[Bf-committers] Proposed Method for GI in blender internal

[Arne Schmitz]

--Boundary-02=_or6wA1W4WE0lcdC
Content-Type: text/plain;
  charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

Am Freitag, 4. Juni 2004 23:15 schrieb Ray Wells:
> I came up with a cool idea for quick GI in blender
> internal..

I do not really get what is new to your idea. Especially your argument that=
=20
one need billions of photons, or otherwise the information will be useless.=
=20
Why do you think you're doing a k-nearest neighbour search in photon mappin=
g=20
then?

In fact I do not see any advantage or big difference of your proposal compa=
red=20
to photon mapping, be it time or space complexity -- if I could figure out=
=20
what the complexity of your algorithm would be...

I agree with Anders that photon mapping is the way to go becaus it is=20

a) quite simple (no difficult math at all)
b) efficient, provided you use some appropriate search structures
c) allows visualization of
	1) diffuse interobject reflections
	2) specular reflections with caustics
	3) volumetric effects such as fog or subsurface scattering

Cheers,

	Arne

=2D-=20
[--- PGP key FD05BED7 --- http://www.root42.de/ ---]

--Boundary-02=_or6wA1W4WE0lcdC
Content-Type: application/pgp-signature
Content-Description: signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.2 (GNU/Linux)

iD8DBQBAw6rozpP3gv0FvtcRAvvUAKC0aukL7ZzhMIQWFDp4pNLQNWzRpACeOZ+0
TgjbgHYe2i1OylG3gGwAZEU=
=5u6H
-----END PGP SIGNATURE-----

--Boundary-02=_or6wA1W4WE0lcdC--