Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!ulowell!cbmvax!jesup
From: jesup@cbmvax.UUCP (Randell Jesup)
Newsgroups: comp.sys.amiga.tech
Subject: Re: BltMaskBitMap function needed
Message-ID: <5894@cbmvax.UUCP>
Date: 3 Feb 89 21:06:12 GMT
References: <254@vixen.uucp>
Reply-To: jesup@cbmvax.UUCP (Randell Jesup)
Organization: Commodore Technology, West Chester, PA
Lines: 19

In article <254@vixen.uucp> joe@vixen.UUCP (Joe Hitchens) writes:
>Or alternatively, I could use BltMaskBitMapRastPort() if I knew that there
>was no significantly greater overhead in using it, when the RastPort in
>question was allocated and initialized by my own code, i.e., not attached
>to a Window, I allocated the RastPort & BitMap structures, and Raster data
>myself.  I know that a Window's RastPort means that clipping will be done,
>and that means slower performance.  If I built the RastPort myself without
>attaching any Layer info to it, will it still be much slower than blitting
>to a BitMap?  I want to handle the clipping myself.

	The rastport shouldn't be noticably slower than bitmaps if there is
no attached layer (clipping).  However, BMBMRP() isn't the fastest call in
the world, since it handles arbitrary rectangles it can't use a cookie-cutter
blit, and therefor uses 3 blits to accomplish the end result.  If your
alignments work out ok, you can use a cookie cutter blit in one pass, but you'll
have to program the blitter yourself.

-- 
Randell Jesup, Commodore Engineering {uunet|rutgers|allegra}!cbmvax!jesup