home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.sgi
- Path: sparky!uunet!spool.mu.edu!agate!ames!sgi!fido!paris.asd.sgi.com!alain
- From: alain@paris.asd.sgi.com (Alain Dumesny)
- Subject: Re: GLRenderCaching (Inventor)
- Message-ID: <pk01h5k@fido.asd.sgi.com>
- Sender: news@fido.asd.sgi.com (Usenet News Admin)
- Reply-To: alain@sgi.com
- Organization: Silicon Graphics Inc.
- References: <MATUMOT.92Sep4190519@calypso.elec.nkk.co.jp> <pedrj4g@fido.asd.sgi.com>
- Distribution: comp
- Date: Tue, 8 Sep 1992 23:40:17 GMT
- Lines: 18
-
- In article <pedrj4g@fido.asd.sgi.com>, mott@sgi.com (David Mott) writes:
- |>
- |> The key here is that your geometry *changes dynamically*. If you
- |> build a cache for the geometry, then change it, the cache is blown,
- |> and has to be rebuilt. Building the cache is more expensive than
- ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
- |> simply rendering the geometry.
- |>
- |>
- |> David
- |>
- |>
-
- What Dave Mott meant to say was that "Building the cache and rendering it
- is more expensive than simply rendering the geometry".
-
- -------------------------------------------------------------------------
- Alain Dumesny alain@sgi.com (415)390-5250 Silicon Graphics Inc.
-