i'm doing a test with depthmap shadow.
the goal is to see the self-shadowing artifact.

so i use 40x40 grid as my obj/geo, then lit this obj with a spotlight using depthmap shadow.
the resolution for depthmap i set to 1x1 pixel. i focus the light to the center of the grid.

so the concept that i learn from other like maya, that the ray is cast from light source through the center
of this dmap pixel (1x1 pixel, so this is a single pixel ).
then when ray intersect the nearest surface, it records the distance from light to that surface. this depth
measurement make up the depth map's pixel.
in this case the distance from light to the center of the grid will be recorded (coz the light focus on this center)

so when rendering, each of point to be rendered (of the grid being shaded) will be calculated : the distance from
this point to the light will be compared to the depthmap value.
if the distance greater than the stored depthmap value, then it means that point must be in shadow.
so as long as the light direction is not perpendicular to the grid ,
when rendering , half area of the surface/grid should be in shadow (self-shadowing),
but i didn't see it.

anyone can explain this?

thx.

I would think in order to do this(although I'm not sure of *why* you want to do this) you need to make your own shadow shader. First off, AFAIK, you can't have a depth map of 1x1 - the minimim is 4x4(check your shadow map with iinfo). Secondly, I would think filtering would mess up what you're trying to do. Mantra's job is to avoid having artifacts.

Cheers,

J.C.

One other thing to check is what kind of depth you're storing (i.e., nearest, mid-point, or furthest). The self-shadowing problems typically only show up when using nearest.

thanx for the response guys…
well, i'm doing this just because of my couriousity.
since the houdini help file didn't provide enough information about this,
so i have to dig it using maya as reference (coz i know a bit of how maya depthmap is working)
so base on this knowledge , i try to compare both, and hoping i'll get
deeper information about houdini depthmap

well, maybe anyone know where i can get quite detail explanation
behind this depthmap concept?

thx.

metaclay
well, maybe anyone know where i can get quite detail explanation
behind this depthmap concept?

If you're interested in how the “classical” depthmap shadow algorithm works, I think the seminal reference would be the Reeves paper:

http://graphics.pixar.com/ShadowMaps/paper.pdf [graphics.pixar.com]

And there have been a million variations since….

Here's one on percentage-closer filtering (PCF):
http://www.mpi-sb.mpg.de/~brabec/doc/brabec_cgi01.pdf [mpi-sb.mpg.de]
And a “soft-shadow” variation using PCF:
http://download.nvidia.com/developer/presentations/2005/SIGGRAPH/PCSS.pdf [download.nvidia.com]

And on, and on, and on… and then there are “deep shadows”:
http://graphics.pixar.com/DeepShadows/paper.pdf [graphics.pixar.com]

… and after that, there is google

that links are really cool,
thanx Mario