Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!samsung!spool.mu.edu!munnari.oz.au!bruce!trlluna!shiva!soh From: soh@shiva.trl.oz (kam hung soh) Newsgroups: comp.sys.amiga.programmer Subject: Re: OOP (was Re: Assembly Language & Programming) Message-ID: <3383@trlluna.trl.oz> Date: 17 Apr 91 02:04:15 GMT References: <67@taloa.unice.fr> Sender: news@trlluna.trl.oz Lines: 34 beust@mimosa.unice.fr (Cedric Beust) writes: > Mmmh... Yes and no. As you say below, this can be a very expensive price >to pay for pure arcade games. I can hardly imagine Psygnosis games written >in something else than ASM. But it is always possible to write most of >the game in C++ and write critical parts in ASM. That is a sensible approach. Writing any major program involves a lot of money and / or time, and there is no reason to write in assembly unless there is an advantage. > The real question for me is "Is there any interest in reusability for >games?". Isn't there a risk to see bunches of clones invade the market, that >are just different by the colors chosen and their titles? Reusable code does not necessarily mean that cloning is any easier. Unless the source code for a program is available, there is no AUTOMATED way of generating a duplicate (short of copying). > Of course, companies re-use part of their code previously written when >they develop a new game, but would it be a large gain if these portions >were written in C++ instead of just being simply callable ASM functions? That is a long drawn argument that erupts once in a while in various comp.lang.* groups. My impression is that a purely functional paradigm is often not sufficient to abstract essential features of a tool. While the algorithms for manipulating say, a list, are usually fixed, the callable functions you've described will have to be rewritten for different types of lists. Regards, ------------- Soh, Kam Hung email: h.soh@trl.oz.au tel: +61 03 541 6403 Telecom Research Laboratories, P.O. Box 249, Clayton, Victoria 3168, Australia