Intel Shows Off Quake Wars, Ray Traced

An anonymous reader writes "At the Research@Intel Day 2008, Intel showed a ray-traced version of Enemy Territory: Quake Wars. Compared to the original game, a water with reflections and refractions and a physically correct glass shader were added. Also, a camera portal with up to 200 recursions to itself has been demonstrated. To show off this ongoing research in the topic of real-time ray tracing, a four-socket system with quad cores has been used that allowed rendering the enhanced visual effects in 1280x720 at 14-29 fps. Just two years before, early versions of Quake 4: Ray Traced ran only at 256x256 with 17 fps. Even though Intel's upcoming Larrabee will be primarily a rasterizer, the capabilities for also doing ray tracing on it should deliver interesting opportunities."

Comments

Voxels?

[Xanavi]

What ever happened to voxels?

Re:Voxels?

[Anonymous Coward]

Same thing that happened to pixels.

Re:Voxels?

[FelixGordon]

Still being pursued as a weird niche. I was absolutely amazed by Voxelstein 3D, http://voxelstein3d.sourceforge.net/

It's "in the spirit" of Wolfenstein, and has totally destructible environments - for example you start in a little cell and using your knife you have to chip your way through the bars to get out. It's neat to see this alternative tech continue getting explored.

Re:Voxels?

[Thuktun]

What ever happened to voxels? Apparently Crysis uses them.

Re:Voxels?

[Skrapion]

Why don't you ask Carmack? Back in 2000 he talked about how, although voxels are interesting, hardware support for polys is just too good to ignore. In fact, the hardware support is so good that even voxel engines usually go through the rather expensive process of converting their voxels into polys in real-time because they'll make up that speed penalty by leveraging the poly drawing power of todays graphics cards.

Although everybody focuses on how easy it is to get correct shadows, reflections, and refractions with a ray tracer, another interesting fact is that ray tracers don't really care how much geometry is in your scene. That means that once you get to a certain level of detail, ray tracers can become more efficient than rasterizers!

There's one big flaw, though. Ray tracers need a fast way to intersect each ray with the world geometry. This is a problem, because creating some way to index all your geometry is really expensive, especially if you have to do it in real-time. We need some clever data structures and algorithms to solve this problem.

So, how is Carmack solving it? Well, you'll be happy to hear that he's bringing back voxels!

Height maps

[tepples]

What NovaLogic called a "voxel space" in Comanche was really just a height map. I guess the reasoning is that a height map is just a run-length-encoded representation of a voxel space.

Re:Height maps

[Anonymous Coward]

I think Outcast used height maps too, but what about TerraNova?

Re:Height maps

[Anonymous Coward]

Alpha Centauri's engine was a voxel-based one called "Caviar". Largely written in self-modifying assembly. Freaky.

Re:Height maps

[jonwil]

Command & Conquer Tiberian Sun and Command & Conquer Red Alert 2 used Voxel based 3d models for various vehicles in the game.

Re:Height maps

[billcopc]

Self-modifying assembly is a long-lost art. If you have a strong stomach, a long long time ago I used to use QuickBasic as a ghetto scripting tool, loading in various assembler modules to do the dirty work. I later switched to Pascal.

Back then, most of my hardware-control loops used self-modifying bits and bobs... sometimes to save a byte, sometimes to avoid a fetch. A few times I used true self-modifying code where the outer loops would reprogram the inner loops on-the-fly. It was the most CPU-efficient way to do realtime multichannel sound synthesis on a 486, and of course it gave me the opportunity to refer to it as a dynamic synth compiler :) The bitches were all over me, dawg!

All that lovely code died a quick, silent death when Windows 95 came along. It wreaked all sorts of havoc and Windows would kill the app as soon as it tried to self-mod. It's a shame I didn't keep up with the skills, I could be one rich despicable virus writer today :)

It's times like this I miss the 90's, I still have that 386 programming manual somewhere safe.

Re:Height maps

[Mozk]

What does that test that I took in 6th grade have to do with anything? I don't believe they tested us on ray tracing and binary space partitioning back then, but I may be wrong.

Re:Height maps

[Anonymous Coward]

I miss the 90's too, when geeks ruled the internet and programming was an art.

Now children rule the internet, and programming is a dead end job.

What a shitty decade this is.

Re:Height maps

[ya really]

All that lovely code died a quick, silent death when Windows 95 came along. It wreaked all sorts of havoc and Windows would kill the app as soon as it tried to self-mod. It's a shame I didn't keep up with the skills, I could be one rich despicable virus writer today :)

I'll have to remember to pay attention more in my next asm course at school. Most college students I know seem to loathe the low level stuff and think it's not worth remembering, but I think it's still in demand in niche areas. If not, there's always virus writing to be done as you said.

Re:Height maps

[Anonymous Coward]

The engine used heightmaps to store the terrain data, and voxels to display them. This is something you might've known if you... I dunno... read the fucking article you linked to, for one.

Re:Height maps

[luther2.1k]

Ah, those were the days. I used self modifying assembly to get around the problem of combinatorial explosion when writing graphics drivers; i.e. where you have lots of states that all require an optimized inner loop with slightly different operators. Sort of the equivalent of pixel shaders, only self modifying.
    That, and a lot of other tricks of mine became redundant when the pentium and predictive branching took off and as such, the amount of assembly in my code has dropped to practically nil, save for a few things that can only be done quickly with the odd specialized instruction call.

Tim.

Re:Height maps

[Anonymous Coward]

. I used self modifying assembly to get around the problem of combinatorial explosion when writing graphics drivers; i.e. where you have lots of states that all require an optimized inner loop with slightly different operators.

This is the sort of problem functional programming excels at. Granted, I don't know of any good "low level" functional programming languages, but if you have to approach this sort of problem again, I suggest you use the common "patterns" in use in functional programming land.

Re:Height maps

[xded]

All that lovely code died a quick, silent death when Windows 95 came along. It wreaked all sorts of havoc and Windows would kill the app as soon as it tried to self-mod. It's a shame I didn't keep up with the skills, I could be one rich despicable virus writer today :) Actually, to use self modifying code in win32, you just have to mark the PE code sections as writable, since by default they're just readable and executable.

And self modifying code is still used today on some software protections, not just viruses.

Re:Height maps

[Anonymous Coward]

Embedded programming is a booming business with no new programmers coming in, because all the college graduates want to be game developers and web designers.

Re:Height maps

[Anonymous Coward]

Where are my mod points? What a beautiful observation...

Re:Height maps

[eulernet]

Self-modifying assembly is a long-lost art It was not an art. This just lead to ugly and buggy code.

The clean way to do this is to generate code in data memory, lock the data memory as executable memory, and execute the code, it will run fine in Windows (I'm using this technique right now on computer crossword filling).

Assembly language has not really disappeared.
It's just hidden under a lot of layers.
For example, you can use dynamic JIT compilers, like LUA, which generates inline assembly code, and if you know well the language and the assembly generation, you can write some pretty code.

Re:Height maps

[Anonymous Coward]

you forgot to add "get off my tubes!"

Re:Height maps

[Tyger]

Back in the 6502 days, self modifying code wasn't just a trick, it was how some things were done. For example, there were no 16 bit indexed memory access methods. The main method for indexed memory references used one of the 8 bit registers for the low 8 bits of the address, and an immediate value for the high 8 bits. To loop over more than 256 values, you had an outer loop that modified the immediate value of the instruction every 256 iterations.

Re:Height maps

[AC-x]

Not just windows fault tho I'm sure, I thought certain CPU caches also broke self modifying code (as I still remember having to turn the caches off to get A500 games to run on an 020!)

Re:Height maps

[Oktober Sunset]

YAY!! I was hoping someone would mention the amazingness that is C&C! I love voxels! *Hugs a cute lil voxel*

Re:Height maps

[somersault]

programming is a dead end job. I'm hoping that's only true if 'programming' means writing HTML (I have heard people refer to that as programming, yep!). Someone's got to write the software for all that shiny new hardware :) If you're in the right market then you'll get paid okay for it too. I started off helping out with IT support for an Engineering company, but now I'm happily developing and maintaining a few in-house apps for them. I already get paid more than I really need, so I'd expect anyone who is lead programmer for a successful application or games company probably does pretty well for themselves..

Re:Height maps

[Anonymous Coward]

God, you just made me so homesick for 1983 ... I wrote a whole text-on-hi-res graphics package for the Apple ][+ doing stuff like that ... even just the phrase "POKE -151" will bring a tear to my eye ...

Re:Height maps

[EvilIdler]

I figured out a way to get around that in my self-modifying code - just put the code chunks in front of the bulk of your static code, and right before the data chunk you use for modifications, do a jump past it.

You could also do cache flushes on a regular basis, but that could slow down things if you wanted to write some invisible, malicious code :)

Re:Height maps

[Impy the Impiuos Imp]

I'd consider a detailed knowledge of HTML to be a kind of programming. It's another language, even if it doesn't have control structures without invoking Java or whatever.

What's truly scary, though, is people claim they "know HTML" when in fact all they know how to do is design a web page using a WYSIWYG layout editor. They don't even know how to FTP it up to their hoste -- the layout editor, built in with the host's offerings, does it all automatically.

No, you can't apply for an "HTML required programming job" because you can edit a little Myspace page and copy and paste whole hog a few news articles as your updates.

Re:Height maps

[Anonymous Coward]

when Windows 95 came along. It wreaked It wrought. Unless maybe you meant to write reeked.

Re:Height maps

[somersault]

My first website was actually done in Word, it's funny looking back :s That was a few years before I learned HTML as part of my University course - I'd never been interested in HTML, I guess I considered it 'below' me or something since I could code in C and Pascal. The history of HTML is pretty interesting though, and it is obviously a very powerful/useful standard for laying out pages and is one of the main building blocks for the internet as we all know it today.. these days I do my HTML coding with Perl :) There was one point a few months ago where I was generating dynamic Javascript menus with Perl, I found it quite amusing writing code in an interpreted high level language that then generated code in another high level interpreted language that then generated some dynamic HTML for a menu structure :P That was just using a menu system I'd nabbed from somewhere else though, I probably would have been better off learning more about dynamic HTML and writing my own. I looked into it a few years ago and it seemed like a complex nightmare with all the differences between IE/Netscape DOMs etc - these days it's better though, browsers are a lot better at handling dynamic changes. Meh I always drift waaaay offtopic! Basically you're right, these kids need to learn what real programming is, and how the internet actually works..

Re:Height maps

[TimKemp]

Actually 16 bit addressing could be done easily on the 6502 with the "indexed indirect" and "indirect indexed" addressing modes. For code in EPROM you either used these or if it was something speed sensitive (since these were very slow addressing modes) you could copy a small routine in low memory RAM and modify the address as needed. This technique was used by Microsoft BASIC to access code.

Re:Height maps

[treeves]

My first website was actually done in Word.

Really?
My first Flight Simulator application was done in vi.

OK, not really.

Re:Height maps

[billcopc]

If by "paid okay" you mean "less money than an unskilled immigrant laborer", you're absolutely correct!

Programmers in the 80s and 90s used to make a zillion dollars, and many of us took pride in the quality of our work.

Programmers in the 21st century earn less than the secretary, have unrealistic deadlines and get blamed for everything.

I think part of the problem is that it's near-impossible for a layperson to gauge whether a developer is good or not, so they assume we're all idiots and treat us as such.

Re:Height maps

[billcopc]

This is quickly descending into an "us vs them" debate. Self-modifying code is a tool, knowing where to use it is even more important than knowing how to use it.

Self-mod code was appropriate for what I was doing at the time. I needed my cost to be as tight and fast as possible, and throwing more hardware at it simply wasn't an option - I was already using crazy-fast SGI machines.

Conversely, using the phrase "grow up and learn" to a veteran software developer, now that's real mature! :P If you think my work consists of a series of disjointed hacks, cobbled together with random luck in-between coke bumps... well I'm afraid there's nothing I could say to open your eyes.

What I did back in the day, was nothing more than extreme instruction-level optimization. Today's compilers do a pretty good job of it, and the CPU decoder does a whole lot more, so hand-tuning is rarely needed, and quite frankly today's hardware is so fast, it doesn't matter anywhere near as much as it did back in the 90s. My current processor is theoretically 800 times faster than the one that ran my old music software. It isn't worth my time to spend 8 hours tweaking 300 bytes of machine code because the sub-optimal code is already fast enough and probably less than 20% away from "perfect code".

What typically happened back then, I would write a prototype in a high-level language (which meant C or Pascal), then convert critical parts into watertight assembler. Particularly for sound and graphics loops, the speed gains were anything from 50% to 5000%.

Key example: bilinear image scaling, on a 486, was very slow. To scale a 320x200 image up to 640x480 took roughly one second in plain old C. The assembler version could run at 20 frames per second, and yes, I used self-mod loops in that one. The non-self-mod version ran at quarter-speed, due to the heavy nested loops in the convolution matrix - that's just how bilinear is done. The self-mod unrolled the loops on-the-fly, used precalculated offsets, and juggled the coefficients in fixed-decimal while avoiding register thrashing.

Sure, it was unmaintainable, but these were small, tight, purpose built functions. There is only one correct algorithm to do bilinear image scaling, it's not a business rule or a kernel driver - it's true computer science that has been tried and proven. I just coded it to be as fast as I possibly could.

Re:Height maps

[billcopc]

Not quite... wrought is the past tense of work. When used in "wrought havoc" it does form a correct phrase, though the meaning is slightly different from "wreaked havoc".

I'll spare you the full details, there's a good writeup at http://www.takeourword.com/Issue048.html /My language beats your language

Re:Height maps

[somersault]

Yeah I heard in the 90s that computer game coders get a really raw deal, which kind of put me off from going into the games industry and so instead I went to work for my uncle's engineering company, and some of the things I write here can be almost as interesting as writing games.

I'm not even sure if I'd be interested in coding games these days as I tend to see most of them boring not-quite-as-good clones of stuff I've already played.

As a side note, while thinking of the types of games that would interest me, I remembered Elite IV. It caught my interest because as well as the standard space trading you can also enter a planet's atmosphere and walk around inside its cities. I really enjoyed X2 but lamented not being able to actually land on the planets (which you could do in Elite 2 of course but you couldn't walk around). According to Wikipedia, Elite IV could actually be out this year!

Re:Height maps

[billcopc]

According to Wikipedia, Elite IV could actually be out this year! Yep, right after Duke Nukem Forever.

*crickets*

Seriously though, engineering would have been a fun thing to do, if only I hadn't had tunnel vision in my teens... back then, I never thought I would grow tired of programming and fed up with the industry. I could sure use a Flux Capacitor right about now.

Re:Height maps

[JFMulder]

But I bet you could have done it in emacs. ;)

Why?

[Iamthecheese]

A lot of power for some eye candy. IANAG(gamer) but it seems to me that more investment into the story line and playability would go a lot further than raising the system requir --oooh shiny!

Re:Why?

[shermo]

Certainly more gameplay and a decent storyline would make it a better game. But sadly, fancy graphics will probably sell more on opening day. (See spiderman sequels)

Re:Why?

[Anonymous Coward]

Some very nominal special purpose hardware would eat this alive. Remember intel is using unaccelerated general purpose processors to do this!

Re:Why?

[Anonymous Coward]

A lot of power for some eye candy.

Only sixteen cores?! For real computing power, you'd could run even more cores-- perhaps (Beowolf?) cluster several million machines so that each is responsible for a single ray/pixel.

Ultimately, this massively parallel distribution will provide data from an even bigger experiment-- what happens when you trace rays from the sun, bounce them off the earth, hit the CO2 layer, bounce back to the earth, back to the atmosphere, back to the earth...

Re:Why?

[Sabz5150]

Certainly more gameplay and a decent storyline would make it a better game. But sadly, fancy graphics will probably sell more on opening day. (See spiderman sequels) Really? See: Wii.

Re:Why?

[Kjella]

Sorry, but expectations change. Fancy graphics don't make a good game, but poor graphics (as relative to the times) does make a game poorer. Once upon a time I was happy with jumping 2D sprites in 16 colors, but when you've played a visually stunning game you think "Why can't [good gameplay game] look like that? It'd be even better. Don't get me wrong, it's being a good game that makes me want to play it long in the first place but when I do play, it shouldn't look like an eyesore. Same as you wouldn't play a lousy game, but if a good game had poor and repetative music you'd get fed up with it. Not that everything has to be ultra-hyper-realistic, I for example love Sam & Max which is hardly the epitome of realism, but I for example like that they upgraded the graphics engine so I could play season 2 in 1920x1200, looks much better that way. I also played through the whole Oblivion (got fed up by the expansion tho) and the fact that it looked so good definately was one of the reasons I was at it that long. And half the reason me and a friend play through Bubble Bobble (you'd think we were emulator cheating if you saw a recording, we cruise through) is the catchy tune, the other half nostalgia. The only downside is that making a game is so much, much work than it was in the old days. Though honestly, I'd rather take 10 excellent games than 1000 ones in any category.

Re:Why?

[Keyper7]

Actually, I believe real-time ray tracing open up some very interesting gameplay possibilities if people know how to use it.

Imagine a FPS, for example, on which you could notice a sneaking bastard on an unusual angle behind you because you saw his reflection on the doorknob you were about to pull. Or maybe cursing at the newbie because he didn't pay attention to the position of a specific lamp and now your team is screwed because your shadows have been noticed.

Then again, I think the whole FPS genre is saturated. Examples of other types of games are welcome here.

Re:Why?

[vadim_t]

I see two things here:

1. It runs on 4 x quad core. Which is about just 4X the CPU power a normal user could have right now. A 4X speed improvement isn't probably that far away. They may be hoping to reach a point where a dedicated video card is no longer needed. With the required performance level being so near, adding some extra support to the CPU may be enough.

2. Raytracing scales differently than methods currently used in games. With raytracing, increasing resolution is what adds the processing time, while adding detail is very cheap. Which I'm guessing means that as soon as you get raytracing going in real time at a decent resolution, adding extra quality is cheap. This would radically change the current situation, and possibly drastically bump the quality level.

3. Raytracing implements effects like shadows and transparency in a straightforward manner, which should make it easier to code. Game developers should like that. Also, in my understanding, raytracing also doesn't need to decompose things like spheres into lots of triangles, so the engine can test a ray's collision with a sphere directly. If you can specify parts of a scene as objects like spheres, toruses and such, it'd result in much finer detail.

What I think Intel is trying to do here to ATI/AMD and nVidia is the same thing fast CPUs did to soundcards. There's no longer a real need to have specialized hardware to play MIDI or add effects to sounds, since the CPU is quite capable of doing it itself. In fact, IIRC, Creative had to *blackmail* John Carmack into supporting EAX, because he could implement the same effects faster using the CPU.

Re:Why?

[nschubach]

One odd case in many. How many people go to the store and look at the back of the box for pictures of the game. How many game sites have screenshots... Graphics sell.

Re:Why?

[Brian Gordon]

Except the graphics kind of look like crap. If it's going to run at 16fps it better look a LOT better than traditional optimized rendering. But.. well, they're very low-res screenshots and the texture detail is Quake III at best. I kind of raised my eyebrows at a few of those ET:QW shots; the environments seem very sparsely populated by anything except solid geometry and the near-solid white skies reminded me of Halo 1.

Re:Why?

[Keyper7]

But one could also argue that graphics sell precisely because gameplay innovations have stagnated. I mean, there's not much to look out for other than the graphics when game companies are making FPS, FPS sequels and FPS versions of other games.

Re:Why?

[CastrTroy]

Can you even tell what the game looks like from the shots on the back? 1.5 inch square photos don't give you much of an idea of the graphics quality of a game. Especially when you don't know if the screenshot is from a cutscene or from actual gameplay.

Re:Why?

[Cheapy]

Well, in this case, there is no real story. ET:QW is a purely multiplayer game, unless you wanna play against only bots. It's quite playable too. (and runs on linux if you download something from id!)

Re:Why?

[mabinogi]

I personally don't want to play a game with a story written by Intel's 3D researchers - I'm quite happy for them to carry on working on what they know best.

Re:Why?

[Typing Monkey]

Yeah the genre could use more creativity. That said, ETQW is not your run of mill FPS.
Multiplayer only. Encourages teamwork by having five different classes that need to work together. Each map has multiple objectives. The two sides doesn't have the same weapons, vehicles and, more then often then not, specialities.
If you like FPS multiplayer games but crave more then deathmatch and capture the flag give the demo a try. Avaliable on both linux and windows(and mac but only full game afaik).
A bit of a plug, but it's a fun game.

Re:Why?

[Zerth]

And yet Dwarf Fortress/Nethack/etc is so much fun, despite being ASCII. Perhaps even because of being ASCII.

Re:Why?

[Anonymous Coward]

Fancy graphics don't make a good game, but poor graphics (as relative to the times) does make a game poorer.

every ps3 owner tells me this same thing. Yet they always are at my house playing my Wii.

Re:Why?

[ArbitraryConstant]

It runs on 4 x quad core. Which is about just 4X the CPU power a normal user could have right now. A 4X speed improvement isn't probably that far away. They may be hoping to reach a point where a dedicated video card is no longer needed. With the required performance level being so near, adding some extra support to the CPU may be enough. The eventual goal is to use large numbers of minimal x86 cores, I think they can increase the performance for a specialized workload quite a bit.

Re:Why?

[lordofwhee]

As a fairly long-time FPS player, I can tell you that no FPS will EVER have any large amount of teamwork (unless you're in a clan, playing with friends, etc), no matter how much the game 'encourages' it.

Re:Why?

[Anonymous Coward]

don't forget simulations of nuclear reactions, bending of space time and gravitational red shift while you are at it.

Re:Why?

[DeathCarrot]

A reflection on a doorknob and shadows are already quite easily achievable by current raster-based techniques. The former with dynamic cube-map FBO/PBO reflections (not perfectly accurate reflections, but given the size of a doorknob, more than acceptable). For the latter, per-fragment shadows (maps and volumes) have been around for quite some time (granted, in certain extremely high detail scenes ray tracing shadows might be faster).

The biggest immediately noticeable pros of ray tracing from what I've seen are reflections in arbitrarily complex geometry (current generation raster shadows are only viable for planar and some spherical reflections, unless there's a technique I'm not aware of). This, however isn't a good enough reason to switch to a purely ray traced paradigm IMO.
From a gameplay perspective it all seems a bit niche, but I'm sure there's someone out there with an idea that could make use of it. I just don't see FPS du jour picking it up any time soon.
Having said that, ray tracing may be a good utility to use alongside rastering techniques for things like sub-surface scattering or ambient occlusion.

.. Not sure that was entirely on-topic, but there's my tuppence on the near future adoption of ray tracing. Of course eventually everything will be done with unbiased rendering (basically just firing photons around and making them behave just like real photons would, see Maxwell Render. Currently, still extremely time consuming)

Re:Why?

[beav007]

The Wii was popular because of certain promised games which are still yet to be Wiileased.

Re:Why?

[beav007]

Forgot to add that the link above is marginally NSFW.

Re:Why?

[Kjella]

1. It runs on 4 x quad core. Which is about just 4X the CPU power a normal user could have right now. A 4X speed improvement isn't probably that far away. They may be hoping to reach a point where a dedicated video card is no longer needed. With the required performance level being so near, adding some extra support to the CPU may be enough. If everyone had a quad-core, which I doubt is "normal" now though I'll admit it's a standard desktop chip. And I usually play at 1920x1200, that's another 2.5x, plus 14-29 sounds too law to play comfortably, I'd add at least 2x there. So 4x*2.5x*2x = 20x away. Furthermore, while we're moving to faster processors it doesn't look like performance per watt improves that fast. According to Anandtech we can expect a 20-30% overall advantage over Penryn with a 10% increase in power usage, in other words maybe 15% performance improvement per watt. At that rate it will take forever to reach 20x. I also think it's obvious from the nVidia chip that supposedly has a 230W TDP that we're past the time where we can just increase performance with more transistors, bigger and more power hungry chips. I don't see this happening unless you can make a dedicated chip that does this much, much better than the CPU. A "raytracing processing unit" if you will.

Re:Why?

[grumbel]

Or maybe cursing at the newbie because he didn't pay attention to the position of a specific lamp and now your team is screwed because your shadows have been noticed. Not really exciting, MetalGear2 had that, SplinterCell had something like that and plenty of other games had similar stuff (Doom3, etc.). You don't need raytracing for that and in terms of gameplay it doesn't really add much, since most games simply are not slow enough that you care about little details like shadows.

Re:Why?

[FuturePastNow]

The eventual goal is to use large numbers of minimal x86 cores, I think they can increase the performance for a specialized workload quite a bit. Yep. http://en.wikipedia.org/wiki/Larrabee_(GPU)

Re:Why?

[Draek]

It depends on the person. Me, I've played Devil May Cry 4, yet I'm perfectly happy playing Jazz Jackrabbit (if you haven't played it, it runs on DOS, 'nuff said), and the fact that I've finished Half-Life Episode 2 doesn't prevent me from enjoying a nice time in DooM or Duke Nukem 3D.

Artistic design is much more important than the technical prowess of the graphics, for me, but even that won't stop me from enjoying a game if the gameplay is good (see also: Far Cry).

Re:Why?

[jaxtherat]

You can't have played Tribes http://en.wikipedia.org/wiki/Starsiege:_Tribes or OpFlash http://en.wikipedia.org/wiki/Operation_flashpoint much.

Even with total strangers they were quite good at forcing people to help each other.

Re:Why?

[Bodrius]

Not sure why this is only moderated as funny - it is quite true.

But that's because the Wii graphics are not really poor - they're just adequate for the games people really play.

Of course, if you try to put something like Assasin's Creed or GTA IV on the Wii - the graphics will suck and affect the sense of immersion and gameplay.

But that's also why no one is really playing those kind of games in the Wii.

Re:Why?

[flappinbooger]

If you went crazy with it you could simulate every second of every day of every person on the earth, like a simulated earth, and then invent some kind of direct-to-brain interface, and then when you keep all the people in some kind of unconscious stasis pod thing, then the machines could take over the world, then ....

That would make a cool movie....

Re:Why?

[William Baric]

I bought Oblivion, looked at the pretty graphics and stop playing this extremely boring game pretty fast. I only did the mage's quests, the arena and went as far as to escort Martin to Bruma in the main quest. That's it. As I wanted to play a CRPG at the time, what I did was play again with Ultima Underworld, which is probably your definition of "eyesore".

Personally, after countless bad experience, I'm to a point where I'm very worry with good graphics. I almost automatically associate good graphics with poor gameplay and I tend to simply overlook those games.

As an example, a year and a half ago I discovered the Gothic series with Gothic 3. It was a fun game, much better than Oblivion, but I also got bored with the game after a while. I read a lot that the previous title were better, so I bought a copy of Gothic 1 on ebay. Of course, graphics were a lot worse, but the game was also a lot better and it was one of the few games I finished. I also bought Gothic Universe simply to have Gothic 2 NOTR.

Of course, the same game would be better with good graphics than with bad graphics, but graphics are still secondary to gameplay. I'd prefer an "eyesore" with good gameplay than a beauty with an average gameplay.

Re:Why?

[WeblionX]

Yeah, but the sequels would suck.

Re:Why?

[billcopc]

I fail to see why we can't have both.

I loves me a good plot, with rich writing and character development, but I also loves me some 27" max-detail graphical virtuosity. Mass Effect is getting a lot of love from me right now, because it delivers a healthy balance of plot and visuals.

Crysis, of course, is a rather sexual experience at 1920x1200. Every now and then I'll perch myself atop a cliff and gaze at the breathtaking imagery... then I go back into cloak mode and snipe the mofos back to hell! I'm not a big FPS fan, but Crysis is one of the few titles that give me a sense of immersion, like I'm actually a gun wielding superhero instead of some synthetic alien cannon fodder.

Would I play Crysis if it looked like ass ? Probably not, because that particular experience hinges on the realistic graphics and all the fine details.

Would I play Quake if it looked like ass ? Hell yes, I would! In fact I did, it was called Quake 1-2-3... They're ghetto by today's standards, but the action was solid and I happily pissed away countless hours railing goddamned teenagers on instagib maps. Just give me a pixel to shoot at and I'm set!

I'm trying to think of a gorgeous game that sucked... memory is failing me right now, but there have been many. Actually, at the risk of getting flamed to death, I'd say Oblivion was one such stinker (for me). The graphics were pretty nice for its time, but I found the actual gameplay sluggish and clumsy. The sandbox concept worked well, but I spent most of my time walking around those stupid hell dimensions looking for stuff to kill, and then dicking around towns waiting for some NPC to come out of hiding at a specific time of day. Much like GTA, I quickly got bored of the storyline and started playing randomly, killing innocents and all the guards I could handle. I stopped playing it after maybe two weeks... epic fail.

Re:Why?

[indi0144]

do you want a $500 Ethernet cable with that?

Re:Why?

[Anonymous Coward]

Yup, Carmack wasn't going to use EAX in Doom 3 so Creative pulled out some patent they own on something Carmack was doing in his 3D engine (nothing to do with sound) and said "Add EAX or we'll sue you".

Of course the patent was on an obvious algorithm that every man and his dog had invested independently.

Re:Why?

[billcopc]

Good one, but seriously.. fire up one of the many horrible Wii games out there, and you'll see that great graphics can save a crap game, but crap graphics on a crap game will make you want to throw chairs.

What makes the Wii so much fun is that it has a couple dozen absolutely fantastic games that make up for the hundreds of shitty ones. It's like the SNES all over again!

Re:Why?

[billcopc]

Yep, ETQW is an interesting game reminiscent of Battlefield to some degree, but it only serves to highlight my deeply-rooted sociopathy, because I fail miserably vs the computer, even on easy difficulty :( I haven't even ventured online yet, I already know I'll get creamed.

Comparatively, I enjoyed UT3 a bit more, largely due to its simplicity: grab orb, make a run for it, capture node / destroy enemy node. No support classes, just Speedball 2 with guns :)

Re:Why?

[ampathee]

Oh, it's been released.

Re:Why?

[aarmenaa]

Actually, with your standard raster renderer, they're mostly cheating to make this stuff work. Yes, you can get away with it in very specific scenes, but it's really kinda annoying to have to count how many doorknobs the player can cram into their field of view at one time, to make sure performance doesn't drop to crap when they're suddenly staring at 10 doors instead of just one or two. What happens when you start running around a hotel? Whaddya know - none of the doorknobs reflect anything anymore! Why? Because PowerPoint isn't in the games section at your local BestBuy.

Dynamic shadows are still optional in just about every game that features them. Why? Because only really fast, expensive cards can do it, and even that's stretching it. Some engines will only render dynamic shadows from one light, and the ones that do it for more than one have to be very careful about not placing too many light sources too close together. This will get better as games finally drop support for DirectX 8 fallback (DX8 is generally not able to do dynamic lights at all), but the other restrictions will likely remain for quite some time.

There's a lot of shortcuts, hacks, and arbitrary map design rules to make this type of thing work. Even things that we're pretty good at these days like water are often more restricted than they seem. You know that nice water in Valve's Source engine? You can't have more than one body of that high-quality water on the screen at the same time, or you're liable to get "unexpected behavior." In other words, it doesn't work. This is specifically mentioned in some of the map making tutorials. So you'll find nowhere in a decently designed map where you can have two different pools with that nice high-quality, reflective, refractive water. You can use the "cheaper" water, though. The cheaper stuff still looks OK, but it's not the real deal and looks strange to see the two side by side. This is mostly an issue where the level of the water is different between two pools of water, such as having a diving pool next to a raised water tank or something. If they're the same height, you can just cheat and use one one body of water clipped through the intervening area for both both bodies of water. This only works, of course, in places where the area in between is unimportant - water floating in mid-air just might be an unexplainable phenomenon in the context of some story lines.

The way I read this, raytracing makes solutions for these types of problems more universal (ie. you just have water, no BS about what kind of water it is and where it can be), and the performance hit for doing it several times in a scene is way less. I'm in no way involved with graphics engines, but I would assume that most of these features require rendering the scene multiple times with different deformations, and that raytracing is somehow faster at this.

Re:Why?

[symbolset]

I'm trying to think of a gorgeous game that sucked...

Myst

Re:Why?

[Anonymous Coward]

I'll agree that better gameplay would go further than better graphics if you'll at least agree that nicer looking games are neat. But I'll also say that graphics wizards aren't going to start crafting great games, and neither are game designers currently wasting their time writing ray tracers. Two totally different groups of people do these things, and until game designers as a group get serious, better games aren't going to happen. And I don't think graphics programmers have much to do with it, so they might as well keep up the good work.

Re:Why?

[Max Romantschuk]

2. Raytracing scales differently than methods currently used in games. With raytracing, increasing resolution is what adds the processing time, while adding detail is very cheap. Which I'm guessing means that as soon as you get raytracing going in real time at a decent resolution, adding extra quality is cheap. This would radically change the current situation, and possibly drastically bump the quality level. Adding detail is cheap in some situation, and insanely expensive in others. With true raytracing and multiple reflective and/or refractive surfaces an awful lot of rays can be going all over the scene... So adding detail might really give an exponential cost addition in certain scenarios.

3. Raytracing implements effects like shadows and transparency in a straightforward manner, which should make it easier to code. Game developers should like that. Also, in my understanding, raytracing also doesn't need to decompose things like spheres into lots of triangles, so the engine can test a ray's collision with a sphere directly. If you can specify parts of a scene as objects like spheres, toruses and such, it'd result in much finer detail. If you're using CSG (Constructive Solid Geometry) it's true that a sphere is in fact a real sphere. But only if you're not starting with a scene with polygons. Most game worlds are polygon based, do going to a largely CSG-based world wouldn't be a trivial task.

Also, some effects, like focal blur for instance, are very expensive to do with full-on raytracing. Shadows, on the other hand, are indeed very cheap.

Re:Why?

[The MAZZTer]

I've played a number of Team Fortress 2 rounds which prove you wrong. The whole game just encourages teamwork. You have the medic class who can go on his own but is best when he helps others. Get a few engineers together and they can set up their sentry turrets quicker, and have a tighter defense with more turrets. Spies can take out opposing sentries to help their team get through.

Granted, true coordination takes a bit more, but often all it has taken is one guy with a mic who suggests a good plan of action.

Re:Why?

[Anonymous Coward]

I call your bluff. ps3 owners do not have friends, so they can't go to friends places to play with their wii's

Re:Why?

[Typing Monkey]

I guess that depends on how you define teamwork. In the etqw context I have engineers coming around to repair my vehicle when it's damaged. Field Ops supplying me with ammo. Medics healing and reviving me so I can complete a hack on the objective. Strogg technicians creating spawn hosts from fallen enemy troops so I can respawn behind enemy lines. Strogg oppressors putting up shields around me so I can disarm an explosive charge without getting shot in the back.
The addition of objectives gives players something to rally around and keeps things focused.
Of course etqw isn't the perfect game. Stacked teams, assholes and terrible games isn't unheard of.
But find a couple of server with a play style and regulars you like, and it is certainly better then many other games I've played.
It also has built in VOIP so the tools for very coordinated teamplay on public servers is there.
The playerbase is generally mature, even the younger crowd. I've almost never heard the voip abused and there isn't much non verbal abuse either(disclaimer: I haven't played on that many US servers).
But yeah, if you define teamwork as huge team pushes with perfect squad balance and shifting ad hoc goals you need to join a clan.

Re:Why?

[Schnoogs]

Probably because PS3 owners bought the console for Bluray where as anyone interested in gaming bought a 360.

The Wii is not for gamers. Instead it appeals to those who are into fads.

Re:Why?

[SupremoMan]

Maybe for the unrefined masses they do. As for me, I have been burned a few times before, so now I know that good graphics can't cover for crappy gameplay. When I look for a game, I try to get a demo and reviewes first. Looking on the box, as fascinating as it is, should be the last thing a person does.

Re:Why?

[Cochonou]

Myst ? It seems this game is really a love or hate case. As far as I am concerned, I found most of the Myst games to be really terrific experiences. So immersive ! Riven (Myst II) went a bit over the top with puzzles, though.

Re:Why?

[Anonymous Coward]

A reflection on a doorknob and shadows are already quite easily achievable by current raster-based techniques. Yeah, but normally you wouldn't do that. Rendering a cubemap of the whole scene for each worthless object is silly. (not that this wouldn't be a silly reason to raytrace..)

Re:Why?

[rastoboy29]

I agree with you completely. That's why I made my own game. I got tired of one well made WW2 or Quake style game after another, and decided to make the game I want to play. And I think you should, too.

The graphics are definitely not first class, but on the other hand I think they're beautiful, because I made it to please me. And at least a few other people like it, too, which is nice.

I'd still be glad to have an artist on board, though :-)

Re:Why?

[Anonymous Coward]

Yet they always are at my house playing my Wii. Tell your friends to play with their own wii's if the find their ps3's boring. and call the police or something.

Re:Why?

[jomiolto]

Some very nominal special purpose hardware would eat this alive. Remember intel is using unaccelerated general purpose processors to do this! Exactly, this is what most people seem to forget. While 4 quad-cores (quad quad-core?) might seem like a lot, it is nothing compared to the recent GPUs that have hundreds of cores. If a specialized ray tracing processor could get you even half of the processing power of a single core of these general purpose CPUs (with a fraction of the cost/power usage), imagine what 100-200 of these on a GPU could do... (Keep in mind that ray tracing scales extremely well when you add processing units).

Re:Why?

[Provocateur]

Have they changed the gameplay recently? Is it all about the scenery? I'd rather a freight elevator or warehouse door whooshed open and suddenly the place would be overrun by angry hordes of zombies, orcs, cacodemons, pain elementals, heck even geohashers, all hungry for blood. Well, maybe not the geohashers. Use the processing power to draw more of my foes. Spare some to draw a raging fire in the background, the city in smoking ruins.

Re:Why?

[Waccoon]

I fail to see the advantage of shiny reflections when the scenery is made out of huge, chunky polygons. Brown and gritty doesn't look any better when it's raytraced.

I would think that good art direction and creative use of processing power would have a greater impact on graphics quality than trying to get every pixel to use every effect all at once. That's why I can look at games on the 360/PS3 running at sub-HD resolutions, running at 30FPS, and I think to myself, "it really doesn't look that good -- AND it's slow". It's a game. Stop messing up my framerates by either running too slow, or in the case of the PC, even too fast for my monitor to handle.

Stop trying to use bump mapping on every blade of grass, and maybe you can model more blades of grass, instead. I don't want to see real-time NURBS around every corner when a good compressor can crunch a pre-tessellated model into a reasonable amount of memory.

Of course, there's no hardware marketing model for "good art direction".

Re:Why?

[Yetihehe]

And now imagine a beowulf cluster of those! I'm actually working on it, Amazon EC2 is going to really rock for some advertisers. Imagine walking through newly designed casino, with all visual details. Only thing required would be laptop with internet connection sufficient for hdtv stream. Heck, you could even send small video to iphone. Imagine realtime raytracing on beowulf cluster with output to your iphone...

Re:Why?

[Anonymous Coward]

every ps3 owner tells me this same thing. Yet they always are at my house playing my Wii. I own a PS3, and I've never been round your house.

Re:Why?

[nyet]

Bah TF2 sucks. QWTF will always be the best.

Re:Why?

[Negatyfus]

Myst IV is still the best-looking game I have ever seen. Nothing can really compare to the panoramic animated graphics. The gameplay wasn't really for me as I'd sooner resort to a walkthrough than struggle for months to get through a single puzzle.

Re:Why?

[Gerzel]

Not necessarily. Oohhh Shiny sells far more copies thus making it a better game.

Re:Why?

[Anonymous Coward]

I love how this is modded insightful :)

Re:Why?

[Anonymous Coward]

I'd prefer an "eyesore" with good gameplay than a beauty with an average gameplay. I completely agree with this philosophy. And that probably explains why I'm okay with being single.

Re:Why?

[HigH5]

I see two things here: 1. It runs on 4 x quad core. Which is about just 4X the CPU power a normal user could have right now. A 4X speed improvement isn't probably that far away. Given the Moore's law, would that be (in theory) 3 to 4 years? That would be actually pretty soon.

Re:Why?

[tuxicle]

If raytracing is most efficient at low resolutions, I'd expect it to take off well in the game console market faster than it would on the PC market, where people seem obsessed with running games at ridiculously high resolutions.

Re:Why?

[Yetihehe]

Please read "Permutation city". Very good book, better than matrix IMHO. And it directly mentions raytracing.

Re:Why?

[cyborch]

abbreviations FTW... I read that as Quake What The Fuck... that may even be appropriate...

Re:Why?

[Anonymous Coward]

Mapping quality doesn't matter, raytracing will work at 16 fps no matter the polycount or texture size, it's one of its best benefits.

Re:Why?

[91degrees]

Don't dismiss eye candy. It can affect playability. Any trick you can use to help convince the brain that it's really there makes the game more immersive. Making objects more distinctive or recognisable makes it easier to tell a story (and bad graphics will distract from the story). If you make something really pretty, or really impressive generally you will get an emotional response.

Re:Why?

[Yetihehe]

So you made another WW2 or Quake style game but with not such good graphics?

Re:Why?

[Anonymous Coward]

It's all about the middle ground - the middle ground is what sucks.

Either extreme works really. You can go super-fancy with graphics and try to make everything look as slick as possible; or you can consciously choose not to have any of that and use only text, for example, or very simple graphics. Kingdom of Loathing is a good example of the latter.

However, trying to make your graphics look super-great and slick and then failing at it and only coming up with something that's run-of-the-mill and bog-standard at best... that sucks. And picking up your Nethack example again, that's why noone actually uses tilesets in Nethack, too.

Re:Why?

[ardor]

As a general rule, raytracing excels at secondary-ray tasks (= when one additional ray is shot originating from point of impact of the first, primary ray) like shadows, reflections, refractions, and the like, but doesn't pay off for primary-ray only tasks (solid wall, ground, ...) This is because rasterization is an optimization for the primary-ray-only case - you need local information only, no need to shoot rays for the entire surface, you can interpolate across it.

Now, those doorknobs will make ... how much % of the visible frame? In typical game scenes, opaque surfaces are the majority. This is not because of hardware limitations, but simply because most game worlds are NOT full of shiny surfaces.

A hybrid is the best approach. You correctly guessed that with rasterization, secondary-ray effects need to be calculated in multiple (and costly) passes. Moreover, since these passes usually render to a texture, you get aliasing issues because the texture pixels do not 1:1 match the screen pixels. You get blocky shadowmaps, blocky reflections, or noticeable noise when moving (in case the map is too large).

Re:Why?

[vidarh]

Of course, in the real world even flat seemingly featureless surfaces has lots of different reflections of light from multiple sources. Just looking around the office now, there are lots of shiny surfaces, and even the walls have complex reflective patterns on them based on reflection of light from the windows, a large number of lamps, and from secondary reflections from windows, interior glass walls etc.

The number of hacks to get anything resembling realistic scenes with rasterization is just so staggeringly high, and given the performance of current systems it's only a question of when raytracing will be "fast enough" to deliver better results in far less development time than what rasterization does.

So Intels current demo uses 16 cores. We buy servers with 8 cores now for less than $2k. Give it a couple more years and you'll start seeing consumer systems that are getting fast enough.

And the nice things about raytracing is that it scales almost linearly with the number of cores.

Re:Why?

[vidarh]

Most game worlds are polygon based because that's what works with current engines. Nobody would start rewriting their current games for a system like this, but next time someone rewrite a significant portion of their engines or write a new engine, they might look at whether raytracing is becoming viable for them. Sooner or later it will be. Once it's viable, it's a question of whether it delivers as good or better quality and whether it can potentially cut their development times.

Re:Why?

[JasterBobaMereel]

Oohhh Shiny sells for the first week but then people discover if the game play is any good and the late adopters depend on recommendations and experience on friends systems to decide if they are going to buy ...

Some games require detailed graphics (or the game is unplayable) some work fine without the highest level (many Wii games) some work fine with almost none (many early console games, Tetris!) If I have to upgrade my PC or Buy a new console just to get some pretty graphics I won't bother, if however the game is fun to play as well then ....

Re:Why?

[dintech]

I'm actually working on it

Imagine realtime raytracing on beowulf cluster with output to your iphone...

Ah, finally! An interview with a Duke Nukem Forever developer!

Re:Why?

[JasterBobaMereel]

For real computing power use a proper chipset not the crippled, still compatible with a chip designed in 1978, garbage that Intel peddles....

Re:Why?

[markov_chain]

I am with you! Graphics is for chumps!

Re:Why?

[Anonymous Coward]

Two weeks of entertainment for $60 sounds pretty good to me... well, at least not an epic fail.

Re:Why?

[RulerOf]

I'm trying to think of a gorgeous game that sucked...
Your short term memory must be failing you as well, because you mentioned Crysis not two paragraphs ago.

Re:Why?

[mgblst]

There is no reason you can't have both. You won't find both from the developers of Quake and Doom, since they are primarily vehicles for showing off the graphics and getting people to buy the engine.

Re:Why?

[KDR_11k]

The Wii is not for gamers. Instead it appeals to those who are into fads.

Yeah and that fad will die any minute now. Aaaaaaany minute. Hey, we've been predicting it for almost two years now, it should come true soon. Oh and that DS fad should be over any day too.

Re:Why?

[KDR_11k]

Personally, after countless bad experience, I'm to a point where I'm very worry with good graphics. I almost automatically associate good graphics with poor gameplay and I tend to simply overlook those games.

In my experience games with bad graphics tend to be shitty in other ways too since it's usually a sign of lacking effort, not a sign of focussing on something else. Good graphics require effort and usually the game underneath tends to be at least passable (as opposed to total crap where you're lucky if touching a wall does not make your character drop through the floor).

I'm not sure but I think the Gothic games (except 3) didn't have bad graphics for their time. If you sample games from the past and judge their graphics by modern standards you'll of course find a lot of games that are ugly but good but that's mostly because the graphics have aged, not because they had bad graphics to begin with.

Re:Why?

[Anonymous Coward]

There's no longer a real need to have specialized hardware to play MIDI or add effects to sounds,

Unless you are in Linux... oh I forgot there is this timidity hack which does not really work.

Re:Why?

[symbolset]

Nintendo marketing and software can turn some shit hardware into the shit game in no time.

Gotta give up the respect for our asian friends at Nintendo.

Re:Why?

[Anonymous Coward]

Don't know if you noticed, but those are digital photos of a monitor displaying the game.

Re:Why?

[Skye16]

Absolutely agreed. Tribes (and Tribes 2) pretty much require it.

Tribes: Vengeance is an atrocity upon mankind and the developers should be facing a war crimes tribunal right this very second. Further, Vivendi should be dissolved and any who continue to link themselves to Vivendi in any way should be considered terrorists.

Re:Why?

[Anonymous Coward]

Go outside, the graphics are awesome!

Re:Why?

[Cctoide]

Meh. ETQW was pretty close to being a commercial bomb, and that's if it didn't really, actually bomb. I got the collectors' edition at release and I'd rather I hadn't. As someone put it, the Battlefield players hate it because it left out some things from Battlefield and the Wolfenstein players hate it because it didn't keep the Wolfenstein format intact.

I can't remember how many ETQW servers there were last time I looked at the server browser, it was either 200 or 700 - either way, it was under 1000, when BF2 at launch probably had over 9000 servers.

Re:Why?

[bestinshow]

Of course because GPUs are actually very programmable now, and ATI's latest card does over a TeraFLOPS of single precision right now (about 4-5x the single precision power of Cell, and probably 4x the single precision power of sixteen Intel cores)

Ah, yes: "Core 2 Quad has a max per cycle of 4 * (4 cores) FLOPS * speed (mhz, we will say 3.33Ghz) = 53.28 * 1,000,000,000 Hz, or 53,280,000,000 FLOPS". Basically, if the raytracing renderer that Intel have implemented on their CPUs was ported to commercially available GPUs today, they would be beating this Intel system, once optimised.

I.e., The GPU is here to stay, even if raytracing comes into vogue. Mixed rendering will likely rule the roost 2010 - 2015 anyway, where raytracing is used for things that would look better for it only.

Re:Why?

[Anonymous Coward]

You are absolutely right! At this point Crysis can't even run... and here we are with a four quadcore cpus trying to run this old game at below minimum specs of 14 - 29fps? I hope their graphics card is as good as the last ones. Garbage. nVidia has the market and will continue to do so. Unless developers start looking at raytracing as a viable option... which I don't see happening for another 10 years. By that time Larabee will be a flop and Intel will be developing their next research flop in the graphics industry.

Re:Why?

[Bones3D_mac]

Imagine a FPS, for example, on which you could notice a sneaking bastard on an unusual angle behind you because you saw his reflection on the doorknob you were about to pull. Or maybe cursing at the newbie because he didn't pay attention to the position of a specific lamp and now your team is screwed because your shadows have been noticed.

This kind of thing can already be done without ray tracing. Projecting shadows and reflections from an object onto another in raster-based engines really isn't all that difficult to pull off in most modern game engines.

Aside from that, exactly how big of a doorknob are we talking about? It's unlikely you're going to notice a reflection on a normal sized one while rushing around, dodging enemy fire in the process. Of course this assumes your setup even has a high-enough resolution to actually make such a minor reflection recognizable as an enemy within a fraction of a second.

Re:Why?

[Anonymous Coward]

some people can't handle the sandbox kind of games. its true its easy to get lost in it, but usually if you stick to the storyline you find its no different than your linear quake or crysis or mass effect type game. in the end, the only person missing out, is you :)

Re:Why?

[p0tat03]

Because when they're *not* over at your house they're too busy playing singleplayer or internet multiplayer, neither segments are really covered by the Wii. The "friend code" thing (as opposed to, say, a memorable user name) has completely destroyed any chance for the Wii to ever become a popular internet multiplayer machine.

See... on the PS360 you got vast, epic RPGs, crazy single-player shoot-em-ups, and lots of awesome multiplayer action games (COD4, Halo 3, etc etc.), on the Wii you've got party games... Each console has its niche, but I do have to say that the signal-to-noise ratio on the Wii is pretty pathetic. For every Mario Kart you have a million other hastily-produced Wii games that aren't worth the discs they're pressed upon.

Re:Why?

[Anonymous Coward]

Wiilease Wodger!
Wiilease Wodewick!
etc.

Re:Why?

[mdwh2]

And they're always round my house playing my SNES.

But wait, they're not. Why isn't everyone playing dirt-cheap old computer/console games (even cheaper with an emulator)? Surely the old technology isn't a problem?

Re:Why?

[Oktober Sunset]

Especially when you don't know if the screenshot is from a cutscene or from actual gameplay. Memories of every final fantasy game ever just flooded my mind. Those lying bastards!!

Re:Why?

[mdwh2]

Actually, with your standard raster renderer, they're mostly cheating to make this stuff work.

I'm not sure "cheating" is a fair word. Yes, it is a problem that you have to be careful with the performance of reflections with standard renderers, but both that and raytracing are not doing 100% accurate simulations of reality.

Dynamic shadows are still optional in just about every game that features them. Why? Because only really fast, expensive cards can do it, and even that's stretching it.

Not in my experience, I can happily switch on shadows on my rubbish Intel GMA graphics. But more to the point, a raytracer is hardly going to run on this kind of hardware, either! If we're going to need expensive next generation hardware to run the raytracer, it's fair to compare to a standard renderer running on that kind of hardware too.

Re:Why?

[mdwh2]

Indeed - see http://en.wikipedia.org/wiki/Shadow_volume#Depth_fail . It's still often known as Carmack's Reverse.

Re:Why?

[tonyreadsnews]

Well, Portal is one game with a recent change to gameplay. Certainly more to it since there are few blood hungry monsters in that.

And the scenery in Bioshock was very interesting and different.

Re:Why?

[Machtyn]

Yes, let's look at the Wii. It came out roughly the same time as Xbox 360 and the new Playstation. The GP stated "fancy graphics will probably sell more on opening day." (emphasis mine)

I remember LONG lines, limited purchasing, etc. for these Playstation things (as a gamer I scoffed... but mostly because I'm almost exclusively a PC gamer.) It wasn't until months later that the Wii really took off. People started to say, "Hey, these games are simple, but they're FUN." And then, since, it has been difficult to find them on the shelf even during the "off-seasons".

My point, the Wii wasn't first in line on opening day, but your point is valid that it seems to be the better system (better in terms of sales, general gameplay for the masses.) Though, I would argue the "decent storyline" is missing from most games.

/currently playing: Ghost Recon Advanced Warfighter 2 - nice graphics, decent (and even somewhat plausible) storyline.
/currently playing: Civ 4:Warlords with a few friends
/currently playing: Trackmania Nations: An incredibly good racing sim, that is FREE through non-obtrusive ad support (think billboards on the side of the road.)
/currently wanting: a Geforce 8800GT or 9800GTX, so I can crank the resolution on Oblivion and max out GRAW2
/just became a tennis pro on Wii Sports.

Re:Why?

[tonyreadsnews]

Odd, I loved Oblivion (and still occasionally play). But I absolutely loathe Mass Effect. Its one of the few I wish I'd never bought.

Maybe there's a part of the game I'm missing, but to me it seems like I spend more time waiting to load the next System/planet/building then actually playing. When I do get a chance to play, the action seems short and not particularly engaging, while the conversations slow and not exactly as groundbreaking as it seemed reported.

Re:Why?

[heartless_]

See World of Warcraft, see puzzle games, see online mini-games.

The problem with your argument is that you are basing it off inept data collecting, like NPD stats, that DOES NOT account for MOST OF THE revenue in PC gaming (MMO subscriptions, game advertising, website games, etc.)

Graphics sell BOX copies of games, and I don't know if you noticed, but box sales have been on a decline for years.

Seriously how many more major development houses and games do we need to see fail before people understand that expensive, top-of-the-line graphics do not sell games.

Re:Why?

[Anonymous Coward]

Why does nobody bother to point out that yes shadows and reflections are possible in rasterization but they are VERY difficult to program. In ray-tracing they are practically free from a programming perspective. This means making graphically rich games will be easier to program and game devs will be free to focus more on making the game-play better.

Re:Why?

[tilandal]

Sure its only 4-8x what desktop processors are today but you have to look at it in comparison to how it performs vs the competition. A low end graphics card coasting $100 can easily render quake wars at a better frame rate.

Intel is basically stuck because the average user has no need for more processing power. Not only that, they can not keep ramping up clock rates. By putting more processors on a core they increase power but do not increase performance for most applications because they can not efficiently make use of so many threads.

In the end Intel has a product they want to sell so they are trying their hardest to make an application for it. The problem is the hardware is not particularly good for the application. You could almost certainly get better performance for cheaper from a dedicated ray tracing chip.

Re:Why?

[robthebloke]

no it won't. As your data set increases, you become increasingly memory bound. You can have 4 cores pumping away at 100%, but if they are sitting there generating page faults and cache misses all over the place, you are going to see some pretty bad frame-rates.

Simply put, a GPU is fast because it can stream data at speeds > 1900Mhz. In comparison, a quad core intel machine with memory speeds of 800Mhz can't feed the cores with data fast enough. In my own tests, if you can keep the data requirements per-frame to approximately the size of the CPU cache, you can see some playable frame rates. As you go over that size, frame rates start to drop exponetially.

Ultimately though, i don't really buy into this 'raytracing is going to revolutionise game graphics' point of view. It's never really revolutionised the film FX world, where even now after 25+ years since Tron, the most commonly used techniques are based around REYES and global illumination models that do not rely on raytracing. Certainly renderers such as renderman have added raytracing extensions, which can be useful in certain situations, but ultimately the majority of FX shots don't really require them.

Re:Why?

[Jesus_666]

Another love-or-hate thing. Even though Mass Effect isn't too big on the action (but to really bad either) and the (IMO mostly interesting) conversations can draw out forever, especially if you just want to wrap up a sidequest and ten NPCs pop up to talk you into doing their sidequest right now, it's still a solid game. Much of what makes Mass Effect good is the solidity of the world and the writing. Bioware just knows how to create huge, immersive worlds where every insignificant locaton has its own backstory - and they are extremely good dialog writers. Even the myriads of elevator rides are lightened up by your team's banter.

Of course you might feel that a detailed universe and witty dialogs can't make up for the fact that the combat isn't executed as well as it could have been. I think Mass Effect is on par with Advent Rising, but I'm a sucker for good writing and immersive universes.

Re:Why?

[Anonymous Coward]

Would I play Quake if it looked like ass ? Hell yes, I would! In fact I did, it was called Quake 1-2-3... They're ghetto by today's standards, but the action was solid and I happily pissed away countless hours railing goddamned teenagers on instagib maps. Just give me a pixel to shoot at and I'm set!

All Quake games (except maybe Quake III) looked like ass from the very moment they were released. The selling point of the first wasn't the graphics, but being a "true 3D" motor.

Re:Why?

[Jesus_666]

I maintain that DoomRL is much superior to Doom 4.

I normally don't like roguelikes too much because they tend to be cruelly hard in the beginning, but DoomRL is relatively fast and easy to get into. If you're a rouelike fanatic who is put off by the game sounding too easy: Play one of the harder challenge modes like Angel of Berserk (melee only) on Nightmare! and we'll talk again...

Re:Why?

[mc900ftjesus]

See: Wii only has 5 games that don't completely suck.

Re:Why?

[mc900ftjesus]

Don't sweat it, Oblivion was crap unless it was your first RPG ever. Morrowind:classic novel::Oblivion:Soap Opera Digest.

Re:Why?

[mc900ftjesus]

Ugh, really? Oblivion was such a generic lifeless game, it would be hard to make it much more boring. There wasn't a surprise in the entire game. At least Mass Effect has a remotely creative story, even if the gameplay isn't your thing.

Re:Why?

[PachmanP]

I suspect it had more to do with the timing. UT3 and TF2 came out right after it. I liked the demo alot, but didn't get it cause I wanted to see tf2 and UT3 first. I have limited time and money, so I was only going to get one of those 3. I ended up with tf2 not because it was a better game but because my friends were all playing it.

Re:Why?

[Ilgaz]

Speaking about Quad Cores, as a Quad G5 owner I think Intel being the largest CPU vendor/Desktop CPU company should spare time to figure out what kind of Applications including core GNU software can be multi threaded to a point where the cores actually get used. They look funny when they try Graphics stuff competing with NVidia and ATI which are way beyond anything they could produce.

It is 2008 and I begin to see multi core aware Flash (check V10 Beta) on my 3 years old OS X system, browsers still use single core, bzip2 needs pbzip to use cores etc.

Re:Why?

[mc900ftjesus]

The Wii is the most overpromised system ever released. I'm waiting for a game to properly use the Wiimote other than No More Heroes. So far there's a giant pile of shovelware (including SSB, and Mario Kart) that don't really use the Wiimote, or the ultimate sin, WORKING BETTER WITH A WAVEBIRD. SSB/MK are shovelware purely because they're GC games with slightly better graphics.

Thanks a pantload Nintendo, but how dare you even offer GC controls in Kart/SSB? Worried your fanboys will get angry or are your new controls not revolutionary enough to use your fantastic new controller?

Not surprising that you changed the name from Revolution to piss. Because you pissed away the most promising system with compromises and the most abysmal software catalog possible. All 1st party titles again? That's my Nintendo.

Yes, the PS3 had a slow launch, but there's already more good games for the PS3 than the Wii. Crap GTA4 has more content than the entire Wii library.

The ONLY titles worth owning are No More Heroes, Wii Sports, and Wii Fit, seriously. The rest are just GC games. Funny that everyone was talking about how Wii Sports was a "tech demo" and how awesome the games were going to be, and here we are 1.5 years later, waiting on something better...

Re:Why?

[Z34107]

Of course games don't need amazing graphics to be fun.

But, I'm glad we can do more than plain ASCII nowadays. Anyone who wishes technological progress stopped circa 1960 because (according to a parent poster) bad graphics must make the story better deserves to be carved up by my link gun's alt-fire.

Well, maybe most of the parent posters need to be carved up with a link gun. I could make the world a better place with a link gun...

Re:Why?

[Zerth]

While I don't agree with the GGP that things should have stopped back in the 60s, my statement that some games are better because of poor graphics is true for 2 reasons:

1: less effort on graphics may leave more effort available for gameplay and puts less limits on gameplay(no need to produce walk animations, IK skeletons, visual physics, etc)

2: less effort on graphics leaves more room for the player to make things up to suit them. Tall elves vs short elves, ugly harpies vs attractive harpies, beards on dwarven women, etc.

Re:Why?

[Ibiwan]

I see three things here...

Re:Why?

[Creepy]

Ray tracing in default form uses a point light source, so you get hard shadows - in fact, traditional Ray Tracing does specular lighting very well but diffuse poorly. To fix the global illumination model for soft shadows/soft lighting you need to use some technique such as photon mapping or BSSRDFs (Bidirectional Surface Scattering Reflection Distribution Function, a technique mostly used for subsurface scattering). You could also use something like Radiosity, but that is horribly expensive (O(n^3) I believe), so not really suited for realtime.

It is much easier to do realistic diffuse lighting in a rasterizer currently, but reflections are easier to do in ray tracing. As I've pointed out before, there are advantages and disadvantages to both, and whatever Intel or nVidia or whomever says, the technologies will likely converge (in fact, many shaders I've looked at recently use mini-ray tracers or pseudo ray tracers, but most of these are "self-aware" as opposed to "scene aware").

Re:Why?

[ZERO1ZERO]

Op flashpoint : I have tried this game a good few times, but I seriously can't live more than about 10 seconds. (unless I hide) It's the most frustrating game ever. What use is a game if I just keep dying all the time? Most annoyingly I can never even see who's shooting at me I just kinda pop up and instantly die. No fun.

Quake or Quake III on the other hand, is more my style :)

Re:Why?

[Anonymous Coward]

Anyone interested in gaming bought a PC.

The 360 is not for gamers. Instead it appeals to those who like to be insulted by 10 year olds.

Re:Why?

[MemoryDragon]

Actually Gothic2 is the best of the series, it reminded me on the best of Ultima 7 in so many ways. Every Ultima fan should play Gothic 1 and 2. Gothic3 basically had the Oblivion syndrome. The bugs aside which the series aways was plagued. G3 simply had the same problem as oblivion doing a big world often ends with a semi boring game because lots of the dense athmosphere is lost by the auto generated parts of the content. I have yet to see a real world game which achieved this to some degree. Gothic2 did but it was a small world hence everything felt tight. Ultima 7 also managed it in a big world, but that was because the entire world was handcrafted. I agree with Oblivion after being a nice tech demo it stopped working for me, due to being extremely boring.

Re:Why?

[Anonymous Coward]

Obviously you didn't check it out!

Re:Why?

[level_headed_midwest]

ETQW isn't your run-of-the-mill FPS...it runs on Linux. :D

Re:Why?

[drinkypoo]

A lot of power for some eye candy.

The nVidia GT200's TDP is 236 watts! Tigerton-based Intel X7350 (2.93GHz, 8MB shared cache) has a TDP of 130 watts. So for little more than the TDP of the newest nVidia card, you get two 2.93GHz quad-cores. If the performance improves only twofold (and hardware raytracing is in its infancy, comparatively) then the performance will be comparable - and those processors are potentially more useful than the GPU for other types of processing simply because they are designed for general purpose computing.

Re:Why?

[Anonymous Coward]

1) Ray Tracing makes shiny graphics(tm) easier to implement

2) "Once you accelerate ray tracing to a decent speed, you can use it for collision detection, or use it for AI [Artificial Intelligence â"ed] to determine certain assets. For example, you could use camouflage to test the AI: You can determine the visibility of something when it is hiding behind different objects such as trees or one or more other solid objects. Ray tracing is not all about rendering. It is about how much more you can do using a single technique." (from TFA)

Re:Why?

[adavies42]

I'm trying to think of a gorgeous game that sucked... memory is failing me right now, but there have been many. Actually, at the risk of getting flamed to death, I'd say Oblivion was one such stinker (for me). The graphics were pretty nice for its time, but I found the actual gameplay sluggish and clumsy. The sandbox concept worked well, but I spent most of my time walking around those stupid hell dimensions looking for stuff to kill, and then dicking around towns waiting for some NPC to come out of hiding at a specific time of day. Much like GTA, I quickly got bored of the storyline and started playing randomly, killing innocents and all the guards I could handle. I stopped playing it after maybe two weeks... epic fail. How about the Myst games? I love the scenery, and the effects once I actually get something working, but I find the concept to be a complete failure, as I can't even find the puzzles most of the time, let along solve them.

Re:Why?

[PitaBred]

Don't forget that with raytracing things like collision detection are "free". That opens up a lot of other things that no longer require a loop through everything to see if your gunshot is actually hitting something.

Raytracing has a lot of advantages. It's main disadvantage is that there's a lot of specific acceleration hardware out there that's tuned to work on a competing algorithm, and it's not really a transferable operation.

Re:Why?

[Kattspya]

There's a mod for the original half-life that's called Natural Selection. The teams are aliens vs space marines and the teams are not "symmetric". The lowest class of alien is a dog sized thingy that kills a marine with two bites (it doesn't take more than half a second). It's fast as hell and can climb walls. Due to these mechanics a lone marine is a dead marine. You have to move together as marines or you're toast. That is the game that's given me the best teamwork experience on public servers.

It's probably not worth picking up due to it being so old. You will get your ass handed to you over and over. I do, however agree that more often than not public is a wasteland when it comes to teamwork.

Re:Why?

[shutdown -p now]

ETQW is a multiplayer-only first person shooter. What's that "story line" you're speaking of?

Re:Why?

[shutdown -p now]

Yes, but when you dig out that old Fallout CD and play it, don't you sometimes wish it looked as good as Oblivion, leaving all other things intact.

Re:Why?

[grumbel]

Nethack example again, that's why noone actually uses tilesets in Nethack, too. The problem with tiles in Nethack is that they don't fix the underlying structure of the game, i.e. in many tiled-nethack clients you still can't tell the difference between a wall and unexplored terrain, both look the same. Same is true with character movement and tons of other stuff. This is more awkward in tile graphics then in ASCII, because tiled graphics normally don't behave like that, they have enough expressiveness to properly represent walls and stuff, while ASCII stuff often behaves exactly like that due to the limited character set.

Re:Why?

[blueswan1]

Didn't Bioware have a hand in neverwinter nights 2 (and probably 1, I'm not sure) which had some of the best dialog I've ever seen in a game.

Re:Why?

[pugugly]

Not that I'm competent enough to have an educated opinion, but heck, never stopped me before.

But from what I have read of raytracing vs raster, it looks to me as if there are two major differences.

A) Raytracing automatically takes care of certian kinds of optical issues (eg, the looking up through water in the screenshots) which just simplifies life. If you have a mixture of items that some had these complications and others didn't, based on that alone, then it might make sense to use a hybrid solution.

B) Raytracing scales logarithmically, whereas rasterization scales linearly.

That second seems to me to be the determining factor in the long run - if you can double your resolution, but only increase your processing needs by log(2), then no matter how optimized your hybrid raster system is, at some resolution raytracing is going to be better *and* simpler.

I'm sure there is a window where Raster efficiency is still worth going after for specific things, but assuming Moore's Law holds for the next 20 years, just how long is that window going to be open and how much effort is it going to be to try to hit it?

That assumes I'm understanding the scaling issues correctly, but if so - is the return on this hybrid system going to be as long as one 18 month processor generation?

Pug

Re:Why?

[Jesus_666]

Actually, Bioware is the developer of NWN 1 and 2. KOTOR (both parts) and Jade Empire, too. The job they did with KOTOR prompted a lot of people to wish that BioWare had written the Star Wars prequels instead of George Lucas.

Re:Why?

[UnknownSoldier]

Obligatory: Reality.

Re:Why?

[tonyreadsnews]

Funny, my feelings are almost completely reversed. For me, the environment in Oblivion was inviting to explore. Plenty of treasures and unique spots to be found by just wandering around.
Mass Effect, however for me, each planet you can land on was almost the same as any others, large areas filled with only 5-10 points of interest. Not to mention, such a pain in the ass to get to most areas, just to see if anything showed up on the radar.

Re:Why?

[Anonymous Coward]

May I ask: Playing what?

Re:Why?

[Swampash]

I'm trying to think of a gorgeous game that sucked...

Unreal, 1998.

Re:Why?

[Mr. Hankey]

I'd like to see a CUDA implementation. With calculation speedups from 10x to hundreds of times over host-based processors for some algorithms on just one GeForce 8800GTS/GTX card (with 128 stream processors working simultaneously.. 8x the cores, even if lower clocked) this could get interesting very quickly.

Re:Why?

[Schnoogs]

Do yourself a favor and educate yourself on how many games the average wii owner has in their library. Then get back to me with your thoughts on the subject matter.

It's a fad plain and simple...fads can last several years. The wiimote is a gimmick and the game selection is incredibly small.

Re:Why?

[Schnoogs]

More like... "gotta give up the respect for the gullibility of the average American soccer mom who just has to buy little Jimmy this season's must have toy"

Re:Why?

[Anonymous Coward]

Ray tracing would allow game developers to focus more on the "playability" items - since they don't have to spend nearly as much time (and $) on what has become a labor intensive process of creating realistic graphics to work around rasterizations shortcomings. We currently spend 1/6 of the production cycle on tasks that could go away if the consoles included capabilities for real time ray traced images - that time could be spent on other things such as AI, physics, etc.

Re:Why?

[Sabz5150]

See: Wii only has 5 games that don't completely suck. So that's why they can't be found on store shelves TO THIS DAY.

Re:Why?

[billcopc]

Crysis seems to trigger a love-it or hate-it response, often for the same reasons.

I liked it because it was short, fast-paced and offered many paths to victory. You could drive along the road and face roadblocks, or you could hop on a boat, perch on an island and snipe from a safe distance. Sometimes I just ran like hell and avoided combat entirely.

The flight level was painful, mostly due to the terrible controls and sluggish response, which made the battles take far too long. All the land missions though, I thought they were pretty cool.

I haven't played any multiplayer maps, but I've played through the single player campaign twice, each time in a single sitting - one long ass afternoon :) It's extremely rare for me to play a game start-to-finish, I get bored too easily...

Re:Why?

[billcopc]

Whoever modded the parent Flamebait needs to take their meds. Disagreement != Flamebait.

I have to agree, the original Unreal was kind of disappointing. Nice graphics but where was the plot ? There was that interesting moment, right at the start, when you enter some building and the lights go dark one-by-one... then you kill whatever the hell was in there, and go back outside to shoot at more buff aliens throughout the entire game. It felt very forced and poorly thought out. But it was damned gorgeous for the time and it made the Voodoo do things never seen before.

Re:Why?

[RulerOf]

I typed a very thorough response to what you said as to why I hate the game so much, and then I closed the tab due to a sudden outbreak of nooblet syndrome. I agree with you on the gameplay points you brought up; I did like those. However, many things were underutilized, or suffered from being horrible gameplay mechanics. For example, the suit was horribly underutilized, and the escort mission was incredibly annoying and stupid, among other things.

Point is though, after I played that game, I regretted wasting my time on Crysis... and that's coming from someone who saw Alexander in the theater. And that movie, though a waste of time, did have a nude Rosario Dawson and *was* comically bad near the end.

Re:Why?

[Anonymous Coward]

For me, the environment in Oblivion was inviting to explore. Plenty of treasures and unique spots to be found by just wandering around.

I found a new cave and excitedly plunged in. Oddly the cave looked very much like all the others I had explored.

Re:Why?

[ardor]

A) is true, and the reason I talked about a hybrid - the benefits in (A) come from secondary ray effects.

B) is true, but irrelevant IRL, because games usually put the complexity on the shaders/materials, NOT on the geometry. Highly detailed geometry is painful to edit, and suffers from visible aliasing. It is much more efficient to bake tiny details into texture maps. The one reason why geometry complexity could increase is usally instancing, like, 10000 soldiers, or a forest where the trees near to the camera are all fully rendered (the trees far away are approximated with impostors).

What you missed is that rasterizers can be very cache friendly with ease, whereas raytracers require non-trivial changes - and especially for complex scenes, cache coherency is VERY important. Also, for the opaque diffuse surfaces rasterization IS an optimization, as said. This means that it will always be more efficient than raytracing in this narrow domain. Many years will pass before rendering diffuse opaque surfaces using raytracing is only slightly slower than by using rasterization. Until then, I have the option of either rendering a room with 100% raytracing, or an entire city using a hybrid.

No Caps?

[Steauengeglase]

No screen captures, just pics taken with a camera? Um, Ok.

Re:No Caps?

[ILuvRamen]

Maybe they didn't have a spare core to process that hehehe. Seriously, this technology is useless. There's just recently been a graphics card that can anti-alias 8x in Oblivion at full resolution and keep 60 FPS. Why don't they work on real features instead of something that's just going to become another DirectX 10 in that nobody cares about it cuz it's a pointless, horrible upgrade to make (Vista and a new, probably expensive card, no thanks!)

Huh

[gadzook33]

With enemeies like that, who needs frames.

Poor quality textures

[ulash]

Is there a reason why the textures (especially those on the chopper but in general as well) look extremely poor quality? Was this done on purpose to increase the FPS?

Re:Poor quality textures

[Anonymous Coward]

No, they are simply colour mapping. They are not using pixel shaders.

Re:Poor quality textures

[JimboFBX]

So are pixel shaders compatible with ray tracing?

Re:Poor quality textures

[91degrees]

Yes, or at least something equivalent. The problem is that this sort of thing doesn't work as well without dedicated hardware. A SIMD shader unit can operate on several pixels at a time whereas each CPU core can only operate on one colour channel of a pixel at a time.

Re:Poor quality textures

[suso]

Sigh. Because raytracing can't catch up with real time texture mapping.

I don't know what this fad is with trying to do real time raytracing recently. As I've said before, so what. It will always look like ray traced graphics from 10-15 years ago, not like that of today. But people don't listen and a lot of time is wasted.

I guess its only interesting now because its possible where as it wasn't quite as possible before.

No Wonder Microsoft Dumped Intel For The 360

[Anonymous Coward]

What an amazingly unimpressive demo or Larrabee sucks as much as people have been claiming.

Enemey Territory

[Pennidren]

Not as dangerous as Enema Territory to be sure, but probably more fun.

Re:Enemey Territory

[Anonymous Coward]

More fun than Enema Territory? Preposterous.

Re:Enemey Territory

[b1t r0t]

Don't you mean "posteriorous"?

Congratulations Intel!

[lowlymarine]

Intel, you've done what only you can do! With $6,000 worth of top-of-the-line processors, you've almost duplicated the performance of a $60 RADEON 2400XT. Except with better reflections. Although even pixel-perfect reflections of crappy textures are, by definition, crappy textures. You're going to crush nVidia any day! I feel it, keep smack-talking!

Re:Congratulations Intel!

[Joe The Dragon]

not only that ati on board video is a lot better then the intel one + is can use side port ram.

Re:Congratulations Intel!

[bigtangringo]

I predict that you'll eat those words one day.

Re:Congratulations Intel!

[Cathoderoytube]

It's just a first step. Give them some time and I'm sure they'll be producing much more impressive stuff. Though I don't really give a rats ass about the applications real time raytracing has for video games. I'm more interested in what it can do for 3D graphics and animation. As it stands now in 3D you have to render everything out to see what it looks like properly lit. It'd mainly be a workflow improvement, but it'd be a welcome one. It's extremely annoying and time consuming to render out a test image that can take 10 minutes just to see how everything looks. That would also cleave through final render times. As it stands now with most projects it can take weeks or even months to render everything out. In theory with this a single desktop computer could be on par with a render farm. Suddenly all those jerks over at CORE won't be so smug.

Re:Congratulations Intel!

[Nimsoft]

Intel are not claiming this is ready for the mainstream, but merely demonstrating they are getting closer!

What you have to remember is this was done entirely on the CPU. Try running a game with simililar visual quality and resolution on a standard software renderer and you'd be lucky to get 1fps, probably much lower (and that's with no fancy filtering or reflections).

Raytracing has the ability to produce stunning images on a future generation of CPUs alone, or in the closer future on custom hardware and it will be exactly this kind of research that will get us there - I for one am glad someone is doing it.

I'll bet on it that expensive graphics cards that are soon out of date will become a smaller and smaller niche just like the sound card, and I personally can't wait :)

Re:Congratulations Intel!

[urbanriot]

Intel, you've done what only you can do! With $6,000 worth of top-of-the-line processors, you've almost duplicated the performance of a $60 RADEON 2400XT. No they haven't. They're attempting to implement real-time ray tracing. A $60 video card can not do this.

Re:Congratulations Intel!

[Anonymous Coward]

A $200 video card should get close to the perf intel is getting here w/ ray tracing.

Re:Congratulations Intel!

[Darkness404]

No but here is the thing, it will run 3-D on Linux out of the box. Unlike that ATI card and nVidia, but at least the nVidia binary drivers will actually run and not cause Linux to panic...

Re:Congratulations Intel!

[Cecil]

Do you even know what ray-tracing is? I agree, this isn't in the same league as a Radeon 2400XT. In fact, it's not even playing the same sport.

Re:Congratulations Intel!

[Oktober Sunset]

And I predict you will need to mash the words up for him, because by then he will be a toothless old man.

Re:Congratulations Intel!

[grumbel]

One day maybe, but not today and not tomorrow. Raytracing still has a very long way to go. And even if it ever happens its far from clear that it will happen in this simplistic form. When one is rendering normal polygonal 3D models raytracing simply has very few practical advantages and most games do fine without shiny spheres. The ability to render more polygons is also not really all that important, normal mapping already gets a 5'000'000 triangle model down to a 10'000 triangle model with a very small loss of detail. And for most stuff you don't really have the time or money to model all that detail in the first place.

In the end one also has to keep in mind that the picture that appears on the screen isn't just a simple straight forward rendering these days. A lot of post processing is applied to it, often on the depth-buffer directly and thus completly outside of classic raytracing or rasterization domains.

Raytracing simply isn't a magic bullet. If we have hardware that is fast for raytracing, we will surely find some use for it, but we won't just throw out our GPUs because of that.

And lets not forget backward compability, even if raytracing would be the magic bullet, you wouldn't want to lose backward compability to pretty much all games created in the last 10 years.

Re:Congratulations Intel!

[grumbel]

but merely demonstrating they are getting closer! But are they really getting closer? I have seen demos of shiny raytraced spheres rendered a decade ago on the Amiga, seening those spheres now again in Doom or QuakeWars isn't exactly impressive. Those spheres are pretty much the "Hello World" of raytracing world. I still miss a demonstration of something impressive that my current graphics card couldn't do, shiny spheres alone just don't do it here.

Meh

[saikou]

I know it's all computationally intensive and impressive in that aspect, but pictures in the article don't really look much better than your average videogame. Same triangular shapes, ugly, clearly "rendered" landscapes.

I wonder if anyone tried to do hardware acceleration with, say, splines or something other than triangles.

Re:Meh

[JustinOpinion]

I've always wanted a realtime graphics engine based on something like the POV-ray ray-tracer (or other procedural modeling). The POV-ray syntax is all "exact". Rather than approximating shapes using subdivision into triangles, exact shapes are created by specifying things like "spheres" or "cylinder" or unions, intersections, and differences thereof. More complex objects can be specified by arbitrary mathematical equations, and complex sequences of operations (e.g. take a spline, sweep it along a path, intersect it with another shape, apply a certain matrix transform, ...). Having done some modeling both ways, I much prefer the "exactness" of procedural definitions, rather than approximation. (I inevitably wish I could go back and add resolution to a triangulation, but that isn't easy to do properly.)

The neat thing is that the resulting objects (if properly defined) have "infinite" detail. The roughness on a surface, for instance, can be based on a noise function, so you can zoom into it without ever seeing triangulation or other artifacts.

The obvious downside is that the computation here is intensive. Objects can be arbitrarily complicated. Calculating the intersection of a ray with a mathematically-defined surface involves very complex calculations. Rendering POV-ray scenes on modern hardware, for instance, can take minutes to days (depending on complexity).

One upside is that the rendering can be tuned to available resources. On older hardware, the number of light-sources (or the intersection accuracy, etc.) can be reduced. This would mean that video game graphics would get arbitrarily "better and better" on newer hardware, without any need for someone to change the code. Having said all this... I think our hardware is not yet powerful enough to make this kind of thing practical. (There are some neat examples that have been coded, but as a general technique we're not there yet.)

Re:Meh

[Khyber]

Softy3D was a modeling program that used splines and spheres, nothign else.

Re:Meh

[larry bagina]

I guess you don't realize it, but POV-Ray is not FREE (as in GPL) software. While it is free (as in beer), users are denied fundamental human rights that the GPL enforces. Using and advocating POV-Ray is counterproductive.

Re:Meh

[Anonymous Coward]

fundamental human rights that the GPL enforces I literally laughed out loud. Let me guess, you still believe in Jesus and Santa Claus too?

Re:Meh

[X_Bones]

I know it's all computationally intensive and impressive in that aspect, but pictures in the article don't really look much better than your average videogame. Same triangular shapes, ugly, clearly "rendered" landscapes.

That's because the demo wasn't about finer geometry or higher-res textures or better bumpmaps. It was about the reflection and refraction bonuses you get with ray-tracing that you don't with rasterization, which won't really affect anything you were talking about.

Re:Meh

[vertigoCiel]

The other obvious downside is that it is hard to model organic objects using such techniques. Ever tried to model a human face procedurally?

All of the earliest 3D renderers (the stuff Lucas Arts developed for Star Wars, whatever they used for the Light Cycles in Tron) used combinations of geometrical primitives as models. They moved to tessellation because most of the things you want to model in movies and video games do not lend themselves to procedural definition very well.

Re:Meh

[grumbel]

I've always wanted a realtime graphics engine based on something like the POV-ray ray-tracer CSG can work with a rasterizer, see for example Ensemblist. The problem with CSG however is that it just isn't very practical for game modeling, its nice for industrial work where you want to have exactness, but not for games where you want it pretty and want it fast. And of course CSG is rather useless when you want to model something organic like a human or a monster.

Re:Meh

[Spikeles]

This would mean that video game graphics would get arbitrarily "better and better" on newer hardware, without any need for someone to change the code. A few games have already applied this idea. Quake 3(and i assume 4 as well as the quake3 derivatives) use Curved surfaces, Messiah i believe used models that had millions of polygons, and the game engine automatically applied LOD when rendering them to keep a good FPS.

Morrowind supports Continuous Tessellation and N-Patch rendering if you have an ATI card, and turn it on using tweaking tools.

With newer and bigger storage there really is no reason NOT to ship with full resolution images and model definitions/meshes on the DVD, and then scale them down to fit the machine during run-time, it's just a shame more developers don't consider doing this.

Re:Meh

[ccr]

The POV-Ray version 4.0 has been "informally decided" to be released under GNU GPLv3 by the devteam. Admittably, the release is probably far off in future, since it'll be almost complete rewrite.

Re:Meh

[Anonymous Coward]

Couldn't have anything to do with the original code base of POV-Ray predating the GPL, could it?

If you really want POV-Ray to be open source, do the open source thing and write some code for the GPL release, or track down all of the original contributors and get their permission.

Why is it, when people complain about "open source" they are told to "Write a patch yourself." but when they complain about proprietary stuff they don't just write it themselves?

Re:Meh

[MidnightBrewer]

Mathematically precise, yes, but also incredibly impractical. The problem is that modeling anything sufficiently complex (like, say, a human body) procedurally is much more readily done using spline-based modeling AKA NURBs than trying to type it in by hand.

POV-Ray supports mesh-based objects as well, but then you lose exactly the advantage that you're looking for.

As far as game-engine graphics go, though, mesh-based modeling is still probably the easiest to optimize, when you can calculate exactly how many triangles you're pushing through, rather than mathematical formulae.

POV-Ray excels for creating pre-rendered animations. In that case, render engines such as LightWave, Renderman and mental ray all support subpixel tesselation of NURB objects, thus giving the user practically infinite detail within an easily managed modeling interface.

Re:Meh

[Yetihehe]

POV ray also supports bicubic patches, which you can use to approximate NURBS. It's like tesselation, but output is still mathematically precise. Plus bicubic patches are a LOT easier to intersect than nurbs. You can also make human body this way without triangles.

Re:Meh

[samael]

I'd like to know how a sphere is any more "exact" than a triangle. They're both arbitrary shapes that you're building other shapes out of...

Re:Meh

[KDR_11k]

Increasing accuracy on mathematical shapes like that is pointless as you rarely if ever want a perfect mathematical shape. Even NURBS are usually just an approximation of a real shape so it doesn't really matter if they're perfectly round or not, the surface they depict isn't either.

Re:Meh

[robthebloke]

> I wonder if anyone tried to do hardware acceleration with, say, splines or something other than triangles.

yes, but only 2 graphics cards to date. The glu nurbs interface, and openGL evaluators have been around for years, but only one card ever implemented them in hardware (can't recall which one).

Then came NV evaluators which were launched with a big fanfare for the Geforce 3, but was then the support was quietly dropped from the drivers a couple of revisions later.

The problem is that hardware support for spline tessellation is basically pointless. The heavy part of the computation is in generating the blending co-efficients for the curve computation, and in both of the above extensions you have to compute those every frame. A simple caching scheme for the co-efficients will always beat any hardware accelerated spline implementation. (i.e. cache the co-efficients periodically so that surface LOD does not change every frame. The remainder of the computation can generally then be put in a vertex/geometry shader).

Of course, there are bigger problems with splines.

1) Artists really don't like modelling in them. 2) They add detail into the wrong places, which ends up generating magnitudes more polys than if you'd just modelled everything in polygons.

Subdiv surfaces are imho a better way to add detail into geometry, but caching is far harder to do when compared with splines. Dynamic surfaces such as character skins are basically out of the question as a result.

The best approach i've seen to date are ATI's PN triangles, which subdivides individual triangles and quads using the face normals and vertex positions. This is pretty trivial to insert into a geometry shader, and gives decent (read: occasionally odd) results. So far this is one of the better ways to deal with the fact that a GPU can only really operate on a triangle at a time (without resorting to horrible hacks like passing data in the form of textures, and nasty render output data to texture approaches).

Re:Meh

[cheesybagel]

AFAIK Star Wars IV used hand made models for the ships and everything was done with traditional film effects techniques. Tron did use ray-tracing since the graphics were done using computer effects by MAGI.

Re:Meh

[smoker2]

A friend of mine is an artist, and it surprised me one day when he said "if it looks right, it is right". Being particularly anal about attention to detail in almost anything I do, that just seemed wrong. But I've tried it out, and in the right circumstances, it's true. The world we see with living eyes doesn't get the same level of scrutiny that any artificial environment does. And I think we go over the top with the look rather than the feel too much these days. I still prefer UT before the flashy skins and fancy maps. Playing CTF in teams of 20 or 30, you didn't have *time* to notice the bloody reflections. A wall is a wall, and that's all it needs to be when you hide behind it. The rest is in your head.
And you certainly don't need the *look* making the *feel* slower.
I don't see that many people in the real world, studying puddles closely to see the ripple effects on the reflection from different angles. Unless they're stoned.

Re:Meh

[vertigoCiel]

Star Wars VI used computer graphics for the hologram of the Death Star in the briefing room of Ackbar's frigate. The Graphics Group division of Lucasfilm would go on to develop the CGI effects for Star Trek II: Wrath of Khan, before being bought by Steve Jobs and becoming Pixar.

Why do i feel that ...

[Skal Tura]

Most people don't understand the beauty of ray tracing .... oh wait a minute, most people are DUMB! that's right >;D

Seeing these comments reflects very well the average human intellect about a subject before talking about it.

Then what's so special in ray tracing versus rasterization?

It's actual real world based mimickery. Ray tracing mimicks how real world works.

Ask yourself would you prefer physics to correlate to real world physics, or something quickly around the corner which is something like that but not quite? That's the difference between rasterization and ray tracing. Rasterization comes close, but never is quite the real thing, while ray tracing works to replicate real world physics of light.

Yes, it actually is physics calculations, in this case, the physics of light and visualizing it.

Then there's things like radiosity tracing aswell ...

With Raytracing / radiosity tracing just provide enough computational power, and you can make it look real, like an photo. With rasterization you can't do that, to rasterize, and make proper looking shaders, they never get quite there, but even doing the shaders, you need to think about how the real world works first, how light travels, how it interacts with what it hits, in other words how the rays interact with objects.

Rasterization is JUST a cheap trick to make it look something like that, nothing else.

It's a bit like comparing veggie soy "meat" to the real thing It's something like it, but not really, just a "cheap trick".

Now they used 16 cores to that, 4 cpus. Moore's law, CPU power doubles every 18 months, that means JUST 3 years before high end home users can enjoy something like that. in 4.5 years it becomes common, 6 years and it's every day. Be prepared for the coming of Ray Tracing. In about 7.5 to 9 years all graphics are probably ray tracing.

And that all without any software, or architectural advantages, it'll probably happer sooner with Intel making a hard effort to make it reality.

Re:Why do i feel that ...

[ulash]

I think you are missing the point a little here by ignoring one of the points people made before you. It's not like Intel has demonstrated amazing graphics using raytracing. Noone would want graphics like this in their games even today let alone in 4 years. Iff Intel can come up with a demonstration using raytracing that actually looks *better* than the "cheap trick" version then we can start counting down the days for when we will get that capability at home for an affordable price.

Re:Why do i feel that ...

[CastrTroy]

It's a bit like comparing veggie soy "meat" to the real thing It's something like it, but not really, just a "cheap trick".

I try not to think of veggie meat and tofu as a replacement for meat, but rather a whole different food that stands completely on it's own. It's quite different from meat, and not a replacement. But if you throw away the whole idea of a meat substitute away, it actually tastes pretty good. Not only that, it's a good source of protein, and extremely low fat compared to other sources of protein. Once you get over the fact that it's tofu, and obviously doesn't taste like meat, and isn't supposed to be meat, then you shouldn't have a problem eating it. If you don't like the taste/texture of tofu, you aren't preparing it right.

Re:Why do i feel that ...

[nuzak]

Psst. It's all tricks. That's not a real helicopter, and it'll never fly out of your screen. Value judgments on what's more "real" are just plain silly.

Re:Why do i feel that ...

[Anonymous Coward]

Most people don't understand Moore's Law

http://arstechnica.com/articles/paedia/cpu/moore.ars

Re:Why do i feel that ...

[Anonymous Coward]

Now they used 16 cores to that, 4 cpus. Moore's law, CPU power doubles every 18 months, that means JUST 3 years before high end home users can enjoy something like that. Moore's law describes an important trend in the history of computer hardware: that the number of transistors that can be inexpensively placed on an integrated circuit is increasing exponentially, doubling approximately every two years.

It has nothing to do with CPU power. What was all that you were saying about most people and the average human intellect?

Re:Why do i feel that ...

[Anonymous Coward]

Your right, ray tracing is conceptually very elegant and rasterization is just a work-around. However, its a hack that works and needs much less computation. Moore's law is an easy excuse, but CPU speed is rapidly hitting a plateau, and while you can always throw more transistors at it since most types of video rendering is very parallel in nature, this still means more power requirements and hardware cost. Not that a chip maker would mind selling more product... Gosh, who is pushing this technology? Id love to see ray tracing become mainstream though, but i suspect it would need to be on a more efficient type of architecture.

Re:Why do i feel that ...

[ArbitraryConstant]

Rasterization is JUST a cheap trick to make it look something like that, nothing else. Right, but it's really fast. It's faster than ray tracing is good.

If you're waiting for humans to get rid of fast approximations when they're good enough, I hope you're patient.

Re:Why do i feel that ...

[Stan Vassilev]

Ray tracing mimicks how real world works.

Raytracing doesn't mimic how real world works. In fact it does exactly the opposite of what happens in real world. In real world you have bazillions of light particles, doubling also as waves, shoot out of many area light sources and bounce/be absorbed by objects around them.

Whatever photons end up hitting your retina, is what you see.

Raytracing instead shoots a ray out of your (virtual) retina straight forward to the scene and may refract/reflect off objects, until it's "absorbed" (means, hits a surface where refraction/reflection isn't calculated).

Rendering a single frame of 3D as it is in the "real world" (with just a fraction of the rays) would mean days on even the fastest hardware out there.

What raytracing gives you is sharp reflections, refractions and shadows, while introducing a bunch of other limitations on the rendering that rasterization doesn't have. It also can't do soft shadows, reflections, refractions, efficiently, nor subsurface scattering, or radiosity.

Best models for rendering in the future will likely be hybrid models similar to what is now used in professional renderers by movie studios. But then again, it's a game, who cares about mathematicaly accurate reflections, when you can fake it close enough with reflection/refraction maps in a fraction of the processing time.

Re:Why do i feel that ...

[DetpackJump]

Wha? There's some decent discussion in this thread, then you come along, call everyone dumb, then toss some stupidly basic definitions at us. I feel like you just walked into a senior level college math class and started to explain high school algebra to everyone.

Re:Why do i feel that ...

[Dubbie99]

His post was the first one on this thread that actually seemed to have some sense in it. The rest of the posts are from amateurs who think that fiddling with POV ray back in the 90's qualifies them as an expert.

Re:Why do i feel that ...

[Anonymous Coward]

Whatever photons end up hitting your retina, is what you see.

Quantum mechanics (aka "God's PVS") begs to differ.

Re:Why do i feel that ...

[HeronBlademaster]

Noone would want ray-trace-quality graphics in their games? How many gamers do you know?

I'm a gamer, and I would love ray-traced games. Half-Life 2 looked good in its day, as did Half-Life several years before it, but we're getting to the point where if we want things to look more realistic (and believe me, gamers do), we're going to need to switch to ray tracing.

Re:Why do i feel that ...

[Anonymous Coward]

You're the perfect manifestation of a person who knows only a little about the subject and talks about it. And I'm confident that you lack the basic understanding of both raytracing and radiosity. Both are crude approximations toward the rendering equation, but they try to capture different aspects of it. In practice, raytracing is better at delivering reflection, refraction or anything that you can do by tracing a ray backward from the eye, but radiosity does global illumination a lot more efficiently than raytracing. That said, raytracing does scale better than radiosity.

Re:Why do i feel that ...

[Anonymous Coward]

You eyes shoot out photons and determine what you see by what they refract and reflect off and what light sources they happen to hit?

Bizarre...

Re:Why do i feel that ...

[blueg3]

For classical optics, modeling the scene in either direction (castings rays from lightsources and only counting ones that hit the viewer vs. casting rays from the viewer and only counting ones that hit lightsources) is valid.

I assume you didn't mean for "efficiently" to be an item in your list, which is the way you wrote it, but raytracers can do all of those things. (I'll make no claims about efficiency.)

"But then again, it's a game, who cares about mathematicaly accurate reflections, when you can fake it close enough with reflection/refraction maps in a fraction of the processing time."

That argument is no more valid that if you say "it's just a game, why don't you just do raycasting, which takes a fraction of the processing time". "Faking it close enough" isn't close enough; it's obvious that you're faking it, and it requires that you either live with it or design your game to minimize the impact of faking it.

Re:Why do i feel that ...

[Anonymous Coward]

I find it funny that you spend several paragraphs talking about "reality" and then bring up "moore's law"...

Re:Why do i feel that ...

[Anonymous Coward]

Ask yourself would you prefer physics to correlate to real world physics, or something quickly around the corner which is something like that but not quite? That's the difference between rasterization and ray tracing. Rasterization comes close, but never is quite the real thing, while ray tracing works to replicate real world physics of light. Can your beloved ray tracing correctly handle a diffraction grating? How about circular polarization and birefringent materials?

If not, then it isn't "real world physics" but some kind of limit (probably something approximating the classical ray-optical limit).

(Note: you can do all of the above with ray tracing, but it'll slow things down quite significantly.)

Re:Why do i feel that ...

[simonv]

Well, it's exactly like the real world works if you only change the reference of time. Who are you to say that we don't emit photons from our eyeballs into light bulbs and local nuclear furnaces!?

Re:Why do i feel that ...

[ulash]

I apologize - somehow I am having trouble expressing my idea here it seems. What I did not say was that gamers would *never* want ray traced games. What I did say however was that *until there exists a ray traced system that produces an experience better than the rasterized version* the gamers will not want the raytraced solution just because they can use their CPUs instead of a graphics card.

Re:Why do i feel that ...

[grumbel]

Noone would want ray-trace-quality graphics in their games? Realtime raytracing as demonstrated here is good for shiny spheres and little else. I don't know about you, but I have rather little use for shiny spheres in my games and if I do a little fake environment map does the job quite well.

How realistic a rendering looks simply has nothing to do if you rasterize or raytrace, its just a different approach to get pixels on the screen. Global illumination and all that stuff is what matters when you want a really realistic look, but you don't get that for free with either rasterization or raytracing. And when it comes to graphical progress in the last years, a lot of stuff is happening on the framebuffer itself, and is as such completly independed of the way your geometry rendering works.

Anyway, I'd be impressed when they manage to render something that looks *better* then current games, not when they manage to render old games with a crappier look, less fps and a few shiny spheres added.

Raytracing simply isn't a magic bullet.

Re:Why do i feel that ...

[bh_doc]

But: Moore's law -> more cores on a single CPU -> more power.

Re:Why do i feel that ...

[cjHopman]

Most people don't understand the beauty of photon tracing .... oh wait a minute, most people are DUMB! that's right >;D

Seeing these comments reflects very well the average human intellect about a subject before talking about it.

Then what's so special in photon tracing versus ray tracing?

It's actual real world based mimickery. Photon tracing mimicks how real world works.

Ask yourself would you prefer physics to correlate to real world physics, or something quickly around the corner which is something like that but not quite? That's the difference between photon tracing and ray tracing. Ray tracing comes close, but never is quite the real thing, while photon tracing works to replicate real world physics of light.

Yes, it actually is physics calculations, in this case, the physics of light and visualizing it.

Then there's things like (insert other random rendering method here) aswell ...

With photon tracing just provide enough computational power, and you can make it look real, like an photo. With ray tracing you can't do that you need to think about how the real world works first, how light travels, how it interacts with what it hits, in other words how the rays interact with objects.

Ray tracing is JUST a cheap (relative to photon tracing, that is) trick to make it look something like that, nothing else.

It's a bit like comparing veggie soy "meat" to the real thing It's something like it, but not really, just a "cheap trick".

  ---

And then we get into some misinterpretation of moore's law.

To be honest, using ray tracing is just giving up one approximation for another, that's what computer graphics is currently about.

Back in reality though, yes ray tracing can make some things look more realistic than rasterization, usually at a significant performance cost. My personal view is that there will be only very little ray tracing used in games for some time. There is not currently spare performance to go around, the performance of rasterization is still a major bottleneck for games, the idea that there will be this change to a less efficient system seems highly unlikely.

Yet, I would not be surprised to see some slight hybridization occur over the next several years, and eventually ray tracing may become the dominant renderization technique in games... I'd say more like 15 - 25 years though... and then another 20 years and maybe we'll be doing real-time photon tracing, who knows.

Re:Why do i feel that ...

[blahplusplus]

"Most people don't understand the beauty of ray tracing .... oh wait a minute, most people are DUMB! that's right >;D

Seeing these comments reflects very well the average human intellect about a subject before talking about it."

Tell that to NVIDIA -- http://games.slashdot.org/article.pl?sid=08/03/07/1659250

Re:Why do i feel that ...

[HeronBlademaster]

I'll agree with that. Gamers largely won't care whether things are ray-traced or rasterized; they'll just care whether it looks nice. I think ray-tracing has the potential to look nicer, if the hardware can do it in real-time... I guess we'll see.

Re:Why do i feel that ...

[HeronBlademaster]

Very true. Like I said to ulash just a moment ago, I don't think gamers will care whether things are ray-traced or rasterized, but I think that ray-tracing has the potential to look nicer if the hardware can do it.

Re:Why do i feel that ...

[Anonymous Coward]

I am a 3d-artist, and i'm using raytracing-derived technology every week to render soft shadows, ambient occlusion, etc. The raytracing of the 1990's may not be very exciting (that's basically what we are watching in real-time here), but double the performance of the chips a couple of times more, and we can do some really cool stuff.

Re:Why do i feel that ...

[religious freak]

You are missing the point. If you've just invented the internal combustion engine -- your invention still has value, even if you haven't mounted it into a car yet.

This tech facilitates major graphics advancements. It's a POC

Re:Why do i feel that ...

[Xzzy]

I've had it explained that the real benefit ray tracing allows is that render time does not increase with geometry. It means you could model every little brick and bit of mortar, down to the tiniest scratch, and it wouldn't slow the game down (up to the limit of the number of vertices you can store in memory).

Image quality is a side effect.. except for reflections, you can get 90% of the image quality out of modern 3D engine tricks that you could out of a ray tracer.

Sure these days, it mostly revolves around making everything brown and slapping on some bloom, but that's more of a phase artists are going through than the limit of the technology.

Re:Why do i feel that ...

[Yetihehe]

Well, yeah.
NVIDIA Comments on Ray Tracing and Rasterization Debate

Ray tracing may be the future of rendering--it's definitely part of the future at least. There is an old joke that goes "Ray tracing is the technology of the future and it always will be!". I don't think that's completely true, but I do believe that ray tracing is not the answer--it's part of an answer. Furthermore, I believe that the C/C++ language programming interfaces for GPU computing are useful for programming ray tracing to run on GPUs. Over time, I expect the graphics APIs will evolve to embrace ray tracing as part of the 3D graphics "bag of tricks" for making images.

And it's not like they are against raytracing, they are acquiring RayScale, raytracing startup. So probably nvidia will use raytracing with rasterisation, IMHO the best approach.

Re:Why do i feel that ...

[Yetihehe]

I think you could be referring to this comic

Re:Why do i feel that ...

[Anonymous Coward]

I bet in 10 years gamers will complain theyÂd like "real soft shadows", as this may give them an edge on close encounters in the Dread Towers of Spartan or something..

Myself, I quit gaming, but it will be interesting to buy a gamer-PC someday just to see whatÂs possible tomorrow..

Re:Why do i feel that ...

[Skal Tura]

I meant those which are sold as meat substitute things.

I eat soy almost daily :)

Oh, and you could say i'm pretty much vegan, sometimes rarely i treat myself with an big mac or something with animal products in it, but 95% of time i eat like a vegan :)

Re:Why do i feel that ...

[Skal Tura]

and you are merely a troll.

Who ever expected that stuff from screen to materialize? The REAL thing in here is the representation looking real. With raytracing you can achieve photorealistic graphics.

Re:Why do i feel that ...

[Skal Tura]

And you missed the point of realistic reflections and refractions :D

I think they were still missing radiosity as well. The demo is proof of concept, intended to show off only reflections and refractions. Google or something for raytraced graphics which have everything included.

Take the demo as it is: Proof of concept. When have you seen proof of concept being a finished product? :)

Re:Why do i feel that ...

[Skal Tura]

well, yes it's not 100% exactly how real world works, in raytracing we leave out the stuff we don't need, thus you could say direction is opposite than in real world, and you could say we "back trace" how the light bounced around :)

hybrid model is still a cheap trick, and faked to somewhere there. However, how movie studios do it has it's perks as well.

The direction is full ray tracing and radiosity tracing. Everything is doable, put brilliant enough mind into the task, and the world doesn't live WITHIN the box anymore. That's a problem with a lot of people, they can't think outside of the box, those who can make things realize, things like real time photorealistic raytracing.

Real time photorealistic ray tracing has been demonstrated times and times over, just look at Demoscene, it's old news, what's new is making the scenes complex like in a game, and maintaining high FPS.

and like someone else said, you can do all of that in raytracing.

Look at POV-Ray for example :)

Re:Why do i feel that ...

[Skal Tura]

I think you didn't look at the other comments :)

Those comments were "raytracing sucks", "only intel can throw in 6,000$ to do $60 radeon's job", and general trolling like that.

And i didn't say dumb, it's your own low self-esteem if you think i said you are dumb. I merely said about the level of knowledge about a subject and human behavior. A human is completely comfortable about talking a subject they know shit about.

I don't even claim to know much about ray tracing, but i do know some of the basics :) Enough to know most of the comments were utter bullshit :D

Re:Why do i feel that ...

[Skal Tura]

Yeah, a lot of today's graphics revolves around shiny and glow.

But rasterization does not come to 90% of the image quality.

Actually increased geometry slows somewhat down, but by far not as much as rasterization. Checking what a ray hits becomes huge, and there probably isn't atleast any easy way to skip 99% of checking what the ray hits, and in what angle is that object.
More geometry, more things to check for the rays to hit.
But the increase in computational power needed by increase in complexity isn't nearly as huge as in rasterization, raytracing doesn't revolve around scene complexity for it's speed, mostly on amount of rays.

The increased complexity performance impact is considered neglible, and i'm sure there's tons of ways to optimize that. Unoptimized code would have a significant impact on that area.

and that's definitely one of the perks of raytracing, which adds to the main perk of ray tracing: realistic graphics.

Re:Why do i feel that ...

[ErikZ]

Or more likely, in 9 years we'll still be using current methods (but more advanced) leaving plenty of CPU power to simulate other physics, and run the AI, and address other things.

They're trying to turn a CPU into a GPU, and it's ridiculous. I already have a GPU.

Re:Why do i feel that ...

[CastrTroy]

I eat a lot of tofu, soy, and even a lot of those veggie meat products. The veggie meat products like salami and pepperoni don't taste the same as the meat based ones, but they are pretty good, and considering the nutritional differences, I'd much rather eat veggie pepperoni and be healthier, than eat the meat based ones and be unhealthy. I'm not vegetarian by any means. However, in my kitchen, we only do meat about 2-3 times a week, at dinner, and otherwise meat is almost non-existent in my diet. People would be a lot healthier in general if we just got the idea out of our heads that we have to have large portions of meat at every meal.

Re:Why do i feel that ...

[grm_wnr]

It means you could model every little brick and bit of mortar, down to the tiniest scratch, and it wouldn't slow the game down (up to the limit of the number of vertices you can store in memory). It wouldn't slow down rendering, but it would certainly slow down development time; someone has to model all those bricks, you know. It's a much more feasible approach to just get a good programmer to write some creative pixel shaders to fake details, especially for static, decorative geometry. Raytracing is a geek's wet dream, because geeks like accurate stuff, but does not autmatically translate to a better look. Getting more bling by cutting a few corners will win over people who don't know the difference between rasterizing any day. I suppose we will get a couple of games that mostly consist of reflective spheres though - kind of like the stencil shadow demo that was Doom 3.

Re:Why do i feel that ...

[KDR_11k]

I've had it explained that the real benefit ray tracing allows is that render time does not increase with geometry.

I think it makes the growth from geometry mostly neglible but it's not zero and if you add too much you will see an effect.

Re:Why do i feel that ...

[RiotingPacifist]

Yeah all these crappy games with their Newtonian physics, when I run time doesn't even slow down, whats that about! stupid approximations!!!

Re:Why do i feel that ...

[RiotingPacifist]

Prove that a photon that doesnt hit a detector (e.g your eye) actually exists.

Re:Why do i feel that ...

[Anonymous Coward]

what chu talkin about willis? Did you mean to say that ray tracing mimicks the real world while ray casting does not?

Re:Why do i feel that ...

[hackstraw]

Raytracing doesn't mimic how real world works. In fact it does exactly the opposite of what happens in real world.

Ever studied how the human eyeball works?

Bassackwards is an understatement.

Re:Why do i feel that ...

[nuzak]

No, your post read like the standard haughty scoffing rant at how inherently INFERIOR a lesser thing like rasterization was compared to the true REALITY of raytracing. And raytracing is just as much a simulation, and the only thing that matters in the end is how good it looks.

The lack of any shaders running on that raytraced example made it actually look worse. Nice reflection effects on the cockpit glass though. Maybe we'll see raytracing done as a shader.

Re:Why do i feel that ...

[cheesybagel]

Healthier bullshit. All vegetarians I know are not healthier. Quite the opposite in fact. I only need to look at their pale grayish skin and anemic physical condition. That says everything.

I have eaten vegetarian burgers. They are edible, but neither taste like meat nor like anything good. Heck even bread, without anything in it, tastes better.

Re:Why do i feel that ...

[cheesybagel]

Wrong. Raytracing supports soft shadows. Radiosity is just an algorithm that achieves diffuse interreflections. There are others, such as photon mapping. It is easier to program reflections and refractions on a raytracer. The artist does not need to make reflection maps either. Not to mention said maps are hacks which only work for static geometry... There was a time people used 2D vector graphics because memory was expensive. Then 2D bitmapped displays started being used, but 3D scanline renderers were common, because a Z-Buffer used large amounts of memory. Now Z-Buffers are used. As processing power gets cheaper raytracing will be increasingly more used. As memory gets cheaper and scenes more complex, with pixel sized triangles, voxel models will replace triangle based models.

Re:Why do i feel that ...

[Anonymous Coward]

"Faking it close enough" isn't close enough; it's obvious that you're faking it, and it requires that you either live with it or design your game to minimize the impact of faking it. If you've seen any movie with convincing visual effects, I bet you'd be very surprised by how much stuff is "faked".

Re:Why do i feel that ...

[Lost Race]

Movies use forced perspective, which makes a lot of things very easy to fake. Games are a lot more fun when they don't force a perspective, which makes a lot of things much harder to fake. "Fake" in this context means that the simulation fails obviously under some conditions. So far game players have been very tolerant of visual artifacts, paying more attention to things like playability, but that doesn't mean they don't appreciate higher quality graphics. All other things being equal, the game with better graphics sells better.

Re:Why do i feel that ...

[Anonymous Coward]

It also can't do soft shadows, reflections, refractions, efficiently, nor subsurface scattering, or radiosity.

... or absorption, or cogitation, nor excogitation, or immersion, or prepossession, or assimilation, or rumination, or cerebration, or speculation...

Re:Why do i feel that ...

[blueg3]

No, I'm quite familiar with how much "faking it" is used in movies and video games. (I'm also personally familiar with "faking it" in live theater.)

However, faking it requires a lot of work and places restrictions on what you can do. The bar for "faking it" continually increases as technology is developed to enable more convincing effects.

Re:Why do i feel that ...

[Bryan Ischo]

There are very few movies with truly convincing computer-generated special effects. In fact, I can't even think of ONE.

Things typically look too plastic-y, not lighted properly, and don't move convincingly. Often times the effect is 'good enough' that we are willing to suspend disbelief. But computer graphics, when they are the center of focus of a scene, are to this day not truly convincing.

I agree that when they are used in the periphery of a scene they can be hard to spot, but that's mostly because I'm not really paying attention to what's in the periphery. Kind of like how they used an old tennis shoe instead of an x-wing fighter in the edges of some shot or other in Star Wars. It's not that tennis shoes making convincing x-wings, it's just that we don't pay attention to stuff in the corner (unless the movie is *really* boring, and then it's got far worse problems than crummy special effects).

Re:Why do i feel that ...

[Skal Tura]

Shaders are a hack to achieve something which should need to be done using raytracing. Therefore, how are you going to accomplish raytracing as a shader? Isn't that a bit counter productive?
Admittably, i do not know specifics how shaders work, perhaps i should boot up my dev environment, and for the first time in about 10 years code some 3d graphics :)

That demo was a proof of concept, with very little of the effects and capabilities implemented, the few that however were implemented WOW.
Tech demos are often like that.

Furthermore, i hear they did not use textures in usual sense.

In any case, raytracing has for long had waayy superior capabilies compared to rasterization, and rasterization has gotten closer to raytracing capabilities due to increased computational power, however, now we are reaching the point where computational power is reaching the point we just can raytrace.

Rasterization does excellent job in what it's meant to do: A quick workaround to do 3d graphics.

Ray Tracing and Pixel Shaders

[Quabbe]

The images clearly show that they are using simple colour mapping for the textures (especially the helicoptor). What I want to know is, can pixel shaders be used with ray tracing?

Re:Ray Tracing and Pixel Shaders

[Ostsol]

Even with ray-tracing, colour mapping for textures will likely never disappear. You do, after all, need something to indicate to the ray tracer how the surface reflects light. Gazing through POV-Ray's documentation, there appears to be a variety of methods for accomplishing this, with bitmap-based color maps being just one (and perhaps one of the fastest). I believe this answers your question, too. A type of pixel shader is used in ray-tracing to describe materials, but somewhat differently from those used in rasterization-based rendering.

Re:Ray Tracing and Pixel Shaders

[Anonymous Coward]

Yes

Re:Ray Tracing and Pixel Shaders

[Quabbe]

So if it is possible to use pixel shaders (all be it a different paradigm) to define material properties/per pixel normals/lighting etc, why then would they use just standard colour mapping? Why not go the whole hog and use these in the tech demos? This would make the fidelity alot more convincing.

Re:Ray Tracing and Pixel Shaders

[grumbel]

I guess it is simply time. They would need to rewrite all the pixel shaders for their raytracing stuff, they can't just use the the normal ones for the GPUs.

Re:Ray Tracing and Pixel Shaders

[imsabbel]

Because it would make the whole thing 10 times slower, and they know it, so they try to hide it by tons of reflections and water.

Re:Ray Tracing and Pixel Shaders

[Ostsol]

It's for the same reason why colour mapping is still used even in the highest-fidelity graphics in rasterization: speed. Keep in mind that there's no rule saying that ray-tracing must procedurally generate textures and that rasterization-based rendering must use colour maps. Both rendering methods are fully capable of using either technique -- with speed being the only major issue. The advantage to using ray-tracing in a general sense is that it allows for realistic reflections, refractions, and shadows without resorting to the sort of complex trickery involved in rasterization. This is why lightmaps are still used for static shadows.

Re:Ray Tracing and Pixel Shaders

[Anonymous Coward]

Yes, programmable shaders existed in commercial ray-tracing packages long before they were used in video games. I'm guessing they did not use shaders in this demo either to avoid the headache of rewriting all of them (you would not be able to directly use the Cg/HLSL/GLSL/whatever, of course) or to achieve better performance.

After close examination of the screenshots

[Anonymous Coward]

Sorry, being an industry insider, I have to post as an anonymous coward here.

There is no lighting, normal mapping, or material fidelity here. So this is a long way from being the quality of a final product, but it is a good demo and a start in the right direction.

Re:After close examination of the screenshots

[enoz]

So in other words, Half Life 2 still looks better?

That reminds me of the original Half Life 2 trailer (before the first game) that appeared to promise such things as realtime material reflections and refractions.

Re:After close examination of the screenshots

[DragonTHC]

agreed, what good is perfect water and glass if the textures are from 2002?

Re:sucks

[Anonymous Coward]

No, it runs on Linux ergo it cannot suck.

The GNU has spoken.

where's the video

[heroine]

Where's the HD video of the enhanced visual effects in 1280x720 at 14-29 fps?

Reflective spheres

[manekineko2]

Every time ray tracing technology is shown off, I can't help but marvel that the long held dream of games filled with reflective spheres can finally be enabled.

Re:Reflective spheres

[Koiu Lpoi]

Well, perhaps it's not circles, but I can see this technology benifiting CUBE

Re:Reflective spheres

[Siridar]

Don't forget! Those spheres need to be on a chessboard!

Re:Reflective spheres

[CODiNE]

Hey at least no more racing games with caveman wheels. YAAABAAADAAAABADOOOOOO!!!!

Re:Reflective spheres

[drsquare]

Maybe they're building up to the ultimate pinball simulator?

Re:Reflective spheres

[Anonymous Coward]

Dammit, you just linked me to www.superdeluxe.com!
That is ten times worse than any rickroll!

Re:Reflective spheres

[poot_rootbeer]

Every time ray tracing technology is shown off, I can't help but marvel that the long held dream of games filled with reflective spheres can finally be enabled.

Perhaps within the next fifty years, we will be able to authentically recreate the visual experience of playing a pinball machine that was build fifty years ago.

Re:Reflective spheres

[BigJClark]

don't forget, the chessboard must stretch out into infinite!

blur of the line

[Anonymous Coward]

If you look at the crytek engine and how it calculates screen ambient occlusion, then you find there is a mix of ray tracing and raster technology. For instance, its suspected, they simulate a ray tracing but against the depth buffer in a fragment shader to get ambient occlusion.

By the way, please don't publish pictures on t.v. sets. It hurts the eyes.

Ultimately, ray tracing is going to win but not at 13 frames a second.

While I agree with you,

[symbolset]

one - you're about to be deluged by haters. I've said these very things here and the haters are quite enthusiastic. They're also wrong.

two - the other guys aren't standing still either. No doubt the other guys are looking into ray tracing now that the level of tech to support it is coming around.

Raytracing, for the win.

Re:While I agree with you,

[Skal Tura]

Yeah, some haters will try to bash, but hey, that's just normal day of business when you actually are able to think outside of the box, and know a thing or two ;)

Yes, definitely others will come around compete with Intel, and "abuse" the work intel has already done, but Intel is the one pioneering the field, thus with most experience and knowledge, and thus they will most likely do the technically best solutions for years to come. Or so i would think.

I think we are soon going to start FINALLY seeing mroe and more of hybrid rendering. Which i think isn't even that hard to do, and am amazed it's not being done in large scale already. And part by part is more and more rendered by raytracing, until finally fully raytraced.

Why do I feel...

[aztektum]

That so many geeks need to get outside more. Oh wait, they do!

Honestly it isn't like either ray-tracing or raster graphics will cure hunger or disease, so to me "close enough" is good enough. It's just entertainment after all.

pov-ray style rendering

[j1m 5n0w]

I like POV-Ray for many of the same reasons; the syntax is very friendly and the available primitives give a lot of flexibility. Most of the fastest real-time ray tracers just support triangles, though, because it makes them simpler and you don't have the overhead of deciding which ray-intersection function to use with each primitive.

I think a typical game developer isn't likely to care if they can make exact spheres and cones and such; the majority of real-life objects aren't perfect quadrics, and are most conveniently represented as triangle meshes. Some might be interested in the added flexibility, though. It'll be interesting to see if any popular ray tracing APIs are going to support anything besides triangles.

POV-Ray itself isn't likely to be competitive with real-time ray tracers any time soon. It just isn't designed for speed. I've been writing my own ray tracer in haskell (link) that implements a lot of the same primitives as POV-Ray, (spheres, cones, boxes, differences, intersections, discs, triangles, planes) and uses a good, modern acceleration structure (BIH). It's not really any faster than POV-Ray, though, and doesn't support the vast majority of POV-Ray's obscure features.

the CD killed the Algorithm years ago

[EdelFactor19]

and the DVD has been defiling its bloody corpse for nearly a decade now...

why bother putting thought and work into actual gameplay and mechanics when we can just make the game prettier?

sadly all too many gamers are locked into the "better graphics = better game" thinking as well which doesn't help the problem.

Re:the CD killed the Algorithm years ago

[SanityInAnarchy]

and the DVD has been defiling its bloody corpse for nearly a decade now... Maybe the streaming download will revive it?

Minimum framerate?

[Barny]

Quoting min-max isn't what's needed, minimum is all that's required, my rig can, if looking directly at the ground, clock over 1000fps in some games, but that is of course a useless measure of the machine.

Minimum usable framerate is around 35fps, if a fps drops under this, don't bother. Particularly don't bother if its going to cost 10 times the price for one tenth the framerate :P

Re:Minimum framerate?

[Anonymous Coward]

Cinema is not useable, as it is only 24 fps.

Don't forget the *many* games that target 30 fps on consoles... not useable.

It feels better a game at rock-solid 30fps than one the goes from 60 to 30 now-and-then. Many people today play on displays that are fed by a 60Hz display. If you are vsyncing your frame buffer, you better have either rock-solid 30 or rock-solid 60.

I still laugh at the thought of those gaming rigs to play games @180 fps when using a TFT monitor fed by a 60Hz signal.

Re:Minimum framerate?

[Aladrin]

My personal 'usable' lower limit is 15 fps. So everyone is different and telling them not to bother if it doesn't hit your personal limit is pointless.

Still need dedicated hardware.

[argent]

Phillip Slusallek was demonstrating full screen real time raytracing using a custom RPU (raytrace processing unit) in 2005, and that unit was running at less than 100 MHz. For a fraction of the hardware cost of a quad quad core system, you could do real time raytracing with less hardware investment than a modern GPU.

Re:Still need dedicated hardware.

[ceoyoyo]

Actually, I suspect Slusallek's custom ASIC cost a great deal more than a quad core system and certainly more than a modern GPU.

Custom ASIC runs are expensive.

Re:Still need dedicated hardware.

[Rufus211]

You know he's at Nvidia now:
http://www.linkedin.com/in/slusallek

Re:Still need dedicated hardware.

[Yetihehe]

But they are still better than current generation of raytracing accelerators for pc (mainly because the latter don't exist yet).

Re:Still need dedicated hardware.

[argent]

I suspect Slusallek's custom ASIC cost a great deal more than a quad core system and certainly more than a modern GPU.

Sure, making anything in small quantity costs a lot more per unit.

But now what would it cost, per unit, if ATI or nVidia was making RPUs the way they make GPUs? Then it's largely a matter of process and gate count, and the RPU had a fraction of the gates in a modern GPU.

Re:Still need dedicated hardware.

[ceoyoyo]

Is it? Without diving into the journal papers I couldn't find any mention of the resolutions his chip can run in realtime. Nobody wants to play a game with 256x256 graphics even if they are ray traced.

I expect the current interest in ray tracing hardware is due to it approaching a point at which gamer relevant resolutions can be computed in real time for a price that is competitive with raster graphics.

Re:Still need dedicated hardware.

[argent]

Without diving into the journal papers I couldn't find any mention of the resolutions his chip can run in realtime.

At 66 Mhz, about 6 million gates, 512x384, 7.5FPS. I don't have the performance for the later 90 MHz prototype on hand, but even the 66 MHz prototype was faster than any desktop CPU in 2005. It also required less memory bandwidth than rendering the same scene using a rasterizer... with about the same hardware resources as a Rage II.

Using an FPGA significantly limits the performance. Using current process technology you'd be looking at 400 MHz and 600 million gates at least. Raytracing is "embarrassingly parallelizable", and the RPU design had to be trimmed down to fit into an FPGA implementation, so I don't see any reason you couldn't get comparable performance to a rasterizer for similar scenes if you gave them a level playing field, otherwise it's like you're comparing a Rage II to a geForce 8800.

Re:Still need dedicated hardware.

[ceoyoyo]

I think we agree: the hardware has advanced to the point where ray tracing and rasterization are fairly comparable speedwise at resolutions approaching the highest you'd want to run them.

A Rage II could do some pretty nice rasterization at 500x300 too, faster than 7.5 fps usually. Today you don't get any attention unless you can do what, fifteen, twenty times that number of pixels? Now that the hardware has gotten to that point the quality and other advantages of raytracing are going to make it take off.

Re:Still need dedicated hardware.

[argent]

A Rage II could do some pretty nice rasterization at 500x300 too, faster than 7.5 fps usually.

I would guess that the performance of Unreal on the Rage II would be a fair comparison to UT2003 on the RPU. The Rage II did about 8-9 FPS at 640x480 with a slower CPU handling the geometry updates for Unreal, and the RPU did 7.5FPS at 500x300 with UT2003. So I'd say it's comparable performance for comparable hardware.

OpenRT, a highly optimised software renderer on a 2.66 GHz Pentium 4, got 8FPS for the same game. A software OpenGL renderer running UT2003 rasterized on similar hardware (2.8 GHz Pentium 4) got about 60 FPS at the same resolution.

Now that the hardware has gotten to that point the quality and other advantages of raytracing are going to make it take off.

Based on the figures in the Siggraph '05 paper and the benchmarked performance of equivalent rasterizing hardware running the same engine, if ATI or nVidia had Dr Slusallek's hardware design in '96 when the Rage II came out, they could have a decent dedicated raytracing engine, in hardware, by the turn of the millenium. Or if nVidia had started cranking up a raytracing card in 2005 after Siggraph they'd probably have something that did 60 FPS at HD resolution for $100 by now.

Also code

[dbIII]

For a few years we've heard of Quake3 and 4 rendered by distributed machines but there's nothing we can download and try out ourselves.

Take that, John Carmack!

[OMNIpotusCOM]

pfft... and John Carmack said that they were going the wrong direction with ray tracing. Shows how much he knows. And they only pulled it off with 4 quadcore processors at 15 FPS. That'll show you! Maybe someone should tell Carmack to go back to developing new shades of black instead of dealing with the light! /sarcasm

I would love to hear John Carmacks thoughts.

[AbRASiON]

John, if you're reading you'd probably be one of the better people to drop some comments here.

Re:I would love to hear John Carmacks thoughts.

[Anonymous Coward]

Don't you think he already knows that?

Real time ray-tracing is good for the coming tech

[jabjoe]

Writing apps for huge number of processors is a bit of a nightmare, but ray tracing is a problem that fits very neatly in this world. A ray tracer can be very small, so once loaded on a processor, it can just munch through data (the scene) until it's finished. A collection of stream processors with read access to main memory (or a large pool of memory where the scene lives) would be perfect. If you have a pixel shader with read access to main memory, draw a single quad over the view, and have each ray tracer in pixel shader form and you get what I mean.

call your bullshit.... Re:Height maps

[leuk_he]

There were workarounds for self modifying code (VirtualProtect) and if you were a developer you were aware of this. The real trouble comes when you want to run this code on a out of order cpu (P5). If it runs you loose a lot of performance with self modifying code.

and of course it gave me the opportunity to refer to it as a dynamic synth compiler :) The bitches were all over me, dawg! I get the technical part, but the part about "bitches all over me" makes it unbelievable.

Rear view mirror!

[mac1235]

Remember camera you could mount on your shoulder, facing backwards in System Shock? It was useful.

Embedded programming and GBA

[tepples]

Embedded programming is a booming business with no new programmers coming in, because all the college graduates want to be game developers I learned embedded programming because I wanted to develop games. The Game Boy Advance and similar handhelds act a lot more like embedded systems than like a modern PC.

Re:Embedded programming and GBA

[Impy the Impiuos Imp]

After spending 10 years doing Windows programming, including the occasional driver, I went to the embedded world and realized they had a whole different look on things.

Embedded has a whole level of bug prevention regular windows programming doesn't have. And if you have a higher safety classification rating (read: something in a car that, if it breaks, can cause an accident = millions in damages + bad rep for your company) it's a whole 'nother level still.

And I don't care how "skilled" a Windoze or Linux programmer thinks they are -- unless you've been in the embedded world, you don't know what you're talking about, no matter how many books you've read.

"Regular" programming has good techniques, too. As embedded RAM gets cheaper, they're toying with C++. And even with just C, there's something to be said for pseudo-encapsulation of all your data, i.e. everybody must use an access function for globals, and modification functions, rather than just modifying data directly in this or that source file, pell-mell.

If you did both for some time, you'd be well-rounded. Take some Lisp on top of that and assembly, and you'd be a god. Like me.

Re:Embedded programming and GBA

[ya really]

I know the old school nintendo and other consoles back then had many of their games programmed in assembly, but I suppose they would have to with the minimal resources they had to work with. That to be hell doing something like mario brothers 3 in asm.

Re:Embedded programming and GBA

[billcopc]

Assembler isn't that much different from C or any other simple language. An assembler is essentially a preprocessor for machine code. It adds human elements to the code to make it more manageable. You can define jump labels, functions, constants, macros - same as any other.

Any decent asm programmer has a bunch of standby routines to call upon, much like a good application developer will have a few custom libraries with shared code. To build a game like SMB3, you would build a few basic libraries for graphics, sound and input, then tie them together with some very straightforward loops and branches. The game logic for SMB3 is simple and deterministic, everything behaves according to fixed patterns, there's no A.I. or pathfinding to worry about.

"Protection"?

[tepples]

And self modifying code is still used today on some software protections, not just viruses. To many posters to Slashdot, software "protections" are just as much malware as any virus.

Here's the video

[Sark666]

http://www.pcghx.com/aid,646920/News/Video_shows_ray_traced_Quake_Wars/

Software rendering

[DrYak]

Except the graphics kind of look like crap. If it's going to run at 16fps it better look a LOT better than traditional optimized rendering. Except that they are currently only using 16 cores to render that, and the final Larrabee is supposed to have much more of them... ...well, when it finally comes out (for now, it start to sounds much more like some vaporware - with technicians from Intel regularly putting demos of how thing could look cool using x86 cores instead of GPU SIMD cores but still no hardware on the horizont).

The other good thing is that ray tracing scales up almost linearly with the number of cores and chips. (Whereas there's a diminishing return for each additional GPU with traditional rendering).

Re:Software rendering

[mdwh2]

Except that they are currently only using 16 cores to render that, and the final Larrabee is supposed to have much more of them... ...well, when it finally comes out

By which time, GPUs will have more cores also.

The other good thing is that ray tracing scales up almost linearly with the number of cores and chips. (Whereas there's a diminishing return for each additional GPU with traditional rendering).

Does it? I'm no expert, but I'm curious for references? (And note, we should be comparing adding additional GPU cores on the same GPU, rather than literally additional processors.)

Re:Software rendering

[Solra Bizna]

I wrote a whole long, insightful comment that went through all the technical issues on both sides, but Discussion 2.0 ate it and now I'm ticked off, so I'll just write the conclusion.

In raytracing, you deal with pixels, so you can get linear performance gains just by adding nodes (even over a slowish network like gigabit Ethernet) and divvying out pixels to work on. In rasterization, you deal with triangles, so you can't.

-:sigma.SB

Re:Software rendering

[Anonymous Coward]

Modern GPU's have several 100's of cores so rasterization must be scaling fairly well.

Myst!

[famebait]

I had completely forgotten I used dream about this, but finally it is here:
Scenery eerily similar to Myst (even the tree models), but raytraced in real time.

Miss the point?

[Funk_dat69]

"Lighting", normal mapping, etc. are short-cut approximations that are tweeked to look good with current graphics systems. You simply wouldn't use these techniques if you had a ray-tracing engine.

To get your beloved 'material fidelity' you would need a machine fast enough to ray-trace with a larger number of 'rays' and be able to 'bounce' and 'absorb' those rays per material.

Re:Miss the point?

[Anonymous Coward]

Normal mapping is extensively used in ray-tracing. Unless you have the hardware to render 5 million poly objects from zbrush/mudbox you use displacement maps or normal maps. Ray-tracers are usually not as good at displacement as reyes renderers like 3delight or renderman so you want to use high poly with a normal map for adding definition.

I submitted this a week ago

[sproketboy]

and it was rejected. Thanks /. editors - for NOTHING.

6809...

[Anonymous Coward]

I worked with the 6809, and let me tell you, when I had to once program on a 6502, with it's freak'n ONE register, I just about crapped my pants! The 6809 waxed the 6502's ass!

Re:6809...

[Tyger]

Technically it had 3. The A, X and Y registers, with the X and Y used for indexing.

Doesn't this sound like something else?

[Ilgaz]

When I heard this Ray Tracing on CPU while GPU would do it in a fraction of time story, I jokingly said "Intel finally admits they can't do Graphics thing."

Next, IBM will show Crysis doing 120 FPS on their Bluegene/L massive supercomputer.

Just ray-trace subset...

[PenrosePattern]

I'm a bit confused. Isn't the likely / adequate / improved implementation (beyond normal) just:
1) render the scene normally
2) for those items that are transparent or reflective (e.g. not all & usually not many), ray-trace them (e.g. refract, reflect)
This give you 'true reflections' on glass/water/shiny/etc. & refraction through water, lenses, etc. A significantly better image than 'shaded' polygons. Limiting the set of pixels limits the cost significantly. You don't need to do the 'whole' scene.

Single source (e.g. camera-eye) Ray-tracing doesn't give you shadows, for that you need to trace from surface back to light source(s).

To be fancier, you really want real-time radiosity.

Re:Real time ray-tracing is good for the coming te

[FunkyELF]

A collection of stream processors with read access to main memory (or a large pool of memory where the scene lives) would be perfect. IBM's Cell processor does this. There are some youtube videos of realtime raytracing using PS3's

This game is already out

[vecctor]

Just in case it isn't clear, this is just a tech demo some guy at Intel put together. It's totally academic/proof-of-concept, and doesn't have anything to do with the actual game.

The game they modded was Enemy Territory: Quake Wars made by Splash Damage - the same guys behind the multiplayer component of Return to Castle Wolfenstein and Wolfenstein: Enemy Territory.

The game is already released. It uses an updated version of the Doom 3 engine and actual game doesn't use ray-tracing.

As for storyline, it is a team multiplayer game like Wolfenstein: Enemy Territory was (and TF2 is, etc) so there really isn't much of one :) Just gameplay.

somewhat off-topic, but..

[vecctor]

This article is about a tech-demo based on the actual game, not the game itself.

But you mentioned demos, so if you want to try out the demo of the actual game, here it is:

http://community.enemyterritory.com/?q=node/225

Team-based objective multiplayer is tons o' fun.

I don't think they are "promising" anything

[vecctor]

They aren't promising anything - this is just a tech demo by an academic guy now employed by Intel - not a game advertisement.

It's essentially a "gee whiz" engine mod - the game itself came out months ago and was made by a totally different company. The actual game uses a modified Doom 3 engine and doesn't use ray-tracing at all.

Raytracing makes graphics programmers obsolete

[badboy_tw2002]

The reason why is that suddenly there's no specific code to write for every material. Right now the graphics artist supplies a description of an object - its shape, its textures, etc. They have "materials" but these are often broken down into very special parameters for a raster based system. If I have a car window in current graphics engine the car window has code associated with it to describe how it reflects the world. In a ray traced system, the car window is described by the artist as "glass", which uses standard fields for a material that reflects how light is rendered. The whole system becomes data driven, and the graphics programmer is left to design a more efficient engine, not extra tricks around making materials render faster.

What's all that mean? It means that better looking game engines are within grasp of smaller and leaner dev teams. Less funding is required to make a game, less time is required to make it, so risk drops and more "out there" games can be made. The reason you see the same game sold every year is that the same game is a safe bet. To maximize return you need to minimize risk, and tech like this helps make it a safer bet to take a risk on a new type of game.

Apple II nitpick

[tepples]

even just the phrase "POKE -151" will bring a tear to my eye ... Almost. The area from $D000 (-12288) to $FFFF (-1) on an Apple II was ROM, and you couldn't POKE it unless you mapped in the II+'s language card or the IIe's virtual language card. Did you mean CALL -151, which starts a "monitor" (machine-level debugger) session?

Parent is *not* flamebait...

[Abcd1234]

I find it truly bizarre that your insightful comment was modded as flamebait... apparently some wannabe hacker was offended.

Re:Real time ray-tracing is good for the coming te

[jabjoe]

I know (I've written code for Cell), but I was thinking more of a sea of stream processors. It's a great way of using them all. :-)

Voxelstein 3-D

[Asphyxium001]

DOWNLOAD VOXELSTEIN 3D!!!! YOU WILL NOT REGRET IT. http://voxelstein3d.sourceforge.net/