Why SM3.0 != NVIDIA only

[Clay]

And since there is not a single "pure SM3.0" announced yet, why do you need it again? Oh, excuse me, Unreal 3 in '06/07 which your 6800 will absolutely blaze thru @ 320x240. In fact, HL2 would be the closest thing to a "pure SM2.0" game and it's not even out yet. So this year, and quite possibly all of next year, there will not be one game displaying effects not possible on ATI's X800s.What about some of these upcoming titles? I'd like to find out more details as to how much they utilize SM3.0:
Lord of the Rings, Battle For Middle-earth
STALKER: Shadows of Chernobyl
Vampire: Bloodlines
Splinter Cell X
Tiger Woods 2005
Madden 2005
Painkiller(in an upcoming patch)
FarCry (in the revised patch or what was thought to be the "next" 1.3 patch that would include HDR, displacement mapping...of which I know are feasible in SM2.0 but at what expense I'm not sure until I see firsthand)

[schuey74]

Rage, yes an old arguement which I don't want to bring up again. I only pointed it out so everyone can see the extreme and just plain silly bias that Mr. Nv40 has. And Clay, none of those games will be "pure SM3.0" as Mr. Nv40 states. I'm just showing how you can't have it both ways. You can't say SM3.0 is necessary and then when it becomes apparent that it isn't, turn around and say wait for the true SM3.0 titles, of which the closest one is Unreal 3 in a minimum of two years.
Now, I do think that some "SM3.0 only" effects will show up in games before that, but that's not what NV40 says - he spins everything to favor Nvidia and it's gotten to the point where he's a complete joke. I don't spend all my time preaching about how 3dc is going to give the X800s a big advantage in HL2 & SS2 because that is yet to be seen, although just like Doom 3 it should be common sense. Both of these architectures have their respective advantages over each other, yet Mr. Nv40 is actually naive enough to think the opposite.

[Ruined]

What about some of these upcoming titles? I'd like to find out more details as to how much they utilize SM3.0:
Lord of the Rings, Battle For Middle-earth
STALKER: Shadows of Chernobyl
Vampire: Bloodlines
Splinter Cell X
Tiger Woods 2005
Madden 2005
Painkiller(in an upcoming patch)
FarCry (in the revised patch or what was thought to be the "next" 1.3 patch that would include HDR, displacement mapping...of which I know are feasible in SM2.0 but at what expense I'm not sure until I see firsthand)


Don't forget MOH: Pacific Assault

[Clay]

Rage, yes an old arguement which I don't want to bring up again. I only pointed it out so everyone can see the extreme and just plain silly bias that Mr. Nv40 has. And Clay, none of those games will be "pure SM3.0" as Mr. Nv40 states. I'm just showing how you can't have it both ways. You can't say SM3.0 is necessary and then when it becomes apparent that it isn't, turn around and say wait for the true SM3.0 titles, of which the closest one is Unreal 3 in a minimum of two years.
Now, I do think that some "SM3.0 only" effects will show up in games before that, but that's not what NV40 says - he spins everything to favor Nvidia and it's gotten to the point where he's a complete joke. I don't spend all my time preaching about how 3dc is going to give the X800s a big advantage in HL2 & SS2 because that is yet to be seen, although just like Doom 3 it should be common sense. Both of these architectures have their respective advantages over each other, yet Mr. Nv40 is actually naive enough to think the opposite.Right, that's along the lines of what I'm expecting for the next six months or so...the SM3.0 only effects along with some potential performance gains from SM3.0 (and hopefully SM2.0b for ATI's current gen). I didn't mean to imply that some of those titles will be SM3.0 exclusive...that wouldn't make any sense at all considering the current user base of cards out there. I'd just like to find out more details about exactly how each game will utilize SM3.0. I'm not addressing what NV40 said, everyone has their opinions and all of us will rarely agree 100% of the time. :)

[schuey74]

Yes, but at some point opinion must be seperated from downright lies. You can not be of the opinion the SM3.0 is necessary one day and then say it isn't necessary the next day. Only one of those can be true, not both. And then when it becomes apparent that ATI's current hardware can handle most, if not all of the SM 3.0 effects Nvidia is touting, turn around and be so hypocritical as to say that ATI is holding back the industry because they are giving more work to developers. It's common knowledge that Nvidia is giving plenty of extra work to developers as well. This guy makes HB's ATI bias look tame in comparison.

[Riptide]

This guy makes HB's ATI bias look tame in comparison.
Oh, you mean with statements like these?
if im a fan it is about technology and computer graphics ..; (hence why i use Nvidia hardware)
Emphasis on that last statement in parentheses. ;)

[evilchris]

Yes, but at some point opinion must be seperated from downright lies. You can not be of the opinion the SM3.0 is necessary one day and then say it isn't necessary the next day. Only one of those can be true, not both. And then when it becomes apparent that ATI's current hardware can handle most, if not all of the SM 3.0 effects Nvidia is touting, turn around and be so hypocritical as to say that ATI is holding back the industry because they are giving more work to developers. It's common knowledge that Nvidia is giving plenty of extra work to developers as well. This guy makes HB's ATI bias look tame in comparison.

Nothing makes HB's blatant fanboism look tame. Nothing! HB had inside sources CONFIRMING 6800U was not a "true" 16x1 part!

:lol2:

[LovesuckZ]

And since there is not a single "pure SM3.0" announced yet, why do you need it again?

Can you tell me one "pure SM2.0" Game since autumn 2002?

In fact, HL2 would be the closest thing to a "pure SM2.0" game and it's not even out yet.

You can play HL2 with a DX6 card, no "pure SM2.0".

So this year, and quite possibly all of next year, there will not be one game displaying effects not possible on ATI's X800s.

Yes and for all the games you don't need a r420 or NV40 because the 5200 displays all effects...

[Demirug]

The abuse of the DirectX shadermodels for marketing purposes is not funny anymore. It was not funny during the time of the first HL2 performanceresults from valve. It is not funny today with the Far Cry patch on shadermodel specific paths.

Fortunately this game will terminated by Microsoft because SM3 is they last shadermodel ever.

Yes, I am annoyed. Valve sells you simplest shader that can easly run with SM1.1 as hightech SM2.0 shaders. And Crytek sells you they same shader twice. The SM30 and SM2B Path use almost they same HLSL pixelshader script. The 2B version is a copy of the 30 code with only 8 lines changed from over 450. This 8 lines allready includes the changes in the comments. Maybe they sell us the same shader sgain as SM2A path. With SM2.0 they have another possible sales of they same thing with a new name.

Shadermodels should IMHO nothing more than an assistance to determine which chips can execute a HLSL shader. Sure a chip that supports all models is a developers friend because it makes development much more easier but in most cases your development will end with HLSL code that can compile to more than SM3.

Features are nice for developers but gamers should ask how much this features can improve the speed of effects. Permit they new effects in real time? Or can a effect used on more pixel on they screen than before? It doesn't matter how they final image is produced. As a gamer i did not care which shadermodel is used. As a developer I am naturally curious about.

It is good that we have a SM3 chip today because it will help to develop they games of tomorrow. And maybe make some games of today faster. But if somebody think about using only SM3 before even reaching the limits of SM2 something is going wrong.

[jbirney]

This guy makes HB's ATI bias look tame in comparison.

:)



Demirug,
good post!

[Pandora's Box]

Can you tell me one "pure SM2.0" Game since autumn 2002?



You can play HL2 with a DX6 card, no "pure SM2.0".



Yes and for all the games you don't need a r420 or NV40 because the 5200 displays all effects...


it comes down to the fact that there might not be a point rendering something in ps2.0 when it could be done with the same image quality in ps1.1.

[Nv40]

Rage, yes an old arguement which I don't want to bring up again. I only pointed it out so everyone can see the extreme and just plain silly bias that Mr. Nv40 has. And Clay, none of those games will be "pure SM3.0" as Mr. Nv40 states. I'm just showing how you can't have it both ways. You can't say SM3.0 is necessary and then when it becomes apparent that it isn't, turn around and say wait for the true SM3.0 titles, of which the closest one is Unreal 3 in a minimum of two years. Now, I do think that some "SM3.0 only" effects will show up in games before that, but that's not what NV40 says - he spins everything to favor Nvidia and it's gotten to the point where he's a complete joke. I don't spend all my time preaching about how 3dc is going to give the X800s a big advantage in HL2 & SS2 because that is yet to be seen, although just like Doom 3 it should be common sense. Both of these architectures have their respective advantages over each other, yet Mr. Nv40 is actually naive enough to think the opposite.

when i say "pure Sm3.0" is pure sm3.0 effects .. not games. pureSm3.0 games will not happen until every gamer even in the low end market have Sm3.0 hardware.if we are lucky that will be 2006.
Sm3.0 pure effects is another thing ,and it is quite possible to see few games this year or next one to use some SM3.0 pure effects that can't be done in non sm3.0 hardware. because sooner or later any gamedeveloper from ATi or Nvidia program or from neither .that wants to be updated with the latest Dx9 standar , "unfortunately" will need to add Sm3.0 to their engine. Valve already said it will add it in their engine. there is no conspirancy here ,its just the progress.the title and what was underlined in the original post of this thread says it all. SM3.0 != (is not equal)Nvidia only.

never said Sm2.0 should be dropped ,only anything that they feel cost/extratime and have no future..->read the first post to understand what i mean.

[Clay]

:)



Demirug,
good post!
Why the smiley face about someone saying "fanboi-A makes fanboi-B's ATI bias look tame in comparison."? We don't need cheerleaders when someone compares the alleged fanboi-factor of one to the other, just makes you look more like a fanboi yourself. ;)

And yes that was a good post Demirug. Although I think some might be putting a few words into others' mouths from both sides of this debate. I don't think anyone with a lick of sense would suggest that SM2.0 is "bad" and only a "pure SM3.0 game" (whatever that means, we wouldn't realistically see that for at least another 2~3 years) is the best idea.

I would be very interested in seeing the code blocks that you (Demirug) are referring to. I've not programmed with the DX API but I do write code for a living so I can understand the logic. If you can't post it here could you PM or email it?

[Demirug]

The two files are "CGRCLightTempl_PS30.crycg" and "CGRCLightTempl_PS2B.crycg". You can find both in ther Shader.pak file.

The first two changes are only in the fileheader and change the text "PS30" in "PS2B". The next change is even part of the header. They change a "PS30Only" in "PS2BOnly".

The next two changes (line 129 and 142) are nessary because they use "COLOR" as usagetype for the inputregister. For PS2B this is changed to "TEXCOORD". Both register store some vectors using "COLOR" usage for this looks very strange. But with PS 3.0 all usagetypes are equal. This is not they case with all SM2 Models. The "COLOR" usagetype will clamp to [0..1) and guaranteed only a low precesion (even lower than FP16).

The last 3 changes (line 431, 449, 451) are necessary because PS 3.0 force you to do they fog calculation in the shader.

In any case the code allways contains many #ifdef for the preprocessor adding an other that switch fog on and off would not change much.

I did not see the C++ Code that control the usage of this shaders but the only difference should the number of lights put in one pass and the place the fog color is set. The number of lights can easy managed with a variable and a little if will do the fog job.

[Clay]

The two files are "CGRCLightTempl_PS30.crycg" and "CGRCLightTempl_PS2B.crycg". You can find both in ther Shader.pak file.

The first two changes are only in the fileheader and change the text "PS30" in "PS2B". The next change is even part of the header. They change a "PS30Only" in "PS2BOnly".

The next two changes (line 129 and 142) are nessary because they use "COLOR" as usagetype for the inputregister. For PS2B this is changed to "TEXCOORD". Both register store some vectors using "COLOR" usage for this looks very strange. But with PS 3.0 all usagetypes are equal. This is not they case with all SM2 Models. The "COLOR" usagetype will clamp to [0..1) and guaranteed only a low precesion (even lower than FP16).

The last 3 changes (line 431, 449, 451) are necessary because PS 3.0 force you to do they fog calculation in the shader.

In any case the code allways contains many #ifdef for the preprocessor adding an other that switch fog on and off would not change much.

I did not see the C++ Code that control the usage of this shaders but the only difference should the number of lights put in one pass and the place the fog color is set. The number of lights can easy managed with a variable and a little if will do the fog job.
Excellent, thanks! :)

[Demirug]

Excellent, thanks! :)

No Problem as I have write the same suff allready in german.

Do you need anything else?

a) SM2B for NV40?
b) SM2B for NV3X?
c) SM30 for NV3X?
d) SM2B for R3XX?

[Clay]

No Problem as I have write the same suff allready in german.

Do you need anything else?

a) SM2B for NV40?
b) SM2B for NV3X?
c) SM30 for NV3X?
d) SM2B for R3XX?
Not at the moment, thanks though. I'm scrambling for enough free time as it is right now so what you've provided will give me enough to check out next chance I get. :)

[jbirney]

Why the smiley face about someone saying "fanboi-A makes fanboi-B's ATI bias look tame in comparison."? We don't need cheerleaders when someone compares the alleged fanboi-factor of one to the other, just makes you look more like a fanboi yourself. ;)

You know clay you have been hounding me for the last few weeks. Every time I post something not even talking to you, you have to come in and jump on my case. This did not happen until I called you out on something you said to DaveB in an old thread. Can you please lay off as its getting old.

[ragejg]

jbirney, do that thru a pm please.

[Clay]

You know clay you have been hounding me for the last few weeks. Every time I post something not even talking to you, you have to come in and jump on my case. This did not happen until I called you out on something you said to DaveB in an old thread. Can you please lay off as its getting old.I don't fully recall what you "called me out on" but I'm pretty sure it had nothing to do with you other than to rush to get your $0.02 in and show your support for Dave Baumann. Regardless, my moderating focus on you of late is not related whatsoever. None of this stuff is anything I ever take personally so there's no "hounding" going on. I do not deny that I am "focusing" on you because of your veiled and suspect posts. We're not here to sit idly by while propaganda is dished out (no matter how subtly it may be being served). If you have a personal issue with my moderating of your posts then feel free to PM me or MikeC.

[jbirney]

I do not deny that I am "focusing" on you because of your veiled and suspect posts. We're not here to sit idly by while propaganda is dished out (no matter how subtly it may be being served).

But yet you do let the pro-NV/Anti-ATI propaganda flow freely here as I did not see you saw a word about NV40 (http://www.nvnews.net/vbulletin/showthread.php?t=33190) or Runnied lastest thread's (http://www.nvnews.net/vbulletin/showthread.php?t=32981). These post were nothing more than attempts to smear ati for lack of PS3.0 support. How does these topics add value to this part of the forum? Those topics have been discussed a bunch already. Been there done that. Time to move on? Yes I understand that this site caught way too much flak during the Fx days. Way too much! But know I see the same people lashing out with the same old argurements against ATI. Is this pay backs for the FX days? What ever it is, if your really trying to stop propaganda, you might want to make more of an effor to stop it on both sides.

[MikeC]

JBirney I'm not sure why you are ATI's watchdog but if you're unsatisfied with the way the site is run you're free to leave. You are at a NVIDIA fansite not ATI's, you can't control what members want to talk about. That's the real problem here and until you can deal with that you're going to be miserable.

[Clay]

But yet you do let the pro-NV/Anti-ATI propaganda flow freely here as I did not see you saw a word about NV40 (http://www.nvnews.net/vbulletin/showthread.php?t=33190) or Runnied lastest thread's (http://www.nvnews.net/vbulletin/showthread.php?t=32981). These post were nothing more than attempts to smear ati for lack of PS3.0 support. How does these topics add value to this part of the forum? Those topics have been discussed a bunch already. Been there done that. Time to move on? Yes I understand that this site caught way too much flak during the Fx days. Way too much! But know I see the same people lashing out with the same old argurements against ATI. Is this pay backs for the FX days? What ever it is, if your really trying to stop propaganda, you might want to make more of an effor to stop it on both sides.Ugh, regarding the first link you provided. ChrisRay was handling that thread just fine...I didn't see a need to join in but I did say this (http://www.nvnews.net/vbulletin/showpost.php?p=384838&postcount=54) which certainly did not SUPPORT how NV40 was coming across. As for the second link, I was busy and left it to other mods. Not sure why you feel the need for me to explain myself to you nor do I even know why I'm bothering really. If you read my posts over time you'll see that I do not support NV fanbois, actually most times I laugh it off because most of them are just kids. I don't think you're just a kid though and your timing is impeccable so that's why I watch/challenge you. If you can't handle that or are unable to offer a rebuttal to my questions then I suggest you either stop whining or just move on.

[ChrisRay]

But yet you do let the pro-NV/Anti-ATI propaganda flow freely here as I did not see you saw a word about NV40 (http://www.nvnews.net/vbulletin/showthread.php?t=33190) or Runnied lastest thread's (http://www.nvnews.net/vbulletin/showthread.php?t=32981). These post were nothing more than attempts to smear ati for lack of PS3.0 support. How does these topics add value to this part of the forum? Those topics have been discussed a bunch already. Been there done that. Time to move on? Yes I understand that this site caught way too much flak during the Fx days. Way too much! But know I see the same people lashing out with the same old argurements against ATI. Is this pay backs for the FX days? What ever it is, if your really trying to stop propaganda, you might want to make more of an effor to stop it on both sides.


Look, It's not Clay's responcibility to come argue every post made by Nvidia fanboys on an Nvidia fansite, Nor is it Mikes or any mods. This is an Nvidia fansite, There are obviously going to be pro Nvidia people. You cant expect us to come out yelling at someone everytime they say something negative ATI,

A Certain level of bias is expected, I imagine most people here understand not to take everything NV40/Ruined says as gospel when it comes to video cards, And I'm certainly up for challenging some opinions at times. But not every mod has the time to come here and babysit and smack peoples hands. Expecially for being an nvidia fan at an nvidia fansite of all places.

Chris

[jbirney]

Look, It's not Clay's responcibility to come argue every post made by Nvidia fanboys on an Nvidia fansite, Nor is it Mikes or any mods. This is an Nvidia fansite, There are obviously going to be pro Nvidia people. You cant expect us to come out yelling at someone everytime they say something negative ATI

Of course I would not expect that. Come on Chris I have been here and posting for long enough time now. I know what to expect.


But not every mod has the time to come here and babysit and smack peoples hands. Expecially for being an nvidia fan at an nvidia fansite of all places.

No but he does go out of his way to smack me (which he even admitted to). And I dont understand why now. I have cut my post count down, stuck mainly to the the Other Video Card and games forum and have tried to clean up my act. I used to be much worse. But eh bigger things in life to worry about :)