<b>I've only been gone three days and I come back to find 264 new messages in the Graphics section and most of them are in this thread! I've only quickly read the posts, so sorry if i've missed anything or am going over anything thats already been covered.</b>
the hardwired T&L in Geforce and Radeon cards will not be included in games for long.
with all new cards comming out with vertex shaders hardwired T&L will be fazed out for the simple reason that is is hardwired, at least CPU's are programable and as they get faster they can always be used for things like vertex shaders, thats something that can't be said for hardwired HW T&L.
It will be good for another year and a half at least. And after that, a good secondary system, The Programmable stuff will almost certainly be primary. Around this time I see games with no Software Geometry Support.
No games can really not work on a non HW T&L card, if anyone was stupid enough to make a game that didn't allow SW T&L then a simple driver trick like geometry assist could be added and the CPU could use the games HW T&L engine and it would probably be faster then using a SW T&L engine
Do you really think it would be stupid to drop software geometry? All the developers decided to drop software renderers when hardware 3D cards started to emerge. They decided to drop all the not so good 3D APIs and develop exclusively for the 3DFX GLIDE API. Do you really think they'll hesitate too much before dropping that.
From what Carmack's been saying, it sounds like Doom 3 will NOT support Software Geometry. Instead the CPU will be very busy with the complex Physics and AI.
<b>What is Geometry Assist?</b>
anyway to me it makes allot more sense to allow for sw T&L even if your game runs slowly because by the time these games come out people will have far faster CPU's, I'm sure in a year CPU's will be just as fast (if not faster) then the Geforce 2 HW T&L unit
It will be a good 5 yrs before any CPU is released capable of matching the raw processing power of the GeForce 2 GTS GPU.
also this is all forgetting that at any decent resolution HW T&L goes out of the window, look at any benchmark and you'll see that a HW T&L card will beat a SW T&L card at 640x480 or 800x600 when neither card is fillrate or mem bandwidth limited, but who plays at those resolution?, once the game goes to 1024x768 the benchmarks level out, go any higher and its the card with the highest fillrate and memory bandwidth (or most efficient fillrate and memory bandwidth) thats going to come out on top and in these cases the HW T&L unit can actually slow the card down by using up precious bandwidth, or use FSAA and again the HW T&L card won't nesassarily win in a HW T&L optimised game unless it has the highest or most efficient fillrate and mem bandwidth.
1024x768 is still excellent with T&L. it beats the Kyro2.
the Kyro II isn't even limited to 8 layers in a single pass, the only reason its 8 layers is because thats the maximum allowed by DX8, if more were allowed the Kyro II could do it?, fancy 10 texture layers in a game?, what about 12?, 16?, the Kyro II could do those all in a single pass, while the GTS is using 8 passes, obviously this is all theoretical stuff but its worth thinking about.
I think The Kyro is limited to 8 layers in a single pass. All the Kyro review sites keep saying upto 8 layers. The Quake 3 Engine supports upto 10 layer cascading in its multi texturing. The minimum is two.
Quake 3 Multitexturing:
=======================
(passes 1 - 4: accumulate bump map)
pass 5: diffuse lighting
pass 6: base texture (with specular component)
(pass 7: specular lighting)
(pass 8: emissive lighting)
(pass 9: volumetric/atmospheric effects)
(pass 10: screen flashes)
-The ones in parenthesis are optional.-
I may be wrong but I don't think the polygons have to be reloaded in the 'traditional' Multitexturing card. From Direct3D6 up to 8 texture operation units can be cascaded together to apply multiple textures to a common primitive in a single pass (multitexturing). The results of each stage carry over to the next one, and the result of the final stage is rasterized on the polygon. This process is called "texture blending cascade".
Well Doom 3 will have no less then 6 texture layers and upto 8
Appart from the Base skin Texture, Doom 3 is only certain to have only two others: Bump Mapping and Dot Products. I'm sure there will be more.
There will be no light maps (3 of the layers in the Q3 Engine) or Shadow maps. Instead, it will sport a fully dynamic lighting system with limited raytracing. Featuring plenty of calls to the T&L!
<font color=blue>"We've been doing hacks and tricks for years, but now we'll be able to do things we've been wanting to do for a long time," Carmack said. "For instance, every light has its own highlight and every surface casts a shadow, like in the real world. Everything can behave the same now and we can apply effects for every pixel." - John Carmack</font color=blue>
No I don't think Doom3 will be using pixel shaders actually or if it is it'll only be in a small way and won't be neccesary to play the game, Carmack said he was more impressed with the Vertex shaders in DX8 so I think thats what he'll be using
I'm not sure where you got the comments, but carmack wants a beast of a graphics card. This is what he said:
<font color=blue>"DX8 tries to pretend that pixel shaders live on hardware that is a lot
more general than the reality.
Nvidia's OpenGL extensions expose things much more the way they
actually are: the existing register combiners functionality extended to
eight stages with a couple tweaks, and the texture lookup engine is
configurable to interact between textures in a list of specific ways.</font color=blue>
BTW, Doom3 will be an openGL game.
<font color=blue>"I am spending a huge amount of graphics horsepower to allow the engine to be flexible in ways that game engines have never been before. It is a little scary to drop down from the ultra-high frame rates we are used to with Q3, but I firmly believe that the power of the new engine will enable a whole new level of game content.
I am hoping that the absolute top-of-the-line system available when the game ships will be capable of running it with all features enabled and anti-aliasing on at 60hz, but even the fastest cards of today are going to have to run at fairly low resolutions to get decent frame rates. Many will choose to drop a feature or two to get some speed back, but they still won't be able to get near 60hz.
Remember, the game won't ship for a long time yet, and today's cards will seem a bit quaint by then" - John Carmack</font color=blue>
and also obviously they can be done in software so the Kyro II or MX or any card can play a game with vertex shaders thats probably the other reason he's using them
As it currently stands, Doom 3 will not contain software Geometry code. But once the game is released, there may be some development for a dreamcast port with severe limitations. The non-T&L community can hope this will also bring a Software geometry pc release.
you can make similar pixel shader style effects with multi-texturing and also every card can use this, even though most cards will be forced into lots of passes, still having to do lots of passes and at least being able to play the game is better then not being able to play at all when pixel shaders are used.
You commend the Kyro for its innovative Tilers but revert to the preference of redundancies with the wasteful Multitexturing over Pixel Shaders?
Besides, the Doom 3 engine will be scalable, allowing the features that require programmable shaders to be disabled. The penalty is much of the eyecandy will also disspear.
I'm hearing from allot of people close to the industry that Nvidia would rather eat there own crap then make a tile based renderer, the reason for this seems to be that they don't want to use 3dfx's tech to bail themselves out of the memory bandwith hole they find themsevles in
That is just rubbish. Not every one aquires technology to pull a "Microsoft" i.e. just to bury the competition and forget about them. You Aquire technology to to use it. Whats better than having that technology in your products rather than in your competition.
They aquired 40 former SGI staff with the Gigapixel technology from 3DFX. They will be busy in implementing their designs into nVidia Technology. This means Tiling just like the Kyro II cards. I don't think nVidia will be furthering the development into napalm or rampage for release products. But they will be researched into, to resolve any potential issues, and to use any good concepts and ideas from those designs.
after QDR I don't think Nvidia have anywhere to go, its either a different design or bust, they can't keep bolting on faster and faster ram.
Ahh, There will always be faster ram, providing you spend enough towards it. Come on, Every one except intel must be grateful about them speeding up the ddr development.
<b>Sorry about the length of the post</b>
<font color=red>"My name is Ozymandias, King of Kings:
Look on my works, ye Mighty, and dispair!"</font color=red>