Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!usc!apple!motcsd!starnet!sschaem From: sschaem@starnet.uucp (Stephan Schaem) Newsgroups: comp.sys.amiga.programmer Subject: Re: Lemmings - a tutorial Part IV Message-ID: <1991Mar31.195148.6667@starnet.uucp> Date: 31 Mar 91 19:51:48 GMT References: <23787@well.sf.ca.us> Organization: Starnet-Public Access UNIX-Los Altos,CA 415-949-3133, login:info Lines: 139 This is getting out of hand!So let me state my personal view. I offer here a way to everybody to make me shutup and I'm sure people wont resist it:-) But persoanl attack on what I do without seing my work is pretty low! I meet Mike Swarthz not a long time ago and we agrea on 100%, he has seen some of my work and frankly only him on the net can give a true value of what I do (good or bad). Rob Peck also visit me 2 year ago to see my work on 4096 hires display. And I'm not stoping anybody:-) I will explain my view clearly, numbered and detailed.So if somebody dont agrea or have more to say it will be clear to follow and we wont get into a big mess. 1)MULTITASKING Some game wont allow you to be multitasking because of the resource they need.The most dificult part is making a custom display fit in the intuition screen enviroment. The OS do not offer enought speed to handle the VP coper list, and the OS do not offer routine to handle special display mode like RAW display or diferent bitplane display. This is mostly a speed issue. Then come the memory issue.Sometime you need special memory access. Like BANKS to keep 16bit pointer address.That require a special version of the allocmem routine to offer obsolute alocation. 16 bit pointers are not alway PCL or STRL (PC Labels or Sructure Labels). And very often you will find your bank already used by the OS (or aplication) and (gladly) the amiga OS dont offer what the MAC offer even if it would be nice in that case. Sharing the blittter is posible, copper is possible... Almost everything is possible by cutting resource for the game. But that require ALOT of work and in the final round will not pay, because competition can offer a better game by not doing so. So this is mainly a time, speed, memory problem.Development time mean money invested, and incentive for that require sales to cover the cost. And from the curent market that doesn't apear true. But maybe other things keep compagnie not to do so? like lazyness or lack of knowledge? I think that people offer what the majority of the customer want, otherwise it would change. 2)OS KILLING Not a necesity for every cases for ANY product.Has long has you have the memory to keep a copy of both in memory you can alway return to the OS.(But is not valid for some system, like server or other of course!) So a game loading from your HD by clicking on a WB icon, and returning to the WB when ever you want is more then possible. The problem is mainly piracy (piracy traduce with loose of money). And including lookup is something that not many will agrea on. Also will need 2 programing style because of 512k users that cannot keep both in memory at once. But that involve a return to the OS each time you need OS resource, like HD access... So a game can also boot from a HD but not also use it at all time. 3)HD SUPPORT Possible in every case.(even if its not at all time) And again piracy is the key to this one. My point will be: why is overdrive curently running from my HD and return to the OS and not the comercial version? I thought people couldnt figure better than lazyness since doing so involve floppy programing. And dont forget the 80K people are talking about, does it include your mounted drive with buffers?! So 1meg will be needed to run 512K games... 4)MACHINE SUPPORT The only point that a programer can be accused of.(that's what start me to reply to Mike Farren). If a game dont work on ANY amiga its because of lack of knowledge of the other CPU and lack of programing knoledge.Or strick OS rules folowing. It doesn't involve development time. 5)MYSELF :-) I do both :system programing and hardware programing. NONE of my program involve reboot, or dont offer A500 to A3000 support. My aplication work on V1.0 has on V2.0 and from what people tell me are complex tools (if you consider programe like DPIII complex.:-) I will be more than glad to show to ANYONE my work in both area. If badge people rey show on 4096 HIRES images in multytasking (on a A1000 with a 25MHZ lucas board) ported from standart 24bit IRIS images I will be glad to show them all the tools, library, editor, duplication software we developed for OVERDRIVE. And also OVERDRIVE starting and returing to the OS gracefully. And also talk about the 400 functions of our amiga library. Also from some request: Musician name Frederic Bouquerel (FRED) GFX Sort Ty PRG Alex Wilky Fred sell his work to software developers, and to music for his group and campagnie Exponantia. Sort Ty participate in 'Wild Street' and 'Fire and Forget II' Alex Wilky , Nitendo reversed engeniring, and custom MAC software. Myself, the lastest is 'Dark Century'. (I do not recomand ANY of the 3 stated game!) We are not proude of the port we done, and thats why we quit doing 3 month PC port and started our own compagnie doing AMIGA ONLY games. We fellomfortable with MOTOROLA macro assembly and really find a incredible pleasure with it, unlike C where its more of a job. People of both worls now what I mean.Especially with macro assembler. For the other check condional assembly, constant definition, structure definition, function call etc.... I took advanced C and advenced 680x0 classes in college and kept ASM even if my early amiga work was done in C... And if you have a good algorithm a C implementation will alway be slower and bigger than an optimized ASM one.(even if its not by mutch). Stephan.