Article View: rec.games.programmer
Article #570Split screen mode: Why? (was Re: 320x400 question)
From: David M. Baggett
Date: Tue, 11 Aug 1992 13:43
Date: Tue, 11 Aug 1992 13:43
52 lines
3363 bytes
3363 bytes
In article <BstLnJ.Lv6@news.cso.uiuc.edu> tmkk@uiuc.edu (Khan) writes: >There's no way I know to draw to both pages at once. What you have to do >is draw your status info to one page and then draw it again to the >other. This takes twice as long to do. OK, now I see what you were getting at. You are doing twice as much work to update score counters and any other status information because this "static" stuff needs to be drawn to both (all) page flip screens at the "same time." Another complexity of course is that you don't *want* to draw on all the page flip screens at once, since one of them will be visible at any given time. (And drawing to a visible page produces flicker.) You need to update your status information on all the nonvisible page flip screens and then save the changes so you can apply them to the currently visible screen once it's no longer visible. But this is just a price to be paid for page flipping -- I don't see how a split screen would eliminate this problem either. If you make the status display half of the split screen a static image (i.e., not page flipped) then you've got a flicker problem. If you make it page flipped, then you're back to the same problem discussed above: you need to "remember" areas of the window you've changed and apply those changes to each page flip screen as it becomes nonvisible. On the Atari, at least, you *must* page flip; if you don't the results are absolutely intolerable. Perhaps multisync monitors or the VGA hardware itself makes the flicker less obvious on the PC? (My impression is that this is so.) So I guess I'm still curious what the advantage is to the split screen mode. One might be that you could use hardware panning on just one half of the split screen (instead of the whole screen). So you could have hardware scrolling *and* a status window, without having to resort to drawing the status window as though it were a sprite (i.e, every frame) like Keen 4 through 6 do. >BTW, I gather that some games do use this technique. There was an issue >of GameBytes which had a screen shot of some game (might even have been >Wolf, I can't remember now). The upper portion of the screen showed the >normal stuff, but the bottom of the screen was garbage. The caption to >the screen shot apologized for the incomplete picture, saying that the >split-screen mode used by the program was incompatible with their screen >capture program. This could also be caused by a mid-screen palette change. The screen capture program would have the data for the bottom half (or third, or whatever) but not the correct palette, so it would look like garbage. Neil Forsyth pointed out to me that Lemmings changes the palette at the top of the icon box -- you can see a flickery line if you look closely. A mid-screen palette interrupt has definite (and obvious) advantages. Dave Baggett dmb@wam.umd.edu
Technical Details
Message-ID:
<1992Aug11.134312.16390@wam.umd.edu>Path:
arc1-ca.pugleaf.net!sparky!uunet!haven.umd.edu!wam.umd.edu!dmbShow Full Headers
Complete Article Headers
Newsgroups: rec.games.programmer Path: arc1-ca.pugleaf.net!sparky!uunet!haven.umd.edu!wam.umd.edu!dmb From: dmb@wam.umd.edu (David M. Baggett) Subject: Split screen mode: Why? (was Re: 320x400 question) Message-ID: <1992Aug11.134312.16390@wam.umd.edu> Sender: usenet@wam.umd.edu (USENET News system) Nntp-Posting-Host: rac2.wam.umd.edu Organization: University of Maryland, College Park Date: Tue, 11 Aug 1992 13:43:12 GMT Lines: 52