Path: chuka.playstation.co.uk!news
From: Chris Chadwick <xjk18@dial.pipex.com>
Newsgroups: scee.yaroze.programming.libraries
Subject: Re: MoveImage() M.O.
Date: Sat, 21 Mar 1998 00:11:24 -0800
Organization: PlayStation Net Yaroze (SCEE)
Lines: 78
Message-ID: <3513762C.7432@dial.pipex.com>
References: <350F9A4B.2062@dial.pipex.com> <6eqlpq$l5r13@chuka.playstation.co.uk>
NNTP-Posting-Host: ae195.du.pipex.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 2.02 (Win95; I; 16bit)

SCEE wrote:
> 
> Hi Chris,
> 
> Using MoveImage that way is fine as the libs copy the rect structure
> internally.

Brill! I was hoping that was the case.

Thanks, Colin.

-Chris


> ( Incidentally you could do something similar using fast sprites, as once
> you have issued the GsSort.. call you can then freely change the structure )
> 
> Cheers,
> 
> Colin.
> 
> Chris Chadwick wrote in message <350F9A4B.2062@dial.pipex.com>...
> >Hi,
> >
> >I currently need to display a sprite object in my game which is
> >effectively a small text box. I'm doing this by copying (with MoveImage)
> >each text character to the box sprite image in VRAM, from a font image
> >located elsewhere in VRAM. I actually copy a single character image AND
> >a cursor image once every couple of vsyncs, to give a 'computer read-out'
> >type effect. I'm doing it this way as I obviously dont want to have a
> >separate sprite for each character etc!
> >
> >So, the basic run of events within the main game loop is:
> >
> > game loop start
> > .
> > .
> > .
> > MoveImage() // to copy character image to box sprite image
> > .<alter RECT var>
> > MoveImage() // to copy cursor image to box sprite image
> > . // Note: I dont/cant use a DrawSync()
> > . // after either MoveImage() call
> > .
> > DrawSync()
> > Vsync()
> > GsSwapDispBuff()
> > GsDrawOT()
> > .
> > .
> > .
> > game loop end
> >
> >I modify the source RECT used by the first MoveImage and use it with the
> >second MoveImage. This seems to work BUT, isn't it possible I will have
> >modified the RECT variable BEFORE the first MoveImage operation is done,
> >as it is non-blocking? Or, does each MoveImage call keep it's own working
> >copy of the RECT variable passed to it? I assume some such mechanism is
> used
> >as when each MoveImage call is made (as above) the GPU will be busy
> >because of the last GsDrawOT call, so the 2 MoveImage operations will
> >be delayed until AFTER the GPU has finished drawing. Is this right?
> >Is it QUARANTEED that any MoveImage call issued AFTER a GsDrawOT will
> >be delayed until the GPU has finished drawing ALL the OT objects?
> >I dont want to be altering a sprite image while/before the GPU has used
> >that image in drawing the previous frame (if you see what I mean!?)
> >
> >I hope that lot makes some kind of sense to someone. :)
> >I quess, what I'm asking in essence is: How and when, EXACTLY, are
> >MoveImage operation requests stored and subsequently dealt with by the GPU?
> >
> >As I say, everything SEEMS to be working fine the way it is (things that
> >run first go give me an uneasy feeling :) I just want to be sure that
> >things wont go pear-shaped somewhere down the line.
> >
> >Thanks in advance!
> >
> >-Chris