Информационный сервер для программистов: Исходники со всего света. Паскальные исходники со всего света
  Powered by Поисковый сервер Яndex: Найдется ВСЁ!
На Главную Pascal Форум Информер Страны мира
   Demo Making    >>    goushad
   
 
 Gourad Shading Effect    Faker/S!P 01.09.1994


How to code gouraud shading. Full pascal source included.
Coded by The Faker/S!P. Fast iNjEcTioN By Access/Antares



14k 
 

Hello Dudes ... I was away for a longer weekend and when I came back I saw there was quite some discussion about certain aspects of my person and my code. I remember posting some controversial statements which I thought would lead to some discussion and widen the horizon of some people who only believe the things what the computer magazines and the so-called gurus say. Well, there was some reaction I didn't quite like - some guys reacted quite childish in the way of this-is-not-the-way-it's- -done-therefore-you-are-an-idiot-and-we-laugh-at-you. Way under there age (or?). Which made me quite angry so I replied hard at them. Most of it was crap anyway - lamers who thought they could prove me wrong and made themselves laughed at by others, when they tried to improve the 16bit code in 32bit mode. NONE of it works, as I was claiming before and was flamed with crap from the people who have to boost their ego. Another thing was the statement about the Gouraud code with 0.25 instructions per pixel. I wrote that only 'coz I was surprised that people still thought it was so expensive ... while I replaced it with Phong shading already. According to the replies I got, where people actually thought I meant 1/4 of a frame and stuff like that, they act like when-I-cannot-do-that-how-can-HE-do-that-MUST-be-some-sort-of- -error-or-else-I'd-be-lame. Some people really think that their code is optimized or so, and nothing can get faster - well it's not impossible to optimize even more, you'd be surprised. NEVER EVER say your code is optimized, somebody can come and make it faster - and if it is with the new Pentium execution unit, or with some new undocumented feature. Therefore below's the code for it. You may want to get my released sources/intros and look at them. Then let's argue again ... Something else about giving code out. It's definitely *NOT* in use for demo groups to give code out. While it may be 'in' for some American demo groups to do that, we Europeans started out without all those tutorials - most of us don't have Internet access or a modem anyway. Enough about this. But I find today's attitude of 'HAVING to give code out, if you mention you have a good method' lame - there's a increasing majority of newbies who actually DEMAND that, 'coz they're used to it. Like the 'if he holds his code back we'll flame him till he gives us!' approach. I fucking hate those LAMERS. Try to see it as a GIFT, and not as your right! Nevertheless I released some, and I'll probably still do ... However, would you give your code/technique/ tricks gratefully to guys who flame you? Most of those lamers don't have anything to prove they can code, actually. Everybody can snap up a few bits of 'on how to do this and that', and no matter if it's wrong or right, flame all who are of a different opinion. I think that's what is called being fascist. Let's take for example DOOM: Lot's of people 'claim' they've coded/are coding it - I don't know who turned up with the argument that DOOM be ray casting, but I bet the ID guys were laughing their ass off when that thread about it was taking place ;) It may lie in their interest to disinform the public as they want to sell their routine's technology, or? (This is only an assumption and no accusation, dudes). You probably could make it ray casting, but I'd strongly doubt you'd reach the speed of the original. Well, how do I come to that conclusion? Me, and some of the leading demo coders agree on that. (There's an example with ray casting, called ACK3D, but it doesn't reach the speed of Wolf3D by far, as you can see, and for floor/ceiling the ratio is worse ...) Laugh at me, but as a demo coder I'm testing algorithms due to their usability and performance. And I don't select the most sophisticated one, but the one who fulfills the needs of the routine. For example, I've never bothered with BSP- trees - I know about the algorithm - but I see no use for it. What I'm trying to say, you shouldn't blindly follow those who call themselves Gurus, but try to look what's behind it. I know that some guys will flame me, either for this attitude, or for some little bugs they find in my routine, or some unoptimized ASM instructions. Those fuckers should really get a life. I thank all those who know me, have seen my routines and support me in this group - You know who you are! ----------------RIP this code here, lamers----------------------- Signed, The Faker (S!P Internet PR) _____________________________________________________________ \ \ \ | "No one told you when to run, | in fake life: | | you missed the starting gun." | Stefan Ohrhallinger | | | St. Laurenz 54 | | SURPRISE! PRODUCTIONS, AUSTRIA | A-4950 ALTHEIM | | | | | "lightyears ahead!" | +43-732-2457-1025 | | __________________________________\_______________________\__ \_/____________________________________________________________/ I really don't care what you're doing with it, 'coz for me it's obsolete ... Why? It's been coded a year before, I never optimized anything except the inner loop, so my Phong stuff is faster now. And it's an example of provement, not a full-documented well- structured nice- ascii-pictured anal-retentive code - I've got better things to do. compile: tp -G+ gourex.pas run: gourex sphere 2 x g