The Ultimate SM3.0 Game Thread

[Bar81]

Originally posted by: BFG10K

If you're not going to stay on point then get out of the discussion.

I am on point, you just can't read and/or comprehend.

You can't just make the discussion about whatever you want.

Isn't it clear that the discussion is about SM 3.0? You can't just ignore arguments because they don't fit into your childish view of the world.

Please get an education and learn what ignorant means.

I'll do that right after you have your twelfth birthday.

Thanks again for adding to the discussion. <thread IQ drops 3 points> Now do us all a favor and get your ignorant self into another thread where you can spout your pointless drivel.

 

[Bar81]

Originally posted by: apoppin

Originally posted by: Bar81

Originally posted by: VirtualLarry

Originally posted by: Jeff7181

Originally posted by: Bar81
I think this could be potentially devestating news for SM3.0 on current cards. If nvidia is making what amounts to a SM3.0B spec for their next generation cards what does that tell us about today's implementation of SM3.0 and it's usefullness in upcoming games played with a 6800 series card?

That's not what's being said at all... did you miss the part saying "I don't think you'll see any radical changes in architecture?" What does that mean? To me it means they'll bump up the speed to handle HDR and displacement mapping without halving frame rates.

But likewise, that implies that current SM3.0-capable parts, have performance issues that will result in halved frame-rates when those gfaphics features are enabled in-game. Hmm.

Also, when an NV spokesperson talks about minimal changes in architecture, that doesn't mean that you might not see radical changed in the hardware implimentation of that architecture. What is I mean is that he might have been speaking of the software architecture/programming-model of SM3.0, which likely there won't be any major revisions to again any time soon. But that doesn't mean that there might not be potential performance issues with current-gen SM3.0 parts, and that NV might not take some radical steps with next-gen parts to alleviate some of those issues (such as implementing "virtual pipelines" and SMT). Still, without proof, this is all just speculative commentary on my part, but it seems logical to me.

Just got back, there's your answer apoppin, said much more eloquently than I could have said it.

what?

might . . . might .. . maybe . . . . doesn't mean . . . . might not . . . . speculative and . . . without proof . . . seems logical

yeah, that's pretty eloquent

I said more eloquent than *me*, so be happy I didn't reply

 

[BFG10K]

Wasn't there some discussion already that HL2 would "perform better" on ATI cards, and "take advantage of ATI's special features"?

That was about shader performance in general given the NV3x couldn't run SM 2.0 with any reasonable level of performance. Also I recall something about changing the AA sample pattern on ATi's cards which could make HL2's packed textures look better when AA was enabled, though I don't know how that ended up in the end.

Geometry-instancing, is more of a driver-level feature AFAIK,

I wouldn't think so given it appears to only be available on R4xx boards. If it's only at the driver level then one would imagine ATi would stick it on earlier hardware for free performance boosts.

Free potential loss of performance is always a good thing.

The potential for bad programming by bad programmers is always present regardless of the language, but that isn't really related to the topic at hand. SM 3.0 provides the tools to make SM 2.0 run better, just like SM 2.0 did for SM 1.x.

 

[BFG10K]

Thanks again for adding to the discussion. <thread IQ drops 3 points> Now do us all a favor and get your ignorant self into another thread where you can spout your pointless drivel.

Would you like a lollipop?

 

[apoppin]

Originally posted by: Bar81

Originally posted by: apoppin

Originally posted by: Bar81

Originally posted by: VirtualLarry

Originally posted by: Jeff7181

Originally posted by: Bar81
I think this could be potentially devestating news for SM3.0 on current cards. If nvidia is making what amounts to a SM3.0B spec for their next generation cards what does that tell us about today's implementation of SM3.0 and it's usefullness in upcoming games played with a 6800 series card?

That's not what's being said at all... did you miss the part saying "I don't think you'll see any radical changes in architecture?" What does that mean? To me it means they'll bump up the speed to handle HDR and displacement mapping without halving frame rates.

But likewise, that implies that current SM3.0-capable parts, have performance issues that will result in halved frame-rates when those gfaphics features are enabled in-game. Hmm.

Also, when an NV spokesperson talks about minimal changes in architecture, that doesn't mean that you might not see radical changed in the hardware implimentation of that architecture. What is I mean is that he might have been speaking of the software architecture/programming-model of SM3.0, which likely there won't be any major revisions to again any time soon. But that doesn't mean that there might not be potential performance issues with current-gen SM3.0 parts, and that NV might not take some radical steps with next-gen parts to alleviate some of those issues (such as implementing "virtual pipelines" and SMT). Still, without proof, this is all just speculative commentary on my part, but it seems logical to me.

Just got back, there's your answer apoppin, said much more eloquently than I could have said it.

what?

might . . . might .. . maybe . . . . doesn't mean . . . . might not . . . . speculative and . . . without proof . . . seems logical

yeah, that's pretty eloquent

I said more eloquent than *me*, so be happy I didn't reply

i'm sorry - after restraining myself re: "exsssive emoticon use" - i just couldn't help it.



my POV remains SM 3.0 will be more useful in recent-upcoming games - especially that ATI has announced HW support for it in r520 - then you believe . . . the only thing i hear 'negative' are just "qualified maybes" . . . . but then we'll know - for sure - soon enough.

Best of all, the decision just isn't that 'critical' . . . either way, nVidia or ati.

 

[MisterChief]

I say we all just Nuke this Cyber-Hiroshima and get the war over with.

 

[TekXoID]

Is there any way for ATi cards to emulate SM3.0?

 

[nRollo]

Originally posted by: BFG10K

Thanks again for adding to the discussion. <thread IQ drops 3 points> Now do us all a favor and get your ignorant self into another thread where you can spout your pointless drivel.

Would you like a lollipop?

Yes, please.

 

[apoppin]

Originally posted by: TekXoID
Is there any way for ATi cards to emulate SM3.0?

SM 3.0 is just an extension of 2.0 . . . . iSM 3.0 can get so much more "done" with much less GPU "work" then its earlier version.

in other words, you may get the same visual effect with 2.0 as 3.0 but your actual in-game-performance may suffer as your GPU "struggles" to render the same thing the 3.0 enabled HW breezes thru.

more-or-less

want more technical? . . . . Shader Model 2.0b supports up to 768 instructions while Shader Model 3.0 supports up to 65536 instructions (BTW, the most complex shader in Half-life 2 is 90 instructions long and FarCry's impressive realistic-looking water shader is around 50 instructions long)

edited

 

[Keysplayr]

Originally posted by: Noob

Originally posted by: keysplayr2003

Originally posted by: Noob
BTW why isn't SC:CT going to support SM 2.0. Is this just to make people with ATI cards miserable? Or is it going to make people with 6800's miserable because UbiSoft might not optimize the code enough for SM 3.0. It seems like people with either brand card are going to lose out on the image quality and performance of SM 2.0.

Lose out? If you can't afford to upgrade in the computer world and expect to keep up with the technology, then don't play the game. The computer world moves fast for enthusiasts. Keep up, or get off the bus.

Did I say I can't keep up. I was merely asking why SC3 won't support SM 2.0 which was answered. It didn't make sense why they would exclude 2.0. And if you looked at my sig you would see that I have an X800 Pro. So you can't tell me I can't keep up. That was such an idiotic comment that you just made.

Thanks for idiotic "compliment" Mr. Congeniality. You said "Lose out". The only way someone can lose out is if they don't, or can't, buy a product that supports everything there is to offer. So in a sense you do lose out. SM2.0b is not up to date and that's exactly the card you own. Out of date the day it hit the store shelves. ATI keeps having to "play" with their drivers in order to get that blessed MS WHQL stamp of approval. And I don't know if they can even pull that off much longer. If you think I'm idiotic, then you are really living up to your name.

 

[MisterChief]

I swear to Bob, the Mods should create a Cyber-Nuke to wipe threads like this off the face of the web:laughmaniacally).

 

[apoppin]

Originally posted by: apoppin

Originally posted by: TekXoID
Is there any way for ATi cards to emulate SM3.0?

SM 3.0 is just an extension of 2.0 . . . . iSM 3.0 can get so much more "done" with much less GPU "work" then its earlier version.

in other words, you may get the same visual effect with 2.0 as 3.0 but your actual in-game-performance may suffer as your GPU "struggles" to render the same thing the 3.0 enabled HW breezes thru.

more-or-less

want more technical? . . . . Shader Model 2.0b supports up to 768 instructions while Shader Model 3.0 supports up to 65536 instructions (BTW, the most complex shader in Half-life 2 is 90 instructions long and FarCry's impressive realistic-looking water shader is around 50 instructions long)

edited

Here's a great article that sums up SM 3.0 benefits without being overly technical (an i think i read them ALL tonight):
Shader Model 3.0 - What's it all about?a couple of excerpts:

As far as Pixel Shader 3.0 goes, the most striking aspect is the minimum instruction count, which jumps way up from the 96 instructions specified in the 2.0 Extended spec all the way up to 512 instructions. How fast a shader with that many instructions would actually run is open to debate (not to mention how powerful the hardware is), but this increase certainly adds to the flexibility on offer. To complicate things further, the ps_2_b profile has also been added to the DirectX 9 HLSL - This gives all the features of Pixel Shader 2.0, but adds in support for up to 512 instructions without the other advantages 2.0 Extended has to offer . . .

. . . Pixel Shader 3.0 offers. To be honest, from a gamer's point of view these extra features aren't included to add to image quality, but rather to make the lives of developers easier, and give them the greater flexibility and power that they yearn for. Again, a lot of these features were in fact included to in the 2.0 Extended specification, but are now also available in PS 3.0. Before going into these in any detail, lets take a brief look at what the main new features are on offer are:

- Static and dynamic flow control
- Predication
- Dynamic branching
- New registers, and more temporary register usage allowed
- New gradient/texture instructions
- Centroid sampling
- Arbitrary swizzle

. . .


Until we start seeing titles explicitly written to take advantage of Pixel Shader 3.0 features, it really won't make any difference at all. We've already seen one title (Far Cry) patched to include 3.0 shader support - This may well have been recompiled to take advantage of some of the features listed above, and although it is unlikely to make a difference to image quality at all, we may see performance improvements using PS 3.0 over 2.0 - Something to test once we get some 3.0 shader capable boards in our hands...

The feature set of Pixel Shader 3.0 does also allow for some nice new visual effects of note - Mainly the ability to use floating point values for texture filtering and frame buffer blending. In particular, frame buffer blending can allow for a realistic looking 'motion blur' effect (brings back memories of 3Dfx's T-Buffer, doesn't it?), and floating point textures should improve the look of effects such as HDR (High Dynamic Range) rendering - An effect that we've already seen on PS 2.0 hardware in Half-Life 2, now taken to a new level. . . .


As we've explored, the potential for 'cool eye candy' is generally greater with the improvements made in Vertex Shader 3.0 than those seen in Pixel Shader 3.0. Certainly, if some of these features are put into effect in future titles, and performance is up to scratch, then we have something to be excited about.

oh yeah, add Age of Empires 3 to the "list" to support SM 3.0.

 

[acx]

SC:CT SM3.0 enables HDR with Tone Mapping, Parallax Mapping, and Soft Shadows.

I've got the singleplayer demo but I can't figure out how to save screenshots. The difference is very noticible between the two shader models whenever you come close to rocks/walls and lighting. Unfortunately, Sam spends a lot of time in the dark so you won't be able to appreciate the soft shadows as often.

 

[BenSkywalker]

Also, when an NV spokesperson talks about minimal changes in architecture, that doesn't mean that you might not see radical changed in the hardware implimentation of that architecture.

Yes, actually it pretty much does when he uses it the way he did-

?Well, from an architecture standpoint we?re just still at the beginning of shader model 3.0. And we need to give the programmers out there some time to continue to really learn about that architecture.

By utilizing 'we' he rather clearly indicates that they are talking about hardware as you would need a MS rep to state that there were any sort of looming changes to an already finalized and shipping standard(which actually, they couldn't do without a revision anyway). nVidia could talk about proprietary implementations of fragment programs for OGL 2.0- but they were talking directly about DX standards, not their own software.

What is I mean is that he might have been speaking of the software architecture/programming-model of SM3.0, which likely there won't be any major revisions to again any time soon.

That is a MS controlled standard, nVidia can not make changes to it. It seems you are trying to make this statement fit your argument instead of reading what it rather clearly states.

 

[Vernor]

Heh, I predicted a year ago here 3.0 to be nothing but a mirage, and that the list of games Nvidia was pimping was mostly bullshit.

These things take **years** to be actually implemented in any meaningful way.


(I didn't predicy that ATI would release one overpriced/underperforming card after another, but that's another story...)

 

[imported_Noob]

Originally posted by: keysplayr2003

Originally posted by: Noob

Originally posted by: keysplayr2003

Originally posted by: Noob
BTW why isn't SC:CT going to support SM 2.0. Is this just to make people with ATI cards miserable? Or is it going to make people with 6800's miserable because UbiSoft might not optimize the code enough for SM 3.0. It seems like people with either brand card are going to lose out on the image quality and performance of SM 2.0.

Lose out? If you can't afford to upgrade in the computer world and expect to keep up with the technology, then don't play the game. The computer world moves fast for enthusiasts. Keep up, or get off the bus.

Did I say I can't keep up. I was merely asking why SC3 won't support SM 2.0 which was answered. It didn't make sense why they would exclude 2.0. And if you looked at my sig you would see that I have an X800 Pro. So you can't tell me I can't keep up. That was such an idiotic comment that you just made.

Thanks for idiotic "compliment" Mr. Congeniality. You said "Lose out". The only way someone can lose out is if they don't, or can't, buy a product that supports everything there is to offer. So in a sense you do lose out. SM2.0b is not up to date and that's exactly the card you own. Out of date the day it hit the store shelves. ATI keeps having to "play" with their drivers in order to get that blessed MS WHQL stamp of approval. And I don't know if they can even pull that off much longer. If you think I'm idiotic, then you are really living up to your name.

To get it through your head 3.0 doesn't make an image quality difference compared to 2.0. Nore has Nvidia been able to outperform ATI even in SM 3.0 games. So that is why I went with the X800. I am saying us ATI users are losing out in SC3 is because we only get SM 1.1. I never said I couldn't keep up with tech. And this is a non debatable fact. The X800 outperforms the 6800. You think I would buy a card that's SM 3.0 supported when there is a huge chance programmers won't get the full benefits out of SM 3.0 in time for next gen games that 6800's can run. No way. Plus in the end the X800's would look better because of their AA and AF. Not to mention the X800's better utilization of memory bandwidth when enabling AA and AF. It barely takes a performance plundge. The 6800 is different and takes huge performance plundge when enabling these options. I am not critizing anybody who has a 6800. When they first came out, I would have bought them to if I hadn't have found out that 3.0 doesn't make an image quality difference. So now that we know it will be a long time before programmers can get the full performacne out of SM 3.0, it will be too late considering the next gen games were SM 3.0 is fully supported would be to demanding for the 6800 or the X800 to run. All I was saying us ATI users won't get the benefits of SM 2.0 in SC3 because it isn't supported. Then you come out with a stupid comment saying "I can't keep up with tech."

 

[nRollo]

Then you come out with a stupid comment saying "I can't keep up with tech."

You're not keeping up with tech Noob. Sure, you can run some games a little faster, but not enough that you can notice the difference.

You traded soft stencil shadows, and IMO after playing Doom3 on my X800XT PE, stencil shadows in general to say "Wow I get 80fps at Far Cry instead of 73fps!"

You gave up being able to see HDR in Far Cry.

You gave up modern graphics in SC:CT.

Etc.

That is not "keeping up"- when you have to play the games without the features the developers put in them, you're falling behind.

 

[imported_Noob]

It has been said by many before. 3.0 doesn't make an image quality difference compared to 2.0. How many times does one have to say it? Plus the 6800 can't even handle these options very well anyway. Look at COR. And the FPS between the 6800 and X800 difference is much bigger when playing at higher resolutions. And wiht AA and AF on. Plus there have been no major advancements in 3.0 yet. So by the time there is, the games will be to demanding for the 6800 or X800 to play.

 

[nRollo]

Originally posted by: Noob
It has been said by many before. 3.0 doesn't make an image quality difference compared to 2.0.

That is true, but there is an image quality difference between the PS 1.1 you'll have to play SC:CT at and the PS3 6800 owners will have.

There's an IQ difference between the HDR mode of Far Cry and what you see.

There's an IQ difference between what you get to see in Riddick, and what I would see.

Plus the 6800 can't even handle these options very well anyway.
Look at COR.

Yes let's look at COR.
The truth, not Noob-speak
Yeah 57fps on a 6800U at 16X12 isn't "handling" it. :roll: Since I have a 6800NU SLI set that performs ~ 6800U level, I would be able to handle that setting as well, or a lower res with some AA/AF.

Plus there have been no major advancements in 3.0 yet. So by the time there is, the games will be to demanding for the 6800 or X800 to play.

Err, you keep telling yourself that!

 

[apoppin]

Originally posted by: Noob
It has been said by many before. 3.0 doesn't make an image quality difference compared to 2.0. How many times does one have to say it? Plus the 6800 can't even handle these options very well anyway. Look at COR. And the FPS between the 6800 and X800 difference is much bigger when playing at higher resolutions. And wiht AA and AF on. Plus there have been no major advancements in 3.0 yet. So by the time there is, the games will be to demanding for the 6800 or X800 to play.

Here please read this . . . . then you'll know what SM 3.0 is really all about and you'll see what you're missing

it rally is a very good explanation and not technical nor does it talk down :thumbsup:

 

[imported_Noob]

Originally posted by: apoppin

Originally posted by: Noob
It has been said by many before. 3.0 doesn't make an image quality difference compared to 2.0. How many times does one have to say it? Plus the 6800 can't even handle these options very well anyway. Look at COR. And the FPS between the 6800 and X800 difference is much bigger when playing at higher resolutions. And wiht AA and AF on. Plus there have been no major advancements in 3.0 yet. So by the time there is, the games will be to demanding for the 6800 or X800 to play.

Here please read this . . . . then you'll know what SM 3.0 is really all about and you'll see what you're missing

it rally is a very good explanation and not technical nor does it talk down :thumbsup:

All I see is HDR. But that makes a little bit of a difference. The only game that they said they say SM 3.0 on is Far Cry. And they mentioned it didn't make an image quality difference. Nor has it made a difference on any other game out there. So you may get a little better motion blur effects. I wouldn't sacrifice that for the performannce and image quality fo the X800. So as I am trying to say. Sm 3.0 hasn't made an image quality difference. I am not saying nor have I said it won't make an image qaulity difference in the future. But when that time comes, this gen of cards won't be able to play the games anyway because they will be to powerful for our 12/16 piper cards to handle. The next gen of cards coming out in a few months are expected to have 32 pipes and 512mb ram. So you can get the game developers will take advantage of that power, even if todays gen cards can't handle it. But don't get me wrong, I fully understand why people would buy a 6800. As I said before, if I had to choose between the 2 cards when they just came out. I would have gone with the 6800 in the snap of a finger.

Thanx for the link though. It explained the difference between PS and VS.

 

[McArra]

you should see Farcry with HDR in motion and Sprintel Cell 3 demo. You wouldn't say that it doesn't make a difference, you would say it makes a big difference believe me.

 

[imported_Noob]

Originally posted by: McArra
you should see Farcry with HDR in motion and Sprintel Cell 3 demo. You wouldn't say that it doesn't make a difference, you would say it makes a big difference believe me.

I definitely don't deny it wouldn't make a differenc ein SP3 because it's 3.0 compared to 1.1. But I don't know about Far Cry. I would have to see for myself. I am jsut taking the word of some reviews about it.

 

[apoppin]

Originally posted by: Noob

Originally posted by: apoppin

Originally posted by: Noob
It has been said by many before. 3.0 doesn't make an image quality difference compared to 2.0. How many times does one have to say it? Plus the 6800 can't even handle these options very well anyway. Look at COR. And the FPS between the 6800 and X800 difference is much bigger when playing at higher resolutions. And wiht AA and AF on. Plus there have been no major advancements in 3.0 yet. So by the time there is, the games will be to demanding for the 6800 or X800 to play.

Here please read this . . . . then you'll know what SM 3.0 is really all about and you'll see what you're missing

it rally is a very good explanation and not technical nor does it talk down :thumbsup:

All I see is HDR. But that makes a little bit of a difference. The only game that they said they say SM 3.0 on is Far Cry. And they mentioned it didn't make an image quality difference. Nor has it made a difference on any other game out there. So you may get a little better motion blur effects. I wouldn't sacrifice that for the performannce and image quality fo the X800. So as I am trying to say. Sm 3.0 hasn't made an image quality difference. I am not saying nor have I said it won't make an image qaulity difference in the future. But when that time comes, this gen of cards won't be able to play the games anyway because they will be to powerful for our 12/16 piper cards to handle. The next gen of cards coming out in a few months are expected to have 32 pipes and 512mb ram. So you can get the game developers will take advantage of that power, even if todays gen cards can't handle it. But don't get me wrong, I fully understand why people would buy a 6800. As I said before, if I had to choose between the 2 cards when they just came out. I would have gone with the 6800 in the snap of a finger.

Thanx for the link though. It explained the difference between PS and VS.

You are welcome for that link . . . . i read many many papers and reviews over the last day or so and this is the "best" (most balanced explanations) i can find.

Please not that this article is pretty old - a year - so don't expect it to be up-to-date on games . . . it's just that the explanations are clear and concise.

they do say:

As far as Pixel Shader 3.0 goes, the most striking aspect is the minimum instruction count, which jumps way up from the 96 instructions specified in the 2.0 Extended spec all the way up to 512 instructions. How fast a shader with that many instructions would actually run is open to debate (not to mention how powerful the hardware is), but this increase certainly adds to the flexibility on offer.
. . .
Away from the simple instruction count, we come to what could be considered the real meat of what Pixel Shader 3.0 offers. To be honest, from a gamer's point of view these extra features aren't included to add to image quality, but rather to make the lives of developers easier, and give them the greater flexibility and power that they yearn
. . .

[FC] may well have been recompiled to take advantage of some of the features listed above, and although it is unlikely to make a difference to image quality at all, we may see performance improvements using PS 3.0 over 2.0

. . .

The feature set of Pixel Shader 3.0 does also allow for some nice new visual effects of note - Mainly the ability to use floating point values for texture filtering and frame buffer blending. In particular, frame buffer blending can allow for a realistic looking 'motion blur' effect (brings back memories of 3Dfx's T-Buffer, doesn't it?), and floating point textures should improve the look of effects such as HDR (High Dynamic Range) rendering - An effect that we've already seen on PS 2.0 hardware in Half-Life 2, now taken to a new level.

. . . . This leaves us with possibly the biggest single leap that the 3.0 shader model offers us over its predecessors:

Vertex textures

The premise of vertex textures is actually quite simple. In the past, vertex shaders have had no way of accessing textures. This has now changed, as a function has been added to allow vertex shaders to do texture lookups. This greatly enhances the number of things you can do with vertex shaders, and once used may well end up with some very nice effects for users.

One potential usage for this feature is the ability to do full displacement mapping. Displacement mapping was a highly-touted feature of DirectX 9, so it was somewhat disappointing when the limitations of current hardware when it came to this feature were established. Currently available hardware (Matrox Parhelia aside) can only use pre-created displacement maps, whereas this new functionality should assist in creating fully programmable displacement maps.

Another great bonus of being able to read textures from the vertex shader will be the ability to create more realistic physics simulations. Being able to use textures and vertex shaders in this new way will be particularly useful for simulating objects like water or cloth in a more realistic manner.

. . .
What does this mean for gamers?

As we've explored, the potential for 'cool eye candy' is generally greater with the improvements made in Vertex Shader 3.0 than those seen in Pixel Shader 3.0. Certainly, if some of these features are put into effect in future titles, and performance is up to scratch, then we have something to be excited about.

i looks pretty good to me . .. . and it's now a year after the article and we ARE seeing new games using SM 3.0 . . . . ati had jumped on the bandwagon and has stopped talking negatively about it.

 

[dev0lution]

I'll take 16 pipelines on a X800 XL that I can use today over 8 in the 6600 GT for the $150 price difference. By the time stuff that takes advantage of SM 3.0 comes out I'll probably need a better card anyhow. Plus I don't want to pay the higher price for the 6800 GT that will still leave me hoping that I got a newer card with the hardware decoding for HD enabled. To each his own...