Stories
Slash Boxes
Comments

News for nerds, stuff that matters

Slashdot Log In

Log In

Create Account  |  Retrieve Password

NVIDIA Doubts Ray Tracing Is the Future of Games

Posted by kdawson on Fri Mar 07, 2008 12:24 PM
from the pick-your-horse dept.
SizeWise writes "After Intel's prominent work in ray tracing in the both the desktop and mobile spaces, many gamers might be thinking that the move to ray-tracing engines is inevitable. NVIDIA's Chief Scientist, Dr. David Kirk, thinks otherwise as revealed in this interview on rasterization and ray tracing. Kirk counters many of Intel's claims of ray tracing's superiority, such as the inherent benefit to polygon complexity, while pointing out areas where ray-tracing engines would falter, such as basic antialiasing. The interview concludes with discussions on mixing the two rendering technologies and whether NVIDIA hardware can efficiently handle ray tracing calculations as well."
+ -
story

Related Stories

[+] Ray Tracing for Gaming Explored 266 comments
Vigile brings us a follow-up to a discussion we had recently about efforts to make ray tracing a reality for video games. Daniel Pohl, a research scientist at Intel, takes us through the nuts and bolts of how ray tracing works, and he talks about how games such as Portal can benefit from this technology. Pohl also touches on the difficulty in mixing ray tracing with current methods of rendering. Quoting: "How will ray tracing for games hit the market? Many people expect it to be a smooth transition - raster only to raster plus ray tracing combined, transitioning to completely ray traced eventually. They think that in the early stages, most of the image would be still rasterized and ray tracing would be used sparingly, only in some small areas such as on a reflecting sphere. It is a nice thought and reflects what has happened so far in the development of graphics cards. The only problem is: Technically it makes no sense."
[+] Games: Carmack Speaks On Ray Tracing, Future id Engines 256 comments
Vigile writes "As a matter of principle, when legendary game programmer John Carmack speaks, the entire industry listens. In a recent interview he comments on a multitude of topics starting with information about Intel, their ray tracing research and upcoming Larrabee GPU. Carmack seems to think that Intel's direction using traditional ray tracing methods is not going to work and instead theorizes that using ray casting to traverse a new data structure he is developing is the best course of action. The 'sparse voxel octree' that Carmack discusses would allow for 'unique geometry down to the equivalent of the texel across everything.' He goes on to discuss other topics like the hardware necessary to efficiently process his new data structure, translation to consoles, multi-GPU PC gaming and even the world of hardware physics."
[+] Games: Students Evaluate Ray Tracing From Developers' Side 84 comments
Vigile writes "Much has been said about ray tracing for gaming in recent weeks: luminaries like John Carmack, Cevat Yerli and NVIDIA's David Kirk have already placed their flags in the ground but what about developers that have actually worked on fully ray traced games? PC Perspective discusses the benefits and problems in art creation, programming and design on a ray traced game engine with a group of students working on two separate projects. These are not AAA-class titles but they do offer some great insights for anyone considering the ray tracing and rasterization debate."
This discussion has been archived. No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
 Full
 Abbreviated
 Hidden
More
Loading... please wait.
  • by peipas (809350) on Friday March 07 2008, @12:27PM (#22677248)
    Kirk should talk to Picard [demoscene.hu] who is quite enthused about real time raytracing.
    • by moderatorrater (1095745) on Friday March 07 2008, @12:39PM (#22677428)
      That lends a lot of weight to the raytracing argument, since I generally prefer Picard over Kirk...
    • Re: (Score:3, Insightful)

      As a counter-counterpoint, the article has quite misleading pictures-

      The Ray-Tracing images are super slick, but are non real-time, highly processed work.

      Whereas the comparison Rasterized images are real-time, game-generated examples. If you were to allow the pro-rasterization side the same time to produce a single picture, it would be super fancy.
      • Re:Counterpoint (Score:5, Interesting)

        by Applekid (993327) on Friday March 07 2008, @01:46PM (#22678504)
        We're not talking about the current technology, we're talking about the future. As in whether Ray Tracing is the Future of Games.

        Graphics hardware has evolved into huge parallel general-purpose stream processors capable of obscene numbers of FLOPs per second... yet we're still clinging tenaciously to the old safety blanket of a mesh of tesselated triangles and projecting textures onto them.

        And it makes sense: the industry is really, really good at pushing them around. Sort of how like internal combustion engines are pretty much the only game in town until alternative save themselves from the vapor.

        Nvidia, either by being wise or shortsighted, is discounting ray-tracing. ATI is trailing right now so they'd probably do well to hedge their bets on the polar opposite of where Nvidia is going.

        3D modelling starts out in abstractions anyway with deformations and curves and all sorts of things that are relatively easy to describe with pure mathematics. Converting it all to mesh approximations of what was sculpted was, and still is, pretty much just a hack to get things to run at acceptable real-time speeds.
        • Re:Counterpoint (Score:4, Interesting)

          by SanityInAnarchy (655584) <ninja@slaphack.com> on Friday March 07 2008, @03:56PM (#22680638) Journal
          From what I read, I think part of this is nVidia assuming GPUs still stay relevant. If we ever do get to a point where raytracing -- done on a CPU -- beats out rasterization, done on a GPU, then nVidia's business model falls apart, whereas Intel suddenly becomes much more relevant (as their GPUs tend to suck).

          Personally, I would much prefer Intel winning this. It's a hit or miss, but Intel often does provide open specs and/or open drivers. nVidia provides drivers that mostly work, except when they don't, and then you're on your own...
        • yet we're still clinging tenaciously to the old safety blanket of a mesh of tesselated triangles and projecting textures onto them.

          Tessellation is also frequently used in ray-tracers as it makes things much simpler and faster.

          Converting it all to mesh approximations of what was sculpted was, and still is, pretty much just a hack to get things to run at acceptable real-time speeds.

          It also makes things much simpler for ray-tracers. Really. Intersecting a line with an arbitrarily curved surface is de

        • Re:Counterpoint (Score:4, Informative)

          by Pseudonym (62607) <ajb@spamc o p . n et> on Saturday March 08 2008, @04:23AM (#22685674)

          Nvidia, either by being wise or shortsighted, is discounting ray-tracing.

          What you may not realise is that NVIDIA sells a renderer/raytracer which uses the GPU for accelleration, targeted at the animation and VFX market.

          They are not discounting ray-tracing. They are embracing it. And they know, from lots of their own R&D, that it's not going to be competitive in the real-time market at any point in the forseeable future.

  • by OrangeTide (124937) on Friday March 07 2008, @12:28PM (#22677266) Homepage Journal
    Saying something sucks if he's already developing a product for it.
  • by DeKO (671377) <danielosmariNO@SPAMgmail.com> on Friday March 07 2008, @12:30PM (#22677292)

    A good way to mix both techniques is Relief Texture Mapping [ufrgs.br]. It's a good way to get smooth surfaces thanks to the texture interpolation hardware, with no extra polygons.

    • Except it can't do jack-squat at the edges of geometry...
        • by Creepy (93888) on Friday March 07 2008, @05:35PM (#22681950) Journal
          It can do curved surfaces, but is far from perfect and really is an approximation placed into a polygon structure (the surface itself is flat, but you don't see the surface itself). The silhouettes are done using 4 of the 16 quadrics as a reasonable subset to get an estimate curvature (4 to save on storage). These are pre-calculated and therefore have some overhead (either in a file or dynamically calculated on load). It is also restricted to hard shadows because it works similarly to ray tracing - tracing lines, not patches (it's hard to describe that, but think of it like when light hits an object, it reflects a line, not a cone), however, you get these shadows for free.

          Incidentally, Steep Parallax Mapping and Interval Mapping also use pseudo ray tracers, but the surface curvature is unique to Relief Mapping (the quadric technique doesn't work with them). Most other techniques can be adapted to support soft shadows, however, and most game developers I know (3 professional) think soft shadows are more important than curved surface relief maps. If surface curvature is important, they'd rather tessellate it out.
      • Re: (Score:3, Interesting)

        Ray tracing is firing rays from each screen pixel, test for collision against the geometry, and figure out the proper color. RTM is firing rays from each polygon's "pixel", test for collision against the "texture geometry", and figure out the proper color. It's just ray tracing in a subset of the screen pixels, in a geometry (heigthmap) represented by a texture (or multiple textures) from a polygonal face. Why do you think this is not related to ray tracing?

  • by Btarlinian (922732) <bjcbell&gmail,com> on Friday March 07 2008, @12:30PM (#22677302)
    Seriously though, does anyone expect Nvidia to say, "Yes, we really do think that our products will all be obsolete and outdated in a few years. Thank you for asking." I personally have no idea as to whether or not ray tracing is the future of games, but I really don't think that Nvidia is the right person to ask either, (just as Intel isn't).
    • but I really don't think that Nvidia is the right person to ask either, (just as Intel isn't).

      True. I believed Intel, when they explained the superiority of Ray-Tracing, and now I believe Nvidia, when they say the opposite.

      From what I can tell, Ray-Tracing is closer to 'reality', and so you'd expect the technology eventually to tend in that direction. But the explanation from the Nvidia dude makes it seem like that point is many years away, owing to the excellent results available now with rasterization, a

      • Re: (Score:3, Insightful)

        From what I can tell, Ray-Tracing is closer to 'reality', and so you'd expect the technology eventually to tend in that direction.

        Not necessarily, humans aren't that picky that the virtual reality be a perfect reality. For example, I don't ever expect to pick up a laser, find a thin enough slit and see the quantum effects on the wall, not the real simulated deal anyway. More natural than modelling it as a pefect beam? Sure, but like what's the point. Graphic cards and Copperfield are in the same business, making the grandest possible illusion with a minimum of resources and maximum immersion. If you get the macroscopic effects close

        • I don't ever expect to pick up a laser, find a thin enough slit and see the quantum effects on the wall, not the real simulated deal anyway.
          I'm assuming you mean within a simulated 'reality'.

          On the one hand, quantum interference can easily be exempted from the potential ray-tracing future, should it prove hard to model. On the other, what's so hard about it?
    • Seriously though, does anyone expect Nvidia to say, "Yes, we really do think that our products will all be obsolete and outdated in a few years. Thank you for asking." I personally have no idea as to whether or not ray tracing is the future of games, but I really don't think that Nvidia is the right person to ask either, (just as Intel isn't).

      One could argue that the Nvidia folks have been well aware of ray-tracing for a long time, and if they thought it was reaching the point where it was going to be usef

      • My understanding is that part of the threat to Nvidia and other dedicated graphics card makers is that ray tracing doesn't lend itself as well to dedicated solutions. Or rather, the type of processor needed tends to be the type that is already being used as a CPU with some minor tweaks to optimize performance. So instead of buying a separate chip for graphics, you get the same performance boost from just getting a second CPU or one with more cores. Instead of a graphics card with more RAM, you just add m
        • by Znork (31774) on Friday March 07 2008, @01:16PM (#22678010)
          So instead of buying a separate chip for graphics, you get the same performance boost from just getting a second CPU or one with more cores.

          Not only that; you get a performance boost from your server. And from your kids computers. You get a performance boost from your HTPC. And potentially any other computer on your network.

          The highly paralellizable nature is the most interesting aspect to raytracing IMO; with distributed engines one could do some very cool things.
          • Re: (Score:3, Insightful)

            If I try to offload to my HTPC something tells me that network latency is sure gonna send my FPS to crap. But, in the current world where raytracing takes a LONG time, sure you can offload all you want.
            • Re: (Score:3, Interesting)

              You should check out the quake3 engien modded to use raytracing. Quite nice, and the computer scientist also has some interesting points about the efficiencies of raytracing (ie those problems Portal had with it's portal rendering (which you didn't see, because they had to hack around it))? Not a problem with raytracing).

              So, quake3 runs easily using raytracing, and that was just something to show how fast it can be...it's feasable in the latest games too. So why is raytacing so slow again?

              Oh, you're thinkin
        • by Abcd1234 (188840) on Friday March 07 2008, @01:35PM (#22678338) Homepage
          Actually, I don't think that's true at all. Raytracing, just like today's rasterizers, can greatly benefit from dedicated hardware for doing vector operations, geometry manipulation, and so forth. This is particularly true as raytracing benefits greatly from parallelization, and it would be far easier to build a dedicated card with a nice fat bus for shunting geometry and texture information between a large number of processing units than it would be to use a stock, general multicore processor which isn't really designed with those specific applications in mind.

          Besides, the whole reason to have separate, specialized gear for doing things like audio/visual processing is to free up the main CPU for doing other things. Heck, we're even seeing specialized, third-party hardware for doing things like physics and AI calculations, not to mention accelerators for H.264 decoding, etc. As such, I see no reason to move graphics rendering back to the main CPU(s).
    • When it comes to ray-tracing complex shapes using spline patches, the recommended approach is to tessellate the geometry into triangles and then ray-trace the collection of triangles, using octrees for optimization (just test a single cube for ray-intersection than a whole set of triangles).

      Graphics cards already do something similar with deferred rendering. They sort the projected triangles according to position on the screen, and render groups of triangles into a local cached copy of the framebuffer. Onl
      • although there are programming techniques to do that with graphics cards too

        Whoops, I think you meant "hacks", there.

        This is the same thing that's been going on with rasterization for years. Developers and hardware designers have built hack upon hack in order to implement what raytracing does so easily. The result is a library of tricks that often can't be easily intermixed, and are always a pain to work with.

        So, if you can switch to raytracing and get similar performance, and end up with a larger feature
    • Just because their products now are focused on rasterization (their current GPUs can do raytracing as well) doesn't mean their next generation ones have to be. I'm sure they'd be happy to produce raytracing hardware, if there was a demand for it and if they could make it fast.

      That is the problem, as the article noted. You get research oriented things that are very pie in the sky about rendering techniques. They concentrate on what is theoretically possible and so on. nVidia isn't in that position, they are
      • Just because their products now are focused on rasterization...
        Ray tracing is a form of rasterization... In other words, it translates a description of a scene into an array of pixels.
  • by $RANDOMLUSER (804576) on Friday March 07 2008, @12:37PM (#22677400)
    Intel says to do it with the CPU, and nVidia says to do it with the GPU. What a surprise.
  • by Itninja (937614) on Friday March 07 2008, @12:39PM (#22677424) Homepage
    Personally, I prefer spites to either ray-trace or polygons. I still thing Starcraft (the game, not the conversion van) had some of the best graphics. But then I am kind of a fuddy-duddy. I also think River Raid was an awesome game.
    • Agreed. Doing anything out of spite is a bad idea. Long live Space Invaders!
    • I still remember, as boy in the early 80's, getting the opportunity to take a cruse on a navy ship. Seeing the targeting equipment on that ship gave me a real appreciation for just how realistic the graphics on Missile Command were. They were darn near indistinguishable from the real thing.
    • Very good point! It does maybe not have that super-realistic "I shot the guy in the head and he's actually BLEEDING from the head", but it just looks... nice! The same with the old black isle games, such as Baldur's gate (especially II), fallout, planespace torment, etc... Clean, simple and nice.

      And, by the way, I thought I'd laugh my ass off first time I went on a camping trip and saw the Starcraft conversion van. :)
    • Re: (Score:3, Informative)

      Talking about sprites, did you see the teaser video for King of Fighters XII? So. Fuckin'. Beautiful. [kotaku.com]
    • If you were talking about pixel detail, I might agree.

      But It's all about the lighting and animation in a changing environment (starcraft was isometric with static lighting). It's hard to light and animate sprites convincingly, unless you have a lot of artists willing to draw a lot of frames.
    • It takes a lot of work to get a 3D model to look as good as a 2D sprite. You gain more freedom and, as the amount of actions increase, can create new animations with as lot less hassle. But it remains very difficult to get a really good "animated" feel with 3D models which need to look good from all angles, and nowadays under all lighting conditions. 2D sprites, while laborious to create, invariably display precisely as the animator intended.

      Games like Ratchet and Clank or Jak and Daxter pull this off well.
      • It goes both ways. Sprites are perfect at a small scale or when you don't need too many frames of animation, or when you want to blend and layer animations. However, a super smooth screen-size sprite (a boss, for example) can easily suck up more resources than a model if you're dishing out lots of animation.
  • IBM doubts the future of the "personal computer"
    Buggy manufacturers poo-poo the new horseless carriage
    etc, etc.
  • We can't do it as well as Intel yet, therefore it sucks. BUY NVIDIA.
  • by 9mm Censor (705379) on Friday March 07 2008, @12:46PM (#22677526) Homepage
    What about game and engine devs? Where do they see the future going?
    • Re: (Score:3, Informative)

      What about game and engine devs? Where do they see the future going?

      IAAGD. My current paid project is to have both a raytracing module and a rasterizing module, and is designed to use them completely interchangeably. Personally, I'm a much bigger fan of raytracing than rasterization, and I'm going to a great deal of effort to make sure that it can be done efficiently with my engine.

      -:sigma.SB

    • by Squapper (787068) on Friday March 07 2008, @12:57PM (#22677700)
      I am an game-developing 3d-artist, and our biggest problem right now is the lack of ability to render complex per-pixel shaders (particulary on the PS3). But this is only a short-term problem, and what would we do with the ability to create more complex shaders? Fake the appearance of the benefits of Ray-tracing (more complex scenes and more realistic surfaces) of course!

      On the other hand, the film industry could be a good place to look for the future technology of computer games. And as it is right now, it's preferable to avoid the slowness of raytracing and fake the effects instead, even when making big blockbuster movies.
    • I am an amateur game developer, so I probably can't speak for large devs, but I can speak for the community. The biggest problem facing game devs (and graphics people in general) is lighting. Doom 3 created an elegant, unified lighting model (i.e. the environment and characters are lit with the same algorithm, making them consistent), but it had severe limitations. Gone were the days where lights could be soft, and bounce around, and generally look nice, and replacing it was a very harsh lighting system tha

  • by Bones3D_mac (324952) on Friday March 07 2008, @12:48PM (#22677550)
    For the most part, I really don't see ray-tracing adding much to the world of gaming that isn't being handled well enough by current methods. Unless someone was specifically creating games that somehow directly incorporated either the benefits or the added calculations involved with ray-tracing itself, it would only be a costly, and highly inefficient gimmick of an alternative to current techniques.

    Sure, ray-tracing has its place in a lot of areas, but real-time gaming would be a terrible misuse of processing horsepower... especially when you could be applying it to other areas of gaming that actually affect gameplay itself. For example, how about more robust AIs for in game elements, or high-end physics processing that can combine things like fabric/hair/ fluid/fire physics processing with the ability to decimate objects completely as vector-calculated chunks based on the surrounding environments, rather than all this predetermined destruction we currently see in games. (Example, a surface could be eroded incrimentally by having a fluid running acrossed it until a hole forms in the shape of the fluid path...)
  • Hardly anything new (Score:4, Interesting)

    by K. S. Kyosuke (729550) on Friday March 07 2008, @12:58PM (#22677714)

    For years, the movie studios were using Pixar PRMan, which is in many ways a high-quality software equivalent of a modern graphics card. It takes a huge and complex scene, divides it into screen-space buckets of equal size, sorts the geometrical primitives in some way, and then (pay attention now!) tesselates and dices the primitives into micropolygons about one pixel each in size (OpenGL fragments, anyone?), shades and deforms them them using simple (or not-so-simple) algorithms written in the RenderMan shading language (hey, vertex and pixel shaders!) and then, it rasterizes them using stochastic algorithms that allow for high quality motion blur, depth-of-field and antialiasing.

    Now the latest part is slightly easier for a raytracer, which can cast a ray from any place in the scene - of course, it needs this functionality anyway. But a raytracer also needs random access to the scene which means that you need to keep the whole scene in memory at all times, along with spatial indexing structures. The REYES algorithms of PRMan needs no such thing (it easily handles 2 GB of scene data on a 512 MB machine along with gigs of textures), and it allows for coherent memory access patterns and coherent computation in general (there is of course some research into coherency in raytracing, but IIRC, the results were never that good). This is a big win for graphics cards, as the bandwidth of graphics card RAM has never been spectacular - it's basically the same case as with general-purpose CPU of your computer and its main memory. But with 128 or so execution units of modern graphics card, the problem is even more apparent.

    Unless the Intel engineeers stumbled upon some spectacular breakthrough, I fail to see how raytracing is supposed to have any advantage in a modern graphics card. And if I had to choose between vivid forest imagery with millions of leaves flapping in the wind and reflective balls on a checkered floor, I know what I would choose.

  • Radiosity does more for indoor scene quality than does raytracing. Radiosity gives you the visual cue of a dark band at an inside corner, which is subtle and a basic part of the human visual mechanism for resolving depth. Raytracing makes shiny things look cool.

    Oh, right, this is for gamers.

  • They will cut back our "first contact" date for at least 500 years.
  • A matter of speed (Score:4, Insightful)

    by CopaceticOpus (965603) on Friday March 07 2008, @02:51PM (#22679616)
    "C is much too slow, to get good performance you must use assembly."

    "Scripted languages are much too slow, to get good performance you must use compiled languages."

    As computers get faster, there is always a move from technologies that are easier for the computer to technologies that are easier for the developer. Since ray tracing involves less hacks and is a more direct model of the effects developers want to create, it seems inevitable.
      • Re:A matter of speed (Score:4, Informative)

        by batkiwi (137781) on Friday March 07 2008, @05:27PM (#22681838)
        Current methods do not scale parallel. SLI does not give you 2x the resolution at the same framerate, or double the framerate at the same resolution. Same for any sort of quad SLI solution (two dual chip gfx cards, etc).

        Ray tracing scales almost perfectly. The same processor with 4 cores will perform about 3.9x as well (pushed pixles per second, so either higher FPS or higher resolution, or a mix) as one with 1 core. This is why intel is pushing this.
  • There used to be an interesting debate between Professer Philipp Slusallek of the University of Saarbruecken and chief scientist David Kirk of nVidia at GameStar.de. The original article has been taken down, but I found a slightly mangled version on the Wayback machine and I've cleaned it up a bit and put it up on my not-a-blog: link [scarydevil.com].

    I'd appreciate a better translation of the German part of the text.
    • Following up on my previous post about the debate between David Kirk and Philipp Slusallek in 2006 (link [slashdot.org] ... apologies to Dr. Slusallek, Slashdot truncated his name).

      According to Dr. Slusallek's LinkedIn profile [linkedin.com] he's currently a "Visiting Professor at NVIDIA".

      The performance of Dr. Slusallek's real-time raytracing engine at only 90 MHz was quite impressive: "In contrast we have recently implemented a prototype of a custom ray tracing based graphics card using a single Xilinx FPGA chip. The first results sho