Added: 4 years ago
From: lsnderick
Views: 21,803
Sort by time | Sort by thread (beta)

Link to this comment:

Share to:

All Comments (35)

Sign In or Sign Up now to post a comment!
  • Will that work with a Radeon 9200 High Performance Graphic Card?

  • SEGA's Model 4 ray tracing capable hardware is currently in development

    Further informtaion is to be found at the following address -

    fgnonlinedotwebsdotcom

  • Can't 'til this is available for games! :)

  • 0:59 its like " HEY dont leave me alone in the room u stupid ball of light ! " lol

  • @ezuan88 really? is it really like that?

  • looks like there is many lights in one place trying to imitate area lighting

  • I think the same.

  • which kit were you using to do the real time ray tracing?

  • great work

  • nature won

  • site last update 2006? where are they now? the 'Still sucking' demo said this was all in software(but was pre-rendered :*( )...put this engine on a GPU, would run better...

    With a p4 3.2Ghz HT I get less than 10 fps (sometimes MUCH less)

  • why..?

  • The demo I downloaded looked nothing like this at all. And it ran at a tenth of the Framerate.

    (I have an extremely good computer) Everything was low resolution (Shadows, light Et cetera)

    Maybe it only works on ATI cards?

  • how good is "good" ?

  • Intel e6850 (Dual-core 3.0 Ghz), Nvidia 8800 GT, 4 Gigs of memory, and Windows XP (32 Bit :( )

    Good enough? My benchmark looked more like the ScreenShots on the website.

  • I guess that is pretty good. I was just wondering since it's "raytracing" and how the work is distributed over GPU and CPU.

  • @Zalo10 That program uses a way to calculate a 3D scene that is very different to the way the 3D card does: the 3D graphics card can not do Raytracing.

  • @EskyHunter They can now because of Nvidia CUDA and OpenCL.

  • @Argoon1981 Ah, yes, but the application must be made for using that.

  • @Zalo10 I know your comment is 2 years old but i thought i should explain why the demo looked like that for you, this demo is 5 years old and based on raytracing so runs on CPU's not GPU's nor ATI/Nvidia cards render this demo, because of that and dual core CPU's were not the norm this demo is single core only, this made the developers make the demo run at fixed resolution of 640x480 so they could get real time frame rates on the single core CPU's of that time.

  • the nail in the coffin will require lifelike human animation. that seems a ways off.

  • F.A.N. guys, where are they now?

  • This is the final "nail in the coffin" of video game realism... Very nice...

    The Source Engine has this... but not in real-time... that's why it has some really photo-relistic STATIC scenes(from certain angles), such as the final shot from the Stress Test.

  • i agree it is truly the final nail. hybrid rendering should be the primary solution with raster/RT/radiosity

  • I was wondering why you can turn on ray-trace rendering in valve hammer. But is just crashes my computer.

  • this method of RT should appraoch games soon in 2010-12

  • While I would dearly love to see raytracing in games that soon.. it's still likely to be limited to tech demo's. There won't be a shift to raytracing in game engines until they can do it at a decent framerate with all the bells and whistles turned on (caustics, sub-surface scattering etc). Nobody is going to want to release a game engine with very basic raytracing in it that makes their game look less advanced.I'd say it's more like 2012 to be realistic.

  • I've been doing progressive research on RT/RASTER in games notably nvidia's Cuda advances and so far David Kirk Nvidia's scientist even John Carmack owner of ID games have concluded that a combination of RT and Raster gfx will be the likely solution in 2011 -2012 since the next gen consoles supposedly will debut,XBOX720 and PS4. RT alone won't be correct since too shiny means a blinding solution. So i agree with you SK1ppr 2012 will be the best timing.

  • Too shiny?

    THats the most stuped explanation ever.

    If a reflection is too shiny then turn down specualary, lol. Its as simples as that.

  • Specularity and reflection are not the same thing but you are correct in your assumption that the values should simply be reduced.

  • yes they are the same thing.

    But because non glossy reflection requires extra samples and thus is much heavier on the cpu they faked it by using 'specular' highlights.

    But if you want to have correct highlights, they are caused by blurry reflections of a lightsource. Thats how modern raytracers calculate them too btw, because now a day computers are fast enough to compute them correctly

  • they are the same thing.

    Specular reflections are reflections of bright lightsources.

    But because in early renderers (rasterizers) calculating raytraced reflections was too heavy they faked it by using faked highlights and called those 'specular highlights'.

    If your raytracer can handle glossy reflections then there is no more need to add fake specular highlights because you'll get them automaticly from the reflections (this is the physicall corect way to do it).

  • @Capeau that is, if the lightsource has a volume so the reflection can see the lightsource in a reversed trace. (Surface to Light, instaid of the usual light to surface)

  • @sgstino

    every lightsource has volume in reality.

    CG pointlight arent realistic to begin with and are just cheap tricks from the 70's

Loading...
Alert icon
0 / 00Unsaved Playlist Return to active list
    1. Your queue is empty. Add videos to your queue using this button:
      or sign in to load a different list.
    Loading...Loading...Saving...
    • Clear all videos from this list
    • Learn more