RT07 Papers List

(L) [2007/08/03] [Xela] [RT07 Papers List] Wayback!

Guys,

you could get my RT07 paper as [LINK http://www.sci.utah.edu/~wald/RT07/vertex_culling.pdf]

(I found that it is easier to place the text on somebody's else web site than navigate the red tape here at Intel. I'm grateful to Ingo for hosting this thing).

I will be in San Diego the whole next week (presenting a poster with the same name) with a sporadic access to e-mail.
Still, when I'll be back, I promise to address possible questions or suggestions.

Oh, BTW, the paper comes with the complete source code. If we forget about vertex culling, the stand-alone intersector in the paper is very fast by itself.
It is faster than other 3D intersectors I know and slightly slower than good ones based on preprocessing (if we ignore preprocessing timing, which might not be wise [SMILEY :)]).

Cheers,
--AR

PS

If somebody will be at Siggraph, you can find me near my poster (J13), I'm required to stand there on Monday and Wednesday from 12:15 to 1:15 [SMILEY :(] but I may ignore that anyway [SMILEY :)]

(L) [2007/08/04] [Xela] [RT07 Papers List] Wayback!

Guys,

you could get my RT07 paper as [LINK http://www.sci.utah.edu/~wald/RT07/vertex_culling.pdf]

(I found that it is easier to place the text on somebody's else web site than navigate the red tape here at Intel. I'm grateful to Ingo for hosting this thing).

I will be in San Diego the whole next week (presenting a poster with the same name) with a sporadic access to e-mail.

Still, when I'll be back, I promise to address possible questions or suggestions.

Oh, BTW, the paper comes with the complete source code. If we forget about vertex culling, the stand-alone intersector in the paper is very fast by itself.

It is faster than other 3D intersectors I know and slightly slower than good ones based on preprocessing (if we ignore preprocessing timing, which might not be wise [SMILEY Smile]).

Cheers,

--AR

PS

If somebody will be at Siggraph, you can find me near my poster (J13), I'm required to stand there on Monday and Wednesday from 12:15 to 1:15 [SMILEY Sad] but I may ignore that anyway [SMILEY Smile]
_________________
[Xela]

(L) [2007/08/04] [tbp] [RT07 Papers List] Wayback!

Loose builder + that intersector indeed sounds like a deadly combo; you're going a bit too fast for my taste over its "failure mode" charateristics, but who am i to complain as you provide a reference implementation [SMILEY Smile]
_________________
May you live in interesting times.

[LINK https://gna.org/projects/radius/ radius] | [LINK http://ompf.org/ ompf] | [LINK http://ompf.org/wiki/ WompfKi]

(L) [2007/08/04] [toxie] [RT07 Papers List] Wayback!

what tbp said..

;)

no, actually i think it's the best rtrt-paper released since a long time, even when the proposed algos still have some edges (but which algorithm hasn't any, so..).
_________________
The box. You opened it. We came.

(L) [2007/08/04] [tbp] [RT07 Papers List] Wayback!

Let's say i was too febrile after the rush to read it to state the obvious, it's a great paper. That also makes that post a me too onto another me too.
_________________
May you live in interesting times.

[LINK https://gna.org/projects/radius/ radius] | [LINK http://ompf.org/ ompf] | [LINK http://ompf.org/wiki/ WompfKi]

(L) [2007/08/17] [playmesumch00ns] [RT07 Papers List] Wayback!

>> Xela wrote:Guys,
I will be in San Diego the whole next week (presenting a poster with the same name) with a sporadic access to e-mail.
Still, when I'll be back, I promise to address possible questions or suggestions.
If somebody will be at Siggraph, you can find me near my poster (J13), I'm required to stand there on Monday and Wednesday from 12:15 to 1:15 but I may ignore that anyway

LOL didn't read this until today. Your poster was practically opposite mine! [LINK http://anderslanglands.com/bssrdf_disk_sketch.pdf]

(L) [2007/08/17] [Phantom] [RT07 Papers List] Wayback!

So... Basically 80% of the papers is available before the conference starts. Is that normal?
_________________
--------------------------------------------------------------

Whatever

(L) [2007/08/17] [Ho Ho] [RT07 Papers List] Wayback!

rayforce.net said dual 1.6GHz quadcore achieves 40-60Mray/s. That is "only" 7.5 per core. IIRC I got around 10M per core with some version of Phantracer on my 3.1GHz core2 with 2M L2. Did I miss something or their ray tracing isn't all that fast?
_________________
In theory, there is no difference between theory and practice. But, in practice, there is.

Jan L.A. van de Snepscheut

(L) [2007/08/22] [greenhybrid] [RT07 Papers List] Wayback!

the preliminary program is [LINK http://www.uni-ulm.de/rt07/News_%26_Downloads/Entries/2007/8/22_Preliminary_Program_and_Information_files/Excerpt.pdf here]
_________________
[LINK http://greenhybrid.net/ greenhybrid.net]

Real Men code Software Graphics.

(L) [2007/09/18] [Shadow007] [RT07 Papers List] Wayback!

Found out 1 paper that is missing from the index :
Adaptive Spatial Sample Caching
[LINK http://graphics.cs.uni-sb.de/~dietrich/]

For completeness, there is also Phantom's keynote :
Keynote:
[LINK http://rt07.raytracing.nl/]

I also found a paper (in french) which seems CLOSELY related to CADET/Lecussan's
Coupled Use of BSP and BVH Trees in Order to Exploit Ray Bundle Performance
"Regroupement des rayons pour accélérer le rendu d’images de synthèse réaliste."
[LINK http://www.irit.fr/EDIT/EDIT07/Articles/siao/cadet_edit07.pdf]

As I don't have mail here at work, I can't send the 2 first links to Kesen Huang...

(L) [2007/09/18] [Shadow007] [RT07 Papers List] Wayback!

Found out 1 paper that is missing from the index :

Adaptive Spatial Sample Caching

[LINK http://graphics.cs.uni-sb.de/~dietrich/]

For completeness, there is also Phantom's keynote :

Keynote:

[LINK http://rt07.raytracing.nl/]

I also found a paper (in french) which seems CLOSELY related to CADET/Lecussan's

Coupled Use of BSP and BVH Trees in Order to Exploit Ray Bundle Performance

"Regroupement des rayons pour accélérer le rendu dimages de synthèse réaliste."

[LINK http://www.irit.fr/EDIT/EDIT07/Articles/siao/cadet_edit07.pdf]

As I don't have mail here at work, I can't send the 2 first links to Kesen Huang...

(L) [2007/09/18] [Shadow007] [RT07 Papers List] Wayback!

>> Phantom wrote:I sent them an e-mail.
Thanks [SMILEY :)]
BTW : (just curious once more [SMILEY ;)] )
>> Phantom wrote:"Early Split Clipping for Bounding Volume Hierarchies": The problem with BVHs is that large triangles do not get split. This paper suggests splitting them before building the tree, by generating multiple bounding boxes. This allows for better splits everywhere, including high in the tree, which considerably improves matters for BVH and possibly also kD. Paper is not online available as far as I know, but I have the proceedings, so come check it out. Piece of cake to implement, will be in Arauna after the weekend.

Now that the week end has passed , would you per chance have any results to share ?
Wouldn't splitting the triangles before building the tree lead to multiple intersections with the same prim ?

(L) [2007/09/18] [Phantom] [RT07 Papers List] Wayback!

Ah, you got that from IGAD, right? [SMILEY :)]
I posted about my initial findings here:
[LINK http://ompf.org/forum/viewtopic.php?t=592]

I experimented a bit after that and got it up to a whopping 11% on ray traversal alone (no shading). So, I guess the results are not very good when this idea is applied to Wald's latest BVH packet tracing thing...

(L) [2007/09/18] [Phantom] [RT07 Papers List] Wayback!

Ah, you got that from IGAD, right? [SMILEY Smile]

I posted about my initial findings here:

[LINK http://ompf.org/forum/viewtopic.php?t=592]

I experimented a bit after that and got it up to a whopping 11% on ray traversal alone (no shading). So, I guess the results are not very good when this idea is applied to Wald's latest BVH packet tracing thing...
_________________
--------------------------------------------------------------

Whatever

(L) [2007/09/19] [Shadow007] [RT07 Papers List] Wayback!

>> hanatos wrote:.. i found another text which is completely unrelated to a paper from the conference
[LINK http://www.uni-ulm.de/~s_jhanik/IntRT2.pdf]

Are you sure it's UNrelated ?
It seems to me it's the real one [SMILEY :)]

(L) [2007/09/19] [moris] [RT07 Papers List] Wayback!

Coupled Use of BSP and BVH Trees in Order to Exploit Ray Bundle Performance

[LINK http://www.hpc-sa.com/downloads/RT07_paper1006_final.pdf]
_________________
GC

If you don't understand me, perhaps there is nothing to understand

(L) [2007/09/19] [Shadow007] [RT07 Papers List] Wayback!

Well found !

more precisely : (with a link to the slides as well)

Coupled Use of BSP and BVH Trees in Order to Exploit Ray Bundle Performance

[LINK http://www.hpc-sa.com/index.php?lang=eng&page=downloads_publi]

(L) [2007/09/19] [Shadow007] [RT07 Papers List] Wayback!

>> hanatos wrote:Shadow007 wrote:It seems to me it's the real one
..that's like your opinion, man..
(slap on the head)

Hey, that's not fair ! It's not as if Hanika were a registered user in the forum ... is it ? [SMILEY ;)]
Well if anyone better informed than me tells me it's not the same ...

(L) [2007/09/19] [hanatos] [RT07 Papers List] Wayback!

>> Shadow007 wrote:hanatos wrote:Shadow007 wrote:It seems to me it's the real one
..that's like your opinion, man..
(slap on the head)
.
sorry for that... just love the big lebowsky, who would continue:
``i'm not mister lebowsky, man... i'm the dude, and that's what you call me..''

(L) [2007/10/23] [Shadow007] [RT07 Papers List] Wayback!

As seen on Huang's papers list, the [LINK http://www.sci.utah.edu/~butler/ The Bullet Ray Vision] paper is also available. (Link at the bottom right), as well as the paper's [LINK http://www.sci.utah.edu/~abe/rt07/2007_RT07_BulletVision-H.264_LAN_960x540.mov movie].

RT07 Papers List back

Board: Board index ‹ Ray tracing ‹ Links & papers