Ray tracing in realtime with PS3's

Started by
44 comments, last by zedz 17 years ago
Quote:Original post by zedz
Quote:Original post by Sneftel
Cool trick, but the PS3 hardware is laughably ill-suited for the demands of real-time raytracing. It's not going to happen in the next five years except on an ASIC.

what about a top of the line $1000+ intel cpu?
FWIW cell slaughters the $1000+ cpu WRT raytracing
cell illsuited, hmmmm, ild be pointing the finger more at other players :)
O...k...

What does that have to do with a custom raytracing ASIC?
SlimDX | Ventspace Blog | Twitter | Diverse teams make better games. I am currently hiring capable C++ engine developers in Baltimore, MD.
Advertisement
Quote:Original post by zedz
Quote:Original post by Sneftel
Cool trick, but the PS3 hardware is laughably ill-suited for the demands of real-time raytracing. It's not going to happen in the next five years except on an ASIC.

what about a top of the line $1000+ intel cpu?
FWIW cell slaughters the $1000+ cpu WRT raytracing
cell illsuited, hmmmm, ild be pointing the finger more at other players :)

true though we wont be seeing any major solely raytraced games coming to the ps3, but raytracing is used in practically all games, eg particle collision tests, line of sight tests + for more advanced stuff eg radiosity, gathering etc


Why isnt it ill suited? The cell uses an asymmetric processor architecture, which means that the SPE's do not have direct access to the main RAM. If you wanted to do ray tracing tests on the SPE's, you would be severly limited with the amount of geometry you could test against in 256KB of memory. In a typical game scene, your geometry would easily number 100k polygons. Obviously you cant fit all of that in the SPE's memory. A single ray test could require multiple iterations of filling the memory with different geometry from the scene. This will introduce additional overheads to the PPE which is already going to be heavily taxed since it is the only real capable General Purpose CPU which will be handling most of the conventional game code and also happens to be underpowered when compared to a Pentium 4 or G5.

In my opinion, theoretically the cell is extremely powerful but in practice the architecture is just flawed for games or almost any other general purpose software.
Quote:Original post by zedz
what about a top of the line $1000+ intel cpu?
FWIW cell slaughters the $1000+ cpu WRT raytracing
cell illsuited, hmmmm, ild be pointing the finger more at other players :)

true though we wont be seeing any major solely raytraced games coming to the ps3, but raytracing is used in practically all games, eg particle collision tests, line of sight tests + for more advanced stuff eg radiosity, gathering etc

Are you drunk?
Quote:Original post by GamerSg
Why isnt it ill suited? The cell uses an asymmetric processor architecture, which means that the SPE's do not have direct access to the main RAM. If you wanted to do ray tracing tests on the SPE's, you would be severly limited with the amount of geometry you could test against in 256KB of memory. In a typical game scene, your geometry would easily number 100k polygons. Obviously you cant fit all of that in the SPE's memory. A single ray test could require multiple iterations of filling the memory with different geometry from the scene. This will introduce additional overheads to the PPE which is already going to be heavily taxed since it is the only real capable General Purpose CPU which will be handling most of the conventional game code and also happens to be underpowered when compared to a Pentium 4 or G5.

In my opinion, theoretically the cell is extremely powerful but in practice the architecture is just flawed for games or almost any other general purpose software.


Well i wouldn't say ill suited, just not optimal, while it's true the cell SPUs doesn't have direct access to the ram it does have indirect access, and it also excels at streaming data from one processor to another, so instead of working on one pixel/ray at the time it just does several at once and then streams the geometry data between the 5 spu's available, so you can in fact use all that processing power without running into the memory bandwidth limit if you just think for a way around that problem.
Although, cell is still underpowered(for RTRT), and you would need 5-10 cells to get any kind of reasonable RTRT.

Quote:Why isnt it ill suited?

u miss the point of my post, A/ cell is a lot lot better at this than a top of the line pc cpu, ie the $1000+ is a lot more illsuited than cell

btw WRT the above demo, true framerate looks about 5fps (which is pants)
but the car consists of 1.6million polygons (hmm someone mention cell will struggle at 100k, well i guess not)
the resolution is 1920x1080 pixels
theres 96 ray cast per pixel

see here for a summary
http://www.gametomorrow.com/minor/barry/iRT-Sumary.pdf

(why has it suddenly gone quiet :) )

the quality is far higher than any existing game, sure they could make the quality lower and the framerate will improve, but im guessing they wanna show u can do movie quality cgi at interactive framerates, on the fastest pc doing the same scene yould be looking at seconds per frame not frames per second
Quote:on the fastest pc doing the same scene yould be looking at seconds per frame not frames per second


Not a fair comparison and/or not true. The PS3 is just released (in Europe that is), so compare it to three quadcore PC's please. Such a setup would do enough rays (easily) to render this at the same speed. Also note that even though 1.6M is a lot of triangles, 96 rays per pixel and no (or few) textures will keep coherency / locality of data (very) high. This is absolutely doable on 3 high-end PC's.

Quote:Are you drunk?


The kind of queries he mentiones all involve ray tracing, so no, he isn't necessarily drunk.

[Edited by - phantomus on April 11, 2007 6:58:01 AM]
I wonder why hasn't anyone yet linked to the last year's research paper that talked about ray tracing on Cell: http://graphics.cs.uni-sb.de/~benthin/cellrt06.pdf

There they describe how to overcome the limitations of Cell architecture and reach about the same speeds for every SPU as singlecore K8 achieves. At 1024x1024 without AA they reach 20FPS on 280k triangle architectural model (conference) with simple shading and shadows on a single 2.4GHz Cell. On a scene similar to that in the video (680k triangle VW beetle) they reach 16.2FPS.

Also, as Core2 has double the SSE throughput, Intel higher clocking quadcores are at least at the same level as single Cell in ray tracing, possibly even faster. Starting from 22'nd April you can get 2.4GHz quad for around $550.

I wonder what improvements does the new 65nm Cell have. I know it should run at extreme clock speeds (>5GHz) but I also hope it has some additional instructions for speeding up ray tracing.
how many raytests/sec can a single high end pc do (of course this is dependant on the scene) but a rough figure would be nice. 100million?
+ no i wasnt drunk (btw i used to work in houten)
Quote:Original post by zedz
+ no i wasnt drunk (btw i used to work in houten)
Uhm...where is houten?
SlimDX | Ventspace Blog | Twitter | Diverse teams make better games. I am currently hiring capable C++ engine developers in Baltimore, MD.
some more recent stuff about raytracing on the ps3
http://gametomorrow.com/blog/index.php/2007/03/07/cell-power-at-gdc-2007
a scene consisting 3 million polygons

Quote:
To put this render job in perspective I just ran one frame of this scene through 3dsMax’s default ray-tracer an it took my Centrino Duo (dual core x86) machine one hour eight minutes to render what Cell and the iRT renders in less than one half of a second.

of course 3dmax is not the worlds fastest ray-tracer but 0.5secs vs 68minutes is a 8160x speedup!!

apparently this was done under ps3-linux (with only 6spe's available), so any raytracing enthusiast's why havent u got a ps3, ive got some ideas for some simplistic raytraced games i will try out once i get a ps3 (me who hasnt owned a console since the atari2600)

Quote:Another point I wanted to clear up is that the iRT is a completely scalable ray-tracer. One SPE can render the entire 3M triangle city model by itself at 1080p.

This topic is closed to new replies.

Advertisement