Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!rutgers!clyde!masscomp!black
From: black@masscomp.UUCP (Sam Black)
Newsgroups: comp.windows.x
Subject: Re: V11 in-memory graphics
Message-ID: <1621@masscomp.UUCP>
Date: Mon, 13-Apr-87 11:42:33 EST
Article-I.D.: masscomp.1621
Posted: Mon Apr 13 11:42:33 1987
Date-Received: Sat, 18-Apr-87 19:23:37 EST
References: <1987.4.11.13.31.21.eddie.caplan@h.cs.cmu.edu>
Reply-To: black@masscomp.UUCP (Sam Black)
Distribution: world
Organization: MASSCOMP - Westford, Ma
Lines: 32

In article <1987.4.11.13.31.21.eddie.caplan@h.cs.cmu.edu> eddie.caplan@H.CS.CMU.EDU writes:
>V11 permits graphics to be done directly to Pixmaps, etc.  My question
>is about how this is implemented.  It seems to me that in-memory
>graphic routines do not need to make the expensive calls to the
>server.  One way this could be implemented is to have a
>device-dependent runtime library for in-memory graphics linked into
>the client's program if the client chooses.

One of the main reasons I see in having the server do PIXMAP drawing
is the sharing of resources.  Another is to avoid large executables
and relinking for different devices (assuming a different library for
each PIXMAP depth or device).  It allows devices with off-screen
memory to utilize their capabilities even for PIXMAPs.  Finally, it
provides a consistent interface for all drawing.

>I believe that this effiency issue is important, especially for those
>of us who want to (and can) create complex images in memory and then
>simply BLT them onscreen.

X11 does not prohibit this; you can use PutImage to do just what you
want.  Remember, just because the capability is there doesn't mean you
have to use it.

----------------------------------------------

There are only two kinds of planes in the world:
Fighters and targets.
			- 1Lt. Steven Brown, 63 TFTS

	...!{decvax,allegra,harvard,seismo}!masscomp!black

----------------------------------------------