For those at SIGGRAPH - see my course(s)

[mhouston]

Aaron Lefohn (Intel) and I (Mike Houston, AMD) organized the "Beyond Programmable Shading" pair of courses this year at SIGGRAPH. We have lots of good speakers from industry and academia talking about using GPUs for things other than traditional graphics pipelines for doing more advanced graphics and applications.

For those interested, the course is on Thursday and is split into 2 parts, "Fundamentals" (architecture and software talks from vendors and academia) in the morning and "In Action" (vendors and academics talking about actually using GPUs to do things beyond traditional graphics) in the afternoon. The course is in room 403AB.

See page 25 and 27 of the SIGGRAPH program or on the web:
http://www.siggraph.org/s2008/attendees/program/item/?type=class&id=549
http://www.siggraph.org/s2008/attendees/program/item/?type=class&id=57

[Simon F]

I'll certainly forward it on to my colleagues who are attending to see if they're interested... :)

[Tim Murray]

Will there be any material from this available online afterward?

[rpg.314]

Will there be any material from this available online afterward?

I want to ask this question myself as well.

[mhouston]

Yep, we will put up a website with pdf versions of hopefully all the slides the morning of the course and I'll post the link that night.

[cho]

http://ati.amd.com/developer/techreports.html


Presentations
Siggraph 2008
Advances in Real-Time Rendering in 3D Graphics and Games, Course, Siggraph 2008, Los Angeles, CA. Full course notes PDF (http://ati.amd.com/developer/SIGGRAPH08%5CSiggraph2008-Advances_in_Real-Time_Rendering_Course.pdf)
Chen, H. and Liu, X. 2008. (Bungie). Lighting and Material in Halo 3. ACM SIGGRAPH 2008: Proceedings of the conference course notes, Advances in Real-Time Rendering in 3D Graphics and Games, Chapter 1, pp. 1-22, Los Angeles, CA, August 2008. PDF (http://ati.amd.com/developer/SIGGRAPH08%5CChapter01-Chen-Lighting_and_Material_of_Halo3.pdf)
Mittring, M. (Crytek). 2008. Advanced Virtual Texture Topics. ACM SIGGRAPH 2008: Proceedings of the conference course notes, Advances in Real-Time Rendering in 3D Graphics and Games, Chapter 2, pp. 23-51, Los Angeles, CA, August 2008. PDF (http://ati.amd.com/developer/SIGGRAPH08%5CChapter02-Mittring-Advanced_Virtual_Texture_Topics.pdf)
Shopf, J., Barczak, J., Oat, C. and Tatarchuk, N. 2008. March of the Froblins: Simulation and Rendering Massive Crowds of Intelligent and Detailed Creatures on GPU. ACM SIGGRAPH 2008: Proceedings of the conference course notes, Advances in Real-Time Rendering in 3D Graphics and Games, Chapter 3, pp. 52-101, Los Angeles, CA, August 2008. PDF (http://ati.amd.com/developer/SIGGRAPH08%5CChapter03-SBOT-March_of_The_Froblins.pdf)
Boulton, M. (Rare). 2008. Using Wavelets with Current and Future Hardware. ACM SIGGRAPH 2008: Proceedings of the conference course notes, Advances in Real-Time Rendering in 3D Graphics and Games, Chapter 4, pp. 103-132, Los Angeles, CA, August 2008. PDF (http://ati.amd.com/developer/SIGGRAPH08%5CChapter04-Boulton-Wavelets.pdf)
Filion, D. and McNaughton, R. (Blizzard Entertainment). 2008. StarCraft II: Effects & Techniques. ACM SIGGRAPH 2008: Proceedings of the conference course notes, Advances in Real-Time Rendering in 3D Graphics and Games, Chapter 5, pp. 133-164, Los Angeles, CA, August 2008. PDF (http://ati.amd.com/developer/SIGGRAPH08%5CChapter05-Filion-StarCraftII.pdf)
waiting for the "Beyond Programmable Shading" part :)

[mhouston]

http://s08.idav.ucdavis.edu/

The course notes. There will be some updates over the next few days as people get their final decks posted as well as some additional material from a few speakers.

[Rys]

I've grabbed everything that's on the ucdavis site currently and bundled it up into a big bz2 to save people some clicking and bandwidth:

Mike, if you'd rather I didn't host the collection like that, let me know and I'll take it down.

EDIT: Removed link, see post later in the thread

[Arwin]

Oh wow. I picked the in-games presentation by John Olick first, and was wondering if there'd be mention of the Cell in there. Well as it turns out there is ... and it seems to be good stuff (from this layman's perspective). They seem to put all six SPUs to good use! Of course there's lots more interesting stuff on paralellism in there.

[idsn6]

Oh wow. I picked the in-games presentation by John Olick first, and was wondering if there'd be mention of the Cell in there. Well as it turns out there is ... and it seems to be good stuff (from this layman's perspective). They seem to put all six SPUs to good use! Of course there's lots more interesting stuff on paralellism in there.

Holy shit. He just went in-depth into the infinite detail sparse voxel octree geometry that isn't scheduled until idTech6.
Don't overlook the explanation bubbles in the upper left.

[cho]

he also imply there are 2+ triangle setup engines in future gpu (page. 118) ?

[CrazyButcher]

does anyone know how the presentation looked liked of the megageometry thingie, ie is that screenshot from the presentation, or just a prerendered/dcc app screenshot of the model ?

And thanks for making that big zip

[mhouston]

I've grabbed everything that's on the ucdavis site currently and bundled it up into a big bz2 to save people some clicking and bandwidth:

Mike, if you'd rather I didn't host the collection like that, let me know and I'll take it down.

We will actually put up a single pdf for everything once we have all the decks updated. I'd prefer that the ucdavis site host everything so that everyone gets the updates as we post them.

[mhouston]

does anyone know how the presentation looked liked of the megageometry thingie, ie is that screenshot from the presentation, or just a prerendered/dcc app screenshot of the model ?

And thanks for making that big zip

Jon showed a live demo and note that he said this is a side research project for him and this is *not* a demo of tech6.

[pocketmoon66]

Nvidia , AMD and Intel all selling their wares is a fantastic sight to behold :) Nvidia do the 'we can do all this NOW - and looks, raytracing!' and Intel saying 'we're not crazy... honest' .

I'd love to see the Nvidia ray trace footage :)

[Rys]

We will actually put up a single pdf for everything once we have all the decks updated. I'd prefer that the ucdavis site host everything so that everyone gets the updates as we post them.
No worries, I've removed the bundle.

[Tim Murray]

Interesting slides all around--thanks for the heads up, Mike!

[PeterT]

That's an amazing collection of slides. I've only looked over the first 6 sets so far, but I'll get to the other ones as soon as I'm able. I particularly like "Running Code at a Teraflop: How GPU Shader Cores Work" -- it's only stuff most of us already know, but I've never seen the ideas laid out, developed and summarized quite so neatly.

Too bad some of the presentations couldn't be more concrete (particularly the Larrabee and Compute Shaders ones). OpenCL looks really CUDA-like. I expected them to be similar but not to this extent.

[galopin]

this courses were really good.

For the john ollic part, the spu/rsx mix is the common admited way to do things as good as possible on ps3, but the raycasting of a preprocessed zbrush mesh on 8800 with cuda was really impressive :)