home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!munnari.oz.au!spool.mu.edu!sdd.hp.com!swrinde!gatech!mailer.cc.fsu.edu!sun13!cs.unc.edu
- From: leech@cs.unc.edu (Jon Leech)
- Newsgroups: comp.graphics.research
- Subject: Re: Scalability
- Message-ID: <11602@sun13.scri.fsu.edu>
- Date: 24 Dec 92 17:58:48 GMT
- References: <11588@sun13.scri.fsu.edu>
- Sender: news@sun13.scri.fsu.edu
- Lines: 47
- Approved: murray@vs6.scri.fsu.edu
- X-Submissions-To: graphics@scri1.scri.fsu.edu
- X-Administrivia-To: graphics-request@scri1.scri.fsu.edu
-
-
- In article <11588@sun13.scri.fsu.edu>, rms@well.sf.ca.us (richard marlon stein) writes:
- |> I would like to PixelPlanes hammer on a 1 Terabyte dataset.
-
- It would take an awfully long time, even if we had that much
- memory :-) I'm kinda hoping to play with large terrain datasets on
- Pixel-Flow, since they're a ready source of interesting, complex
- geometry. I don't think even the Magellan Venus global topography map
- approaches 1 TB, however.
-
- I'm also not sure what your point is. You originally made a claim
- that Pixel-Flow was not "truly scalable", which I and Marc Olano, who
- are both members of the Pixel-Flow team, have refuted. Pixel-Planes 5
- is an older and different architecture. Part of the purpose of PxFl is
- to build a machine that does not suffer from the constraints and
- problems we learned about from Pxpl5.
-
- Switching gears (maybe this should be a different thread):
-
- |> One things for sure though, I know when I've been beat. Since the SCVS
- |> does not "subdivide" the screen by assigning processors to a specific
- |> region of a single display surface, I don't think you are in a
- |> position to judge the architecture. Ask Molnar if he'll sign the
- |> non-disclosure agreement, and then you can judge.
-
- I started with your one-sentence description:
-
- "I advocate a tessellated display approach myself, where each
- tessellation element owns a processor, framebuffer, and
- display."
-
- and made some comments about inherent problems with the
- screen-space decomposition this *appears* (that's why I said "If you
- mean that...") to refer to. We have a lot of experience with these
- problems, since Pxpl5 suffers from them.
-
- I'd be interested in hearing what SCVS actually is, but if you
- won't or can't provide any more detail, there's no basis for further
- discussion. Good luck building your machine. I look forward to
- discussing it whenever you can describe it in public.
-
- Jon (leech@cs.unc.edu)
- __@/
-
- --
- Moderated by SCRI Vis <> Submissions to: graphics@scri1.scri.fsu.edu
- Guy, John R. Murray <> Administrivia to: graphics-request@scri1.scri.fsu.edu
-