home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: sci.virtual-worlds
- Path: sparky!uunet!usc!rpi!usenet.coe.montana.edu!news.u.washington.edu!milton.u.washington.edu!hlab
- From: bcsaic!davidk@cs.washington.edu (Dave Kerlick)
- Subject: Re: SCI: Voxels vs. Polygons
- Message-ID: <1992Aug13.002747.16412@u.washington.edu>
- Originator: hlab@milton.u.washington.edu
- Sender: news@u.washington.edu (USENET News System)
- Organization: Boeing Computer Services AI Center, Seattle
- References: <1992Aug5.054046.958@u.washington.edu> <BsKD6F.4JA@watserv1.
- Date: Tue, 11 Aug 1992 17:01:03 GMT
- Approved: cyberoid@milton.u.washington.edu
- Lines: 27
-
-
-
-
- >>[many valid observations about the virtues of voxels deleted]
-
- Voxels are hardly a panacea for scientific work, and are probably most
- appropriate where the metaphor is imagery. Voxel arrays are basically
- uniform in scale, dimension, and optical thickness. They are not very
- good for situations (like visualizing viscous flows) where some cells
- are .0001 as thick as others. Unless the "jelly" inside the thin
- voxels is made much more opaque, then the are simply not seen. Also,
- it is difficult to get spatial relationships from looking at voxels.
- This is not so bad when you are looking at a CAT scan and know e.g.
- where the ears are, but not so good if you are looking at a casting
- and want to localize the defects.
-
- For quantitative work (e.g. estimating the volume of a tumor) one has
- to go to some implied geometrical representation anyway.
-
- Probably 3-D texture maps are a more flexible idea for general visual
- simulation including VR. One could keep several scales around
- (mipmaps) and transparency.
-
- --
- G. David Kerlick, Research & Technology Boeing Computer Services
- P.O. Box 24346, MS 7L-24, Seattle, WA 98124-0346
- Work phone: 206-865-5051 Fax: 206-865-3327 Email: davidk@espresso.boeing.com
-