home *** CD-ROM | disk | FTP | other *** search
- Path: columba.udac.uu.se!sabik!m93kma
- From: m93kma@sabik.tdb.uu.se (Kristofer Maad)
- Newsgroups: comp.sys.amiga.programmer
- Subject: Re: Demo/game to OS frien
- Date: 7 Feb 1996 11:36:30 GMT
- Organization: Uppsala University
- Message-ID: <4fa2ru$13mk@columba.udac.uu.se>
- References: <4f6r3u$2db@sinsen.sn.no> <4f79mo$emb@serpens.rhein.de>
- NNTP-Posting-Host: sabik.tdb.uu.se
- X-Newsreader: TIN [version 1.2 PL2]
-
- Michael van Elst (mlelstv@serpens.rhein.de) wrote:
- : tbk@sn.no (Thore Bjerklund Karlsen) writes:
-
- : >If you assume an
- : >Amiga with native display, blitter and CIA should be there.
-
- : There you are. Assumptions instead of knowledge.
-
- : >Well known hardware.
-
- : So well known that most c0d3r stuff breaks on hardware they didn't have for testing
- : (and sometimes even there).
-
- Well, this discussion is getting a bit like a holy war, or so it
- seems. I thought I might as well jump in with my two cents here...
-
- The point in demo programming getting the _most_ out of a certain
- hardware configuration. As soon as the performance of commonly used
- machines is increased, new demos are released to utilize it.
-
- Tell me, if I write a demo that crams the most out of a stock A500
- with a rather complex display, would anyone be happy if it worked on
- an A4000 looking exactly the same? I don't think so. The point is, to
- the people on the seen the visual appearance of a demo is a rather
- minor point. It's if it is _hard_ (or, preferably _impossible_) to do
- that counts. (This is my impression, anyway, but feel free to correct
- me if I'm wrong...)
-
- So, let's not deny that things will _always_ run faster (if it runs at
- all) on a certain machine if the code is hand optimized for that
- special piece of hardware with the best algorithms. This is quite
- obvious, I think.
-
- On the other hand, if you want to write software for anything else
- than to show off the limits of the hardware, then you should of course
- instead put the main emphasis on the compatibility issues. The loss in
- speed will mainly hit the low-end users, but if you concentrate on
- using good algorithms, the result will be acceptable anyway. For
- example, take Frontier. It runs with the workbench in the background -
- there's even a patch available that lets you quit the game and return
- to the workbench. So, it _is_ possible to do good games fully OS
- compliant.
-
- I think that people used with hardware bashing are a great resource to
- us, and that they could do a great job implementing algorithms that
- are OS compliant. Of course, they will want to continue to develop
- on-the-bone demos just to see where the absolute limits are...
-
- Flame, anyone? ;)
-
- /Kristofer
-
- --
- Name: Kristofer Maad | "The satisfaction in making a 'crash-
- Snail: Stureg. 12A, 3tr. | proof' system screw up completely
- 753 14 Uppsala | almost makes it worthwhile." /Kris
- Phone: 018-51 11 44 |
- E-Mail: m93kma@student.tdb.uu.se | <http://www.update.uu.se/~kris/>
-