Ray tracing is the current holy grail of gaming graphics, the rendering technique that might finally make the licensed game based on Pixar's latest look as good as the film itself. But, the typically random nature of rays has made rendering them on traditional hardware inefficient, a problem Caustic Graphics
claimed to have solved, and is now backing that up by giving
PC Perspective some further details and demos. The company's tech will rely on a new graphics co-processor called the Ray Tracing Processing Unit (RTPU), working in concert with existing 3-D accelerators to deliver rays at frame rates high enough for interactive applications. How high? Early hardware dubbed CausticOne (that giant slab of silicon above) manages 3 - 5 frames-per-second in the demonstration video after the break. That's not nearly enough for twitchy first-person shooters, but second-gen hardware due next year is looking to deliver 14 times that -- plenty to get your high-reflectivity frag on.
posted Apr 20th 2009 9:21AM
this sounds like physx stuff, it might be good for some people but the cost isnt worth it for most
It's just like PhysX... except it's completely different.
PhysX is a physics supplement that offloads physics, not graphics, calculations from the CPU.
This is an entirely different type of rendering from current GPUs though. A game's engine would have to be created from the ground up to utilize ray tracing.
Its's not physix it's the next generation in graphics. A new method in presenting realistic graphics. It will be quite a bit of time until we consumers can get our hands on raytracing cards.
I don't know why it wasn't included, but there is an article link attached to this picture and video:
http://www.pcper.com/article.php?aid=694
Ray tracing is a rendering technique developed way before any graphic cards exists. It is used in every non-interactive rendering package, so at least don't call it new.
Is it me or did everyone miss the point of this comment? It's exactly like physx in that if there is any consumer product it would be a very expensive coprocessor card which nobody will buy. At least this one might have real applications (outside the consumer market, ie not games).
I'm sure it's technically impressive but whatever the framerate, that demo looks horrible, and I say that as a Wii owner. :)
In order to succeed, first you have to fail.
"the rendering technique that might finally make the licensed game based on Pixar's latest look as good as the film itself"
NEWSFLASH: Renderman sucks at raytracing.
Man! And I thought I was at a geeky tech site. Get your facts together next time...
Just wait till Nvidia buys it and puts it in every video card they make xD
this will be interesting 3-8 years time when it's integrated into graphics cards. No add-ons. you have to start somewhere and whether this will be successful or not doesn't really matter. It will change things.
That demo reminded me of the AWESOME CG in early playstation/saturn/pc games.
Of course, eventually this will make it TO graphics cards. Where's PhysX now? Nvidia.
Ray tracing is the future I watched the video yesterday and I didn't see anything that looked important. But it might be a long time before we see anything for consumers.
Raytracing is the future?
Yea, maybe in 1997. The tricks video games are pulling off these days using existing real time rendering solutions often rival ray tracing. Realtime AO (like SSAO) and GI are more realistic goals in my opinion.
Yeah, if it's anything like raytracing in traditional computer graphics, be prepared for 10-20 years of amateurish talentless "art" that gets accepted only because "HAY KEWL IT'S RAYTRACING".
In all seriousness, I can see big benefits to this - I just hope it's integrated well and that the fact it's so different from traditional computer game graphics doesn't make it useless to developers.
@Jamie
"The tricks video games are pulling off these days using existing real time rendering solutions often rival ray tracing..."
That's just it- they're "tricks". They rely on hacks like programmable shaders and complicated culling algorithms (BSP etc) just to keep things fast enough. The reflections and refractions we see in games are generated using environment mapping either a pre-rendered or realtime cubemap, and the effects fall apart. Shadows are kludgey- regardless whether we use shadow maps or z-buffered/volume-based algorithms- they still suck.
Ray tracing algorithms are simple and easy to optimize and embarrassingly parallelizable and they attempt to mimic actual light. Combining raytracing with radiosity (likely in later gen hardware) renders worlds indistinguishable from real life. And new features recently added to GPUs are still almost trivial to do in a ray-traced scene. Realistic shadows, reflections/refractions, displacement mapping, etc don't create as much a performance hit as they do in rasterization.
There are always downsides, which I think will be solved by making future rendering techniques a hybrid between rasterization and raytracing (similar to how professional scanline renderers work today).
yeah, it probably will be a while. as long as it requires this special hardware, i can't see anyone making blockbuster games that use it (since i assume it would be a real pain to include two completely different rendering methods in one game). we'll probably have to wait until the hardware becomes fairly wide-spread before big titles start using it.
As Cornelius says, the tricks are just that, tricks, they fall apart, where this type of hardware would come into its own is with 3D displays, where current tricks are obvious, this type of tec would remove the need for those tricks and provide a much more interactive lighting environment. Plus this would be fantastic if integrated into 3D modelling applications so you could see your work in real-time.
The elephant in the room with ray tracing is that in order to calculate each ray, you need to have every 3d object and every texture on that object available to collision test. There aren't very many shortcuts you can take since much of the 'realism' of ray tracing usually involves reflections, and reflections can bounce off objects in any order.
Sure, you can throw another gig of memory on the graphics card, but then game designers will add 1.5 gigs of detailed models with high rez textures. It's a losing battle because ray tracing puts the full brunt of the most inflationary part of software squarely on one of the most expensive parts of hardware.
On the other side of the equation, rasterization, even with fancy shadow calculation, scales much more nicely, and all those 'tricks' are easy to pipeline, and don't compound as badly or inflate as quickly. You just get a much higher rate of return investing in better tricks than trying to brute-force the problem with ray-tracing.
@GTMoogle
You've missed the point of dedicated raytracing hardware then.
Rasterization is easy to pipeline? Wha?!?! There are many, many stages in the traditional graphics pipeline that are blocking. For an idea, break the modeling setup apart, you'll realize that geometry processing needs to be finished prior to vertex transformations. Vertex transforms block lighting. And so forth. Yes, each vertex transform and fragment program can be parallelized, but the entire pipeline cannot. Frankly, I'm amazed that GPU manufacturers managed to work out most of the kinks in multithreaded rendering contexts, but there are still limitations.
In raytracing, each ray can run on separate thread. Not counting redundancies (which can be optimized), these can be performed simultaneously. Also, at least in eye-casting applications, we get to remove clipping/frustum culling and visibility optimizations (what isn't seen is no longer drawn, which isn't always the case in rasterization).
Btw, Rasterization scales linearly (O(n)) for the number of polygons in the scene. In raytracing, growth is linear as well, but static structures can be optimized into k-d trees for logarithmic growth.
@cornelius
"renders worlds indistinguishable from real life"
If I got a penny every time I heard that will be 'next' I'd have enough to buy several of these raytracer cards.
Hmm...the next Bitboys, perhaps?
dude. I don't need to see that avatar this early in the morning, or ever, really.
Well, it would be pretty cool if they could get reasonable performance out of some moderately priced hardware. Raytracing is a much nicer solution than the current stuff in the traditional rendering pipeline. Less complicated and full of hacks!
Now, if they can just pull off some real time radiosity too...
Can't wait to see this type of stuff in the new MW2
Damn you - I got all excited thinking that perhaps you meant "Mech Warrior 2" and that a remake of it was coming.
I want to see this rendering Far Cry 2. Seriously, the engine in that game is probably what they use to render real life.
Except not, considering they'de have to remake the entire game engine anyways in order to ditch their current techniques in lieu of raytracing.
Things like this come in cycles. Right now, many people think it's a good idea to have separate processors for everything. Then, in a few years, somebody will have the bright idea that it can all be done on a single die. Rinse and repeat ad nauseum. Who else remembers the 80387, the 68881, etc.?
Hmm, sounds like work to me. Centralisation sounds like a good idea, so let's do that and get everything under control. Scroll forwards a few year and it's, let's de-centralise and give power to the experts in each location. Rinse and repeat.
The reason things like this are developed as discrete add-on cards is because no one wants to invest the massive time and effort to integrate it into an otherwise OK product and then have it bogged down with an expensive feature no one wants. They only integrate it later when it's a must-have.
No the reason for it is practical. Usually pertaining to heat/space issues. First gen anything almost always is on an external card with a co processor. That and its less risky for someone to do it this way. In that if its a failure you didn't just spend a **** load of money on developing one integrated system that now is a waste of money to produce. Better to have an external card that can be sold to Dell, HP, etc as an addon to their systems.
Can't wait to see what that will look like. Expect perfect shadows, transparency and refraction.
There's a feature for the PS4.
[i]But, the typically random nature of rays[/i]
What's random about rays? They travel in straight lines.
They scatter from surfaces and objects. The Scatter effect of different materials is what makes Ray Tracing so life like.
@TrojanLL: But scattering isn't random. It's well defined within the laws of classical physics.
The scattering in many raytracing algorithms IS random though.
It really depends on what raytracing/pathtracing approach is taken.
It isn't random if you want it to look shitty.
Randomization is how you avoid aliasing.
How powerful can they be with those crappy cooling units?
The 90's called, they want their heatsinks back.
See this isn't Nvidia or ATI who don't actually do anything more then pack more transistors into the damn thing and call it *holy chorus sings above* INNOVATION. I wouldn't be at all surprised if they did more with less because they are a smaller firm that doesn't need to beat their competition to the punch by cutting corners. People like this is where REAL innovation comes from.
Sorry if I sound jaded. But if you look at what Intel is doing with their processor line it proves that they learned the lesson that packing more Mhz into something isn't always the best thing to do. Intel and the Core series is doing more with less power. It would be nice if ATI and Nvidia would take a break from the arms race and think about how to do graphics hardware going forward. Instead we will continue to see graphics cards grow hungrier and hungrier, eventually needing their own power supply. . . oh wait that already has happened.
@Jon Doe
what? where have you been for the past year? ATI has done the exact same thing as Intel with the 48xx series, drawing more efficiency out of their designs and targeting the mainstream instead of the hardcore.
In the video they say it's running at 100mhz.
Hence why they would love for it to catch on so that they can release a commercial card running in the GHZ.
Get it up to 3ghz like a modern CPU and put 4 cores of that on a card and you're looking at a rediculously fast raytracer.
If they had NVIDIA's access to fab technology then they would have a use for a nice big loud fan.
Do you have any idea how a GPU works? Or do you think a GPU is ZOMG LIEK A COER TOO DUOO BUT AWESOMER!11!1!!
@monkeypox
ATI is NOT doing a good job of making their GPUs more efficient right now.
If you compare nv and ATIs flagship cards (the GTX295 and the 4870x2), the ATI card consumes far more power and generates much more heat than the nvidia one (and generally for slightly less performance). ATI seems to be taking the route Intel took with the P4 ("crank up the frequencies, to hell with the increased power consumption and heat!") at the moment.
I don't think this will catch on in consumer realm for a long time.
Normal raster based GPU's will be much faster, and much cheaper. So until they can make this run most games as fast as a GPU, and at most within a 10% higher price, then at most Pixar n co will buy these to get a vague look at their work in real-time.
Sure ray-tracing look's pretty, but that video was some water, which Dx10 made water look kinda nice. Show me (oh this is so cliched) Crysis running Ray-traced at a decent speed. Cos even a normal GPU can do 2 car's n a pond.
Read the article. These cards reside alongside the GPU and process the ray tracing required by the scene for lighting, etc. The GPU will still be used to do post processing and other things. It's an add-on card like PhysX was, not a GPU replacement.
That was 3-hfps, with a GPU backing it up. Ouch.
I can knock out HD-quality render's of simple object's at about 12fps, and that's CPU based. So currently this tech isn't selling me. The dream of Ray-Traced real-time render's is a nice one, but beyond tying it to a Quadro for 3D film work I personally don't see any future in this. especially in gaming. Give it a decade or so n GPU may hit a wall like the CPU did in the ghz race, but until then GPU's are cheaper, faster and good enough.
Exactly. We got so good at cheaply faking the appearance of ray tracing that this tech has become moot. 'But we got true accurate caustics working at 2-3FPS at VGA resolutions, and all it took was this $500 card!'. Who cares! Of the 10% of people that might actually spot the difference accurate ray tracing would make to a games, only 10% of that set will actually care, and only 10% of that subset would be willing to spend the extra money. To be honest, I don't really see the point for the likes of Pixar either. Sure they want accurate ray calcs for the big screen, but they also want the kind of programmable flexibility that hardware renderers never provide. Larabee might help but I bet this won't.
Well, anyway, I certainly don't wish fail on these guys, but I don't expect them to set the world on fire either. Ray tracing might become a standard (as the GPU people need to have Something to sell as 'the next big thing'), but it won't have really been necessary.
The demo was fully utilizing the card. The final product would be an estimated 14x more powerful. At first it could be used to assist raster rendering by accelerating their already ray traced lighting tricks.
Please read the article instead of speculating and spreading fud.
Also, this is a prototype card. It's running at 100Mhz instead of the 350Mhz they would like and it's using FPGA Processors instead of final hardware.
So, 14x faster = 60fps at VGA resolutions. Whoop-dee-doo.
Why, given that ray tracing gives you pretty much all "realistic" effects for free, are they demoing a dull grey landscape with a couple of cars? It's not exactly a mind blowing demo, especially given what people like intel were demoing a couple of years back on distinctly less specialised hardware.
Elmer you completely miss the point.
This wasn't using the GPU at all. This ISNT a hardware renderer. This lets you use the full power of your raytracing shading pipeline. But accelerates it. It's a hardware co-processor for a SOFTWARE renderer.
A GPU is a hardware renderer. It's extremely limiting and extremely frustrating to work with. The Caustic card ISN'T A RENDERER. It's a render accellerator. It would be like calling PhysX a game engine. It's not a game engine, it's a co-processor.
Now if your game wants to use raytracing it can call data like it would the CPU. This is exactly the same as using a CPU only faster. This is also very clearly still 'developer art'. The simplicity of the shaders is a result of the demonstration not the capability of the card. Anything that uses Raytracing can use raytracing EXACTLY the same as it does currently just faster.
if this thing ever makes it way to mainstream then they could have a winner on their hands..btw before anyone asks...yes is can play crysis!
I think its quite the news that these guys are developing this on a mac. Which stands to have a very interesting possibilities, since mac gaming hasn't been the best place to find gaming hardware or software.
Yah because this won't ever be on the PC. Nope. Not at all because the core components of the Mac are so different from the PC that it will never happen. Sorry, but Mac gaming has been, and always will be a joke.
*shrugs* There is nothing wrong with co processors. They generally are the stop gap before having it integrated into the system board. Take a look throughout computer history and this has almost always been the case. Everything from audio, to video, to NIC, to modem, to memory (Look at some of the smaller laptops out there memory is integrated onto the board now.), to the HD controller. It all eventually gets rolled into one set of chips as will this with time.
bad part is once they're integrated progress slows down.
Physx is integrated into most of the newer video cards as well.
Co-processors don't work for the reason PhysX failed: chicken and egg problem for consumers over whether to buy the hardware when the software doesn't exist, and for developers on whether to spend resources to develop for the hardware when nobody owns it.
However, it's possible for this to work for the reason why PhysX didn't fail completely: if this card can do raster based graphics as well at a good speed, then people would buy it in good conscience while they wait for developers to get on board. Because as far as I can remember, the whole problem about RT is that it requires a massive, massive amount of parallel processing, but that the architectures wouldn't actually be that different, just that today's GPUs are optimized with things like shader units in mind that RT wouldn't need.
It actually needs a discrete video card if I read it correctly...
Same reason Physx failed - additional card that doesn't do anything and costs twice as much as a normal video card. Either Nvidia/ATI/someboy buys them or they die from hunger.
Yeah, going to be a tough sell for the mainstream crowd unless they can get it cheap and fast enough. If it really adds a new level of realism to games they might get my Benjamins, but from the looks of the demos they're simply giving me what I already have (or less) simply without visual "tricks" and I simply don't effin' care.
So, it's a Poser-accelerator? THAT'S just what the world needs.
More desktops of highly reflective balls in space.
Why not just toss in another GPU and run raytracing code on it?
___Read the article.___
It explains exactly your question as much as the company was willing to release. Essentially, GPUs do not have the per stream memory to be able to modify the models in real time and other things.
Ray tracing doesn't look as good as scanline rendering anyway. All the techniques we've developed for things like anti-aliasing don't work with ray tracing. Plus, for any non-trivial scene, ray tracing takes exponentially more work.
I don't get why people are interested in real-time raytracing anymore. I thought the idea of it died in about 1998 when scanline rendering started to look so much better than raytracing.
Actually, if you read the article, you can see AA being implemented.
I thought we were moving away from add-on boards and graphics cards;
Graphic acceleration cards, Riser boards, co-processors, were all developed back in the day when processors were single core running at 500Mhz and not able to processes information fast enough.
Developing more add-on boards and such seems like a step backwards for me. Why not work on better core APIs that allow multicore CPUs to shift a core into GPU mode another into a Physics mode, and then us the remaining 2 cores for maths (they were talking about this when multicore became mainstream).
no matter how good your add-on is, in the end, you still get chocked by the Mobo's bus, the more the processor core(s) do on their own the better.
They because even modern multicore cpus still suck at math crunching, just look at the i7-965@3.2 GHz 51.20 gflops source
http://www.intel.com/support/processors/sb/cs-023143.htm
This company is gonna come out with it's first gen cards and the Nvidia is license the tech and through it in their drivers.
I'd rather Nvidia just buy them out and do like they did with Physx. That way I wont have to buy a separate card and risk it becoming redundant with my next videya card purchase.
The accelleration happens thanks to hardware not software.
PhysX was a problem which could easily be solved with GPGPUs. Raytracing isn't. In short it requires access to memory in a different way than a GPU operates. So Raytrace accelleration isn't terribly fast. As the movie describes it isn't just a matter of throwing more CPU clocks at the problem, there are other bottlenecks that need to be addressed. The caustic card addresses these hardware bottlenecks.
Now NVidia could license the hardware specs and add it onto their GPUs. But they can't emulate it and get any kind of performance improvement.
All this talk about Pixar and raytracing is bollocks anyway. The vast majority of Pixar's work has been done with REYES, which is a micropolygon rendering system, much more similar to scanline than raytracing.
OK thats nice you can have a low spec System with this Card in there and still get that same performance as a highend system
+1
im still trying to figure out what is supposed to be impressive in this video.
LOL. "raytracing is the future"
I've been hearing that for twenty years. The nice thing about the "the future" is that it never fucking gets here you so can keep on saying that and it'll still be true!
Raytracing is a correct model of our world, no cheats. If it could work in high fps it would be extremely easy to create nice/correct looking games. This is true simulation. Nice work.
@Cornelius
Ray tracing may optimize for O(ln(n)) for singular searches, but that's not the bottleneck. At least in the present incarnations of technology, we're talking about "how much geometry is represented in the scene". Adding more geometry means adding more memory. The bottleneck is the datapath to main memory. Ray tracers actually do not reduce the total geometry needing to be touched to render a scene by anything significant. Current 3d rendering methods can employ ray-tracer esk geometric culling to get that number more in line with a ray tracer, with significantly less silicon. That is the present bottleneck for rendering geometry.
It boils down to cost, the theoretical "advantage" of ray tracers in being able to view a large amount of geometry is just that, theoretical, it breaks down in application and will never cost anywhere near current hardware for the same level of performance.
Cost will drive adoption. Sum, I never expect mainstream adoption of ray tracing 3d hardware.
Ray Tracing is undoubtedly the future, but it still has a long way to go. When we can sustain a solid 60 frames per second, then I'll be interested.
F#!$% im going to need more PCIE slots for one of these....
I don't understand too well the implications of this product but everyone here is mostly way off on what is being implied.
Caustic One is an enterprise product not a consumer product, hence why it has a $4000 dollar price tag and is being advertised to developers on their website.
What Caustic One is supposed to do is reduce the time it takes companies to reduce their time creating ray traced scenes by a factor of 20. Reducing every dollar and minute they spend to 5 cents and 3 seconds would be a ridiculous amount of savings if it works.
As it stands Caustic One is useless for gaming. Ray tracing takes a lot of time to render and even the reductions this version would create would be insufficient to work for gaming in real time.
Caustic Two is the next iteration that they hope to have available by next year. That's the one that will make ray tracing for gaming feasible and this will only be considered if developers that currently ray trace now adopt Caustic One as the new standard in ray tracing development.