici fill rates.

If you go by those numbers we should be able to draw something like 20 million polygons to the screen at 30 fps, but it doesn't work that way because lighting slows down the gpu some, then the shader operations will slow down some more, etc, etc,etc.

 

[Marvelous]

LOL really again you don't know anything about that. I'm going to go through this step by step,

First of all, fillrates can be split up to two different catagories,

a)polygon/vertex fillrates
b)pixel fillrates.

Texture bandwidth has nothing to do with either of these, since textures are sent the vram to the gpu, the fillrates are the calculations neccessary to give you the effect on your monitor.

Now as you said the gt will always get hurt since it uses both types of shaders to draw to the screen, but that goes to ATi too, can't have a pretty enviroment without some sort of textures either procedural or hand drawn, both are stored in vram and both go through the same process just one is hand drawn and the other calculated at load time. Then go through a pixel shader for the final results, depending on that shader is what we were talking about.

Just as a side note, all fillrate numbers that reviewers show are PEAK THEORETICALLY FILL RATES, and the graphics cards sites too, they also have another called VERTICES PER SECOND, which pretty much is your polygon/vertici fill rates.

If you go by those numbers we should be able to draw something like 20 million polygons to the screen at 20 fps, but it doesn't work that way because lighting slows down the gpu some, then the shader operations will slow down some more, etc, etc,etc.

There you go again with your mumble jumble (do you feel special by talking all that garbage?), I understand basic of 3d rendering your don't have to explain. But in your summary you suggested that all the outdoor scenes are Pixel Shader related. Even though you don't have any proof I went by your word. In your logic you say 6800 has good pixel shaders but to use that shader it will also need access the vertex part of the 6800 which you say is slow on the 6800. So what does this point out? That it will be slow regardless.

But in my logic fillrate, bandwidth=power. But in your logic SM30 will bigger impact.

If this was the case regular 6800 will run all over the Radeon XT or even XTPE on SM30 games. But it doesn't so I proved my point here.

 

[joecuddles]

Since I do prefer ATI and would like the higher image quality, there's really only one way to choose which card.

A coin toss.

Getting PC on October 3rd, so I'll be holding out for an XT a bit longer, if not, I'll post the momentous coin toss occasion results, or maybe I won't. I personally think HardOCP needs to do some flashed OC'd Pro's vs. a few GT's to Ultra speeds in a head to head, rather than random people posting random results. It'd be nice, of course it won't happen, but I can dream.

Thanks for the replies and interesting arguments everyone

 

[rancor]

There you go again with your mumble jumble (do you feel special by talking all that garbage?), I understand basic of 3d rendering your don't have to explain. But in your summary you suggested that all the outdoor scenes are Pixel Shader related. Even though you don't have any proof I went by your word. In your logic you say 6800 has good pixel shaders but to use that shader it will also need access the vertex part of the 6800 which you say is slow on the 6800. So what does this point out? That it will be slow regardless.

But in my logic fillrate, bandwidth=power. But in your logic SM30 will bigger impact.

If this was the case regular 6800 will run all over the Radeon XT or even XTPE on SM30 games. But it doesn't so I proved my point here.

But in my logic fillrate, bandwidth=power

Your logic is wrong thats why the 5900's failed!

If this was the case regular 6800 will run all over the Radeon XT or even XTPE on SM30 games. But it doesn't so I proved my point here

We weren't talking about sme 3.0 games were we, SM 3.0 will speed things up to a certian point but most of thier shaders will require 2 passes. It will be perfectly possible with nV's current technology.LOL you can't understand a single thing I said because I'm not talking about the 3d render pipeline, I'm not only talking about the 3d render pipeline, 3d render pipeline is only a small part of it. Just for your information performance has nothing to do with 3d render pipeline

I understand basic of 3d rendering your don't have to explain. But in your summary you suggested that all the outdoor scenes are Pixel Shader related.

You have no clue of what your are talking about, go back to your "Barney Funtime Music Maker Sound Book" because I just explained it to you about 20 different times, and my 10 year nephew understands it and is laughing.

But in your summary you suggested that all the outdoor scenes are Pixel Shader related. Even though you don't have any proof I went by your word. In your logic you say 6800 has good pixel shaders but to use that shader it will also need access the vertex part of the 6800 which you say is slow on the 6800

Without a pixel shader you can't draw to the screen period. You have twisted logic because ATi also needs to use the same functions as nV does (fixed by Dx or Ogl), so go back to to Barney Funtime Music Maker Sound Book and enjoy yourself.

I'm wasn't trying to mock your inabality to understand, but talking ignorantly of something that is at this point can be seen as way beyond your intellectual grasp because I have explianed the same things for the last page and half in numours ways.

Anyhow my nephew wants to play ut 2004 ciao! great bunch of laughs, oh btw my nephew really feals sorry for you, don't be ashamed, he already made his first simple 3d engine lol

 

[Marvelous]

Your logic is wrong thats why the 5900's failed!



We weren't talking about sme 3.0 games were we, SM 3.0 will speed things up to a certian point but most of thier shaders will require 2 passes. It will be perfectly possible with nV's current technology.LOL you can't understand a single thing I said because I'm not talking about the 3d render pipeline, I'm not only talking about the 3d render pipeline, 3d render pipeline is only a small part of it. Just for your information performance has nothing to do with 3d render pipeline




You have no clue of what your are talking about, go back to your "Barney Funtime Music Maker Sound Book" because I just explained it to you about 20 different times, and my 10 year nephew understands it and is laughing.



Without a pixel shader you can't draw to the screen period. You have twisted logic because ATi also needs to use the same functions as nV does (fixed by Dx or Ogl), so go back to to Barney Funtime Music Maker Sound Book and enjoy yourself.

I'm wasn't trying to mock your inabality to understand, but talking ignorantly of something that is at this point can be seen as way beyond your intellectual grasp because I have explianed the same things for the last page and half in numours ways.

Anyhow my nephew wants to play ut 2004 ciao! great bunch of laughs, oh btw my nephew really feals sorry for you, don't be ashamed, he already made his first simple 3d engine lol

Is that what you do in your spare time. Hump Barney. What you do with your own time is none of my business.

I'm trying to have genuine conversation. You sit there tease hardy har har. You make me laugh.

I'm going by your logic but when I try to use your logic you just go ahead and call me names. Haha I go read Barney now..


Your hopeless. Without fillrate and bandwidth SM30 is hopeless. When real games that actually use SM30 come out we wil have much faster and better cards. Usually first cards that come out with new technology cannot even utilize the technology at all or is too slow. For instance T&L hype from nVidia when geforce ddr were made. Damn where's the T&L now. In that situation it did helped like 5fps. But fill rate does the same thing it pushes more fps.

Whether you like it or not XT has more fill rate and pushes more fps.

5900 failed because it was poorly designed. 4 pixel pipeline and 2 texture units will not beat a 8pixel and 1 texture unit.

 

[little squirt]

Hi Marvelous, I'm Rancor's nephew, I asked him to write those things in.

I'm sorry if I offended you.

I didn't like it when you said my Uncle was talking Garbage.

You don't know anything close to what he knows.

I hope when I get older I won't be as arrogent as you!

 

[rancor]

Is that what you do in your spare time. Hump Barney. What you do with your own time is none of my business.

I'm trying to have genuine conversation. You sit there tease hardy har har. You make me laugh.

I'm going by your logic but when I try to use your logic you just go ahead and call me names. Haha I go read Barney now..


Your hopeless. Without fillrate and bandwidth SM30 is hopeless. When real games that actually use SM30 come out we wil have much faster and better cards. Usually first cards that come out with new technology cannot even utilize the technology at all or is too slow. For instance T&L hype from nVidia when geforce ddr were made. Damn where's the T&L now. In that situation it did helped like 5fps. But fill rate does the same thing it pushes more fps.

Whether you like it or not XT has more fill rate and pushes more fps.

5900 failed because it was poorly designed. 4 pixel pipeline and 2 texture units will not beat a 8pixel and 1 texture unit.

Thx little squirt, anyways back to the topic

Whether you like it or not XT has more fill rate and pushes more fps.

http://www.hothardware.com/viewarticle.cfm?articleid=438
http://www.devhardware.com/c/a/Video Cards/FX-5900-Ultra-vs-R9800-Pro


Wrong again!

The fx line had very poor pixel shader execution which the texel fillrates couldn't over come.

I'm going by your logic but when I try to use your logic you just go ahead and call me names. Haha I go read Barney now..

Can you grasp all of the logic or only half the half you want to hear?

Poorly designed the 4x2 TMU's still provided a huge bandwidth advantage too. Bandwidth is out of the picture!

Games that utilize sm 3.0 will work just fine with this cards, but just not at the highest of settings. As the 9700's show its tech will last for a good 2 years, and this really has nothing to do with fill rates, well very little, as you can see performance on these cards are very close, and by using x number of passes will slow down these cards x number of times, even far cry goes above 100fps at 1200x1024 with no aa and af, cut that down by 3 you can do shaders up to 3 passes and they will still work fine (with branching and looping there really shouldn't be any need to go more then 2, so now you can increase geometric complexity)

http://www.planetquake.com/features/articles/editorials/pixels-1.shtml

As they say fillrates don't matter!