Beamtracing etc

(L) [2007/09/16] [Phantom] [Beamtracing etc] Wayback!

On the GameDev forum, someone mentioned that in the field of acoustics, people are looking at beam tracing as 'the' way; it properly solves the aliasing problem caused by the point sampling which rt is (I believe this problem was brought forward on this forum as well). It struck me that I never read any papers in different but related fields, such as acoustics and collision detection (perhaps there are more related fields?), and I have a feeling that this is not common to do in general.

So... I want to read up on acoustics. I'm going to find some papers, but since there are people on this forum that are already active in this field, are there any good papers that you recommend as a quick introduction to the field in general and perhaps beam tracing in particular? If I find cool stuff, I'll post it too, obviously.
_________________
--------------------------------------------------------------

Whatever

(L) [2007/09/16] [Wussie] [Beamtracing etc] Wayback!

Well, I wouldn't really know about the papers, but I'd imagine beamtracing could be a valuable addition to a current raytracing platform. Although I'm not terribly familiar with the field, I'd imagine that similar algorithms and methods could exists for say acoustics and caustics, they both rely on deflecting (parts of) a ray/beam and detecting a new point of impact. Looking forward to some reading material [SMILEY Smile]

(L) [2007/09/16] [nAo] [Beamtracing etc] Wayback!

Interesting read:

[LINK http://www-csl.csres.utexas.edu/gps/publications/beamtrace-egsr07/ A Real-time Beam Tracer with Application to Exact Soft Shadows]
_________________
Isn't it enough to see that a garden is beautiful without having to believe that there are fairies at the bottom of it too? [Douglas Adams] [LINK http://pixelstoomany.wordpress.com/ [my blog]]

(L) [2007/09/16] [Phantom] [Beamtracing etc] Wayback!

Here's a first paper on the subject:

[LINK http://people.csail.mit.edu/addy/research/funk-beamtracing-jasa04.pdf]

('A beam tracing method for interactive architectural acoustics')

It discusses both beams and rays. Year is 2003. Interestingly enough, it quotes ray tracing (gfx) papers.

EDIT: simultaneously posted with nAo. [SMILEY Smile]
_________________
--------------------------------------------------------------

Whatever

(L) [2007/09/16] [Zakalwe] [Beamtracing etc] Wayback!

The problem as I understand it with beam tracing is that while the beam starts out simple as a cone or frustum it quickly becomes more complex as it is partially occluded by objects. Now you can of course subdivide the complex beam, but there's overhead there.

MLRTA operates in a similar enough fashion if you think about it, but the frustum always remains cuboud in cross section if you perform splitting.

(L) [2007/09/18] [bitshit] [Beamtracing etc] Wayback!

Beamtracing sure looks interesting, there where two papers released on it not so long ago, see:

[LINK http://research.microsoft.com/research/pubs/view.aspx?id=1276&type=Technical+Report]

and

[LINK http://www.cs.columbia.edu/cg/pdfs/133-BeamTrace.pdf]

They use the GPU for rasterizing the final result, but if I understand correctly raytracing could be used as well.

(L) [2007/09/18] [davepermen] [Beamtracing etc] Wayback!

instead of beamtracing and perfectly adjust the beam to all triangles and everything, what about using the beam to "collect" the triangles for a region, and then quickly trace only on those.. should enable quite high supersampling where needed, without tracing into the full scene all the time.. i think the vertex_culling.pdf presented recently covers this..

(L) [2007/09/19] [dr_eck] [Beamtracing etc] Wayback!

If anyone wants the code from Reshetov's article, all pretty and compilable, just pm me.
_________________
Opinions? Those are *facts* son.

Beamtracing etc back