home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Light ROM 4
/
Light ROM 4 - Disc 1.iso
/
text
/
maillist
/
1994
/
july94.doc
/
000084_owner-lightwave-l _Fri Jul 8 03:57:16 1994.msg
< prev
next >
Wrap
Internet Message Format
|
1995-03-23
|
2KB
Return-Path: <owner-lightwave-l>
Received: by mail2.netcom.com (8.6.8.1/Netcom) id WAA26677; Thu, 7 Jul 1994 22:58:58 -0700
Received: from xmission.com by mail2.netcom.com (8.6.8.1/Netcom) id WAA26670; Thu, 7 Jul 1994 22:58:56 -0700
Received: by xmission.com (4.1/Xmission/SMI-4.1) id AA21524; Thu, 7 Jul 94 23:56:45 MDT
From: dingebre@xmission.com (David Ingebretsen)
Message-Id: <9407080556.AA21524@xmission.com>
Subject: Shadows VS. Smoothing
To: lightwave-l@netcom.com
Date: Thu, 7 Jul 1994 23:56:44 -0600 (MDT)
X-Mailer: ELM [version 2.4 PL22]
Content-Type: text
Content-Length: 1624
Sender: owner-lightwave-l@netcom.com
Precedence: list
Reply-To: lightwave-l@netcom.com
Hi all,
It's nice to have net access again!
Topic: Shadow maps, ray traced shadows, and smoothing
This is best described with an example. I created a sphere 32 sides
by 16 segments 1 meter in radius and jittered it by 1 cm. I exported
the sphere to Lightwave and turned smoothing on, no shadow maps, no
ray traced shadows, changed the default light into a spot with a
default cone and moved it so the whole sphere was visible in the
cone. Rendered. Looks like a bumpy ball, nicely smoothed surface.
Now, turn on shadow mapping. Ooops, I see small polygonal shaped dark
spots. Turn off shadow maps, turn on ray traced shadows, same
thing. The only difference is that the shadow mapped spots have soft
edges and the ray traced spots have sharp edges (no real surprise).
What I believe to be happening is that shadow maps and ray traced
shadows are not being smoothed over the boundary of two adjacent
polygons whose included angle is within the "smoothing angle". It is
as if these type of shadows don't know that I have enabled smoothing
for the surface.
NewTek dismissed this saying that they couldn't see or reproduce the
problem, even having sent them two disks of scenes, objects, and
images. I reproduced the effect on three systems and all who viewed
the output agreed with me that something was wrong.
Since then (this started September 1993) I've seen references to this
behavior in other publications. Has anyone else noticed this? Am I
seeing things (again)?
I've had to go to other software at times because this "digital acne"
ruined the final output.
Is this behavior fixed in upcoming releases?
David