Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

John Carmack's Comment on Hardware Ray Tracing

arstechnica.com

Post reply on HN
Print view View original

11–20 of 69 posts

Re: John Carmack's Comment on Hardware Ray Tracing

#11

Post by wmil »

chii wrote: post #5

Having no real knowledge of ray tracing (or rendering for that matter), can someone tell me why we can't be moving towards a massive cluster of gpu (in the order of 10,000) ray tracing a scene in parellel? I imagine each ray trace call is quite independent of any other. SOooner or later, cpu (and gpu) will have many cores available for such form of rendering. Is it because of the chicken/egg problem?

Current rasterization methods also parallelize well. The problem is that ray tracing is just less efficient for most scenes.

Re: John Carmack's Comment on Hardware Ray Tracing

#12

Post by ChuckMcM »

electrograv wrote: post #3

Excellent response by Carmack. "Ray Tracing" has always been an overhyped or misunderstood technology, at least in my experience. Because of the impressive lighting effects it's famous for producing, people view it as some "holy grail" of superior computer graphics technology that we just need to optimize a bit for use in games. As Carmack described, those highly realistic ray traced renders come at a price: billions…

Pretty much agree, the only thing that we might disagree on is the 'billions' number, while I expect good ray tracing to trace billions of rays, if transistors are cheap enough this becomes more interesting. The typical HDMI 1080p display is 2 megapixels (or megatexels) if you have a rendering engine with 2,073,600 cores each of which is looking at a billion ray 'view' based on where it sits in the scene, its easier to set up the scene and light it. That presumes that you can do those cores like DRAM instead of like current processors of course. My point is that cheap transistors keep amazing me and whenever I say "Oh that will never happen" some fool goes and shows me I'm wrong.

I'm trying to imagine how complex a 'core' would be that computed the incident rays on a single pixel. Then figuring out how big that is in a 18nm process technology and then trying to see if I can fit 2M on a reasonable size die. My head exploded sadly.

Re: John Carmack's Comment on Hardware Ray Tracing

#13

Post by yk »

electrograv wrote: post #3

Excellent response by Carmack. "Ray Tracing" has always been an overhyped or misunderstood technology, at least in my experience. Because of the impressive lighting effects it's famous for producing, people view it as some "holy grail" of superior computer graphics technology that we just need to optimize a bit for use in games. As Carmack described, those highly realistic ray traced renders come at a price: billions…

I think people rather think of ray-tracing as much better motivated by physics ( and intuition) compared to all the complicated matrices in raster engines. Instead of thinking it is superior because of some demos.

Re: John Carmack's Comment on Hardware Ray Tracing

#14

Post by cmccabe »

zxcdw wrote: post #9
chii wrote: post #5

Having no real knowledge of ray tracing (or rendering for that matter), can someone tell me why we can't be moving towards a massive cluster of gpu (in the order of 10,000) ray tracing a scene in parellel? I imagine each ray trace call is quite independent of any other. SOooner or later, cpu (and gpu) will have many cores available for such form of rendering. Is it because of the chicken/egg problem?

AMD's vision of "heterogenous computing"(I think they came up with a new name for this?) would essentually be the "true" fusion of a CPU and a GPU in a sense that the chip consists of general purpose execution units(current CPU cores) and of special purpose parallel processing units (curent GPU SIMD/VLIW units) which operate concurrently in such sense that both sequential and parallel code can be executed efficiently…

Rasterization also requires "acceleration structures," kept in memory. I could be wrong here, but I think the point John was trying to make was that there was again a constant factor handicapping ray-tracing. But constant factors are, well, constant, and in 2050 we may well have ray-tracing done in hardware delivering scenes that are indistinguishable from reality.

I wrote a ray tracer once, but it was primitive. So I'm not completely talking out of my butt, just mostly.

Re: John Carmack's Comment on Hardware Ray Tracing

#15

Post by yk »

ChuckMcM wrote: post #12
electrograv wrote: post #3

Excellent response by Carmack. "Ray Tracing" has always been an overhyped or misunderstood technology, at least in my experience. Because of the impressive lighting effects it's famous for producing, people view it as some "holy grail" of superior computer graphics technology that we just need to optimize a bit for use in games. As Carmack described, those highly realistic ray traced renders come at a price: billions…

Pretty much agree, the only thing that we might disagree on is the 'billions' number, while I expect good ray tracing to trace billions of rays, if transistors are cheap enough this becomes more interesting. The typical HDMI 1080p display is 2 megapixels (or megatexels) if you have a rendering engine with 2,073,600 cores each of which is looking at a billion ray 'view' based on where it sits in the scene, its easier…

Throwing more ( or finer etched) silicon at an algorithm will not help the algorithm, compared to a superior algorithm. For almost all scenes you get better results with rasterization, the two exceptions I can think of is scenes where the geometry of light rays is not flat, so for example involving lenses or black holes, or where there is a lot more optical complexity than you can reasonably use. ( A forest where each of the leafs is an object, you will never be able to notice that the advantage of this over an forest where the trees in the background are just textures.)

Re: John Carmack's Comment on Hardware Ray Tracing

#17

Post by Osmium »

yk wrote: post #15
ChuckMcM wrote: post #12

Earlier quoted context omitted.

Pretty much agree, the only thing that we might disagree on is the 'billions' number, while I expect good ray tracing to trace billions of rays, if transistors are cheap enough this becomes more interesting. The typical HDMI 1080p display is 2 megapixels (or megatexels) if you have a rendering engine with 2,073,600 cores each of which is looking at a billion ray 'view' based on where it sits in the scene, its easier…

Throwing more ( or finer etched) silicon at an algorithm will not help the algorithm, compared to a superior algorithm. For almost all scenes you get better results with rasterization, the two exceptions I can think of is scenes where the geometry of light rays is not flat, so for example involving lenses or black holes, or where there is a lot more optical complexity than you can reasonably use. ( A forest where eac…

e.g. mirrors?

Re: John Carmack's Comment on Hardware Ray Tracing

#18

Post by ibrahima »

damian2000 wrote: post #4
hayksaakian wrote: post #2

Its so cool that someone as high up in the games industry as carmack actually knows his technical shit.

He might be high up now, but he started off as a programmer. He co-founded id software and (along with others) developed Wolfenstein 3D, Doom, etc. http://en.wikipedia.org/wiki/Id_Software

If you follow him on Twitter, he posts almost exclusively highly technical content.

Re: John Carmack's Comment on Hardware Ray Tracing

#19

Post by jamieb »

Here's a commercial, realtime raytracing tool. Runs happily on a dual-core macbook pro. No custom card needed. Yes, it runs better on a $20,000 64-thread Tigerton, but it will run usably on a laptop.

http://www.youtube.com/watch?v=yUiLBmioRI4

Carmack is right about games (surprise!). I can't imagine "Imagination Technologies" is pushing the Caustic R2500 at games.

Re: John Carmack's Comment on Hardware Ray Tracing

#20

Post by gsg »

chii wrote: post #5

Having no real knowledge of ray tracing (or rendering for that matter), can someone tell me why we can't be moving towards a massive cluster of gpu (in the order of 10,000) ray tracing a scene in parellel? I imagine each ray trace call is quite independent of any other. SOooner or later, cpu (and gpu) will have many cores available for such form of rendering. Is it because of the chicken/egg problem?

Ray tracing enjoys a ridiculous amount of inherent parallelism. That's not much of an advantage though, since rasterisers also enjoy a ridiculous amount of inherent parallelism.

Large scale "render farms" that exploit both already exist and are in commercial use in the VFX industry.

Post reply on HN