9fans - fans of the OS Plan 9 from Bell Labs
 help / color / mirror / Atom feed
From: Paul Lalonde <plalonde@telus.net>
To: Fans of the OS Plan 9 from Bell Labs <9fans@cse.psu.edu>
Subject: Re: [9fans] A sad story and a question
Date: Sat, 10 Nov 2007 21:34:01 -0800	[thread overview]
Message-ID: <971A7853-CE8C-4ED2-9937-AD217C3730E2@telus.net> (raw)
In-Reply-To: <7871fcf50711101712i519ff70cx1f5befbff7f50f59@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Modern graphics libraries essentially treat the GPU like a remote  
host - even over fast PCIe it's far enough away that when you're  
aiming for 16.6ms frame times you can't afford much bus traffic.   
Instead, the intention is that most of your data is resident in GPU  
memory and only a small fraction of the used data is transfered from  
the host every frame.   9P is probably not ideal, but the advantages  
of being able to refer to resources so easily could be compelling.   
I'd love to be able to mount a tar file of textures and vertex  
buffers and have a cache on the GPU manage those for me.  It's a much  
simpler user model than the current myriad ways of setting up  
different data types.

Paul

On 10-Nov-07, at 5:12 PM, Joel C. Salomon wrote:

> On Nov 10, 2007 6:28 PM, Paul Lalonde <plalonde@telus.net> wrote:
>> If I were going the route of enabling 3D graphics on Plan9 I'd
>> probably flat out ditch the OpenGL model.  Instead...
>
> The little I know of OpenGL led me to think that it was also created
> to be run over a network, but I have no idea how big a mismatch there
> is between the model it assumes and 9P.  I assume you have a better
> view of things?
>
> --Joel

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (Darwin)

iD8DBQFHNpRJpJeHo/Fbu1wRAqqYAJ4n79e2BG5QJ8AtGYuVtw8fAuvQPgCgu/ZY
aamBhTVOi4qVztGmFZJS1Zk=
=5DLh
-----END PGP SIGNATURE-----


  reply	other threads:[~2007-11-11  5:34 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-11-10  2:13 Pietro Gagliardi
2007-11-10  2:49 ` geoff
2007-11-10  3:00   ` Pietro Gagliardi
2007-11-10  4:50     ` ron minnich
2007-11-10  4:55       ` Pietro Gagliardi
2007-11-10 13:04         ` erik quanstrom
2007-11-10 17:31         ` ron minnich
2007-11-10 18:58           ` David Leimbach
2007-11-10 19:00             ` Pietro Gagliardi
2007-11-10 23:48               ` David Leimbach
2007-11-12  3:19                 ` Russ Cox
2007-11-10 19:24             ` Bakul Shah
2007-11-10 19:46               ` andrey mirtchovski
2007-11-10 21:20                 ` john
2007-11-10 23:28               ` Paul Lalonde
2007-11-11  0:27                 ` Anthony Sorace
2007-11-11  1:12                   ` Paul Lalonde
2007-11-11  1:12                 ` Joel C. Salomon
2007-11-11  5:34                   ` Paul Lalonde [this message]
2007-11-12 20:32                 ` 3d graphics (was " Bakul Shah

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=971A7853-CE8C-4ED2-9937-AD217C3730E2@telus.net \
    --to=plalonde@telus.net \
    --cc=9fans@cse.psu.edu \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).