larc.csci.unt.edu— #1 Real game programmers aren't afraid of mathematics.#2 Real game programmers use C++.#3 Real game programmers aren't afraid of learning new things. etc. etc.
Aug 3, 2006View in Crawl 4
"Come on. What if there are hidden bugs. Moreover newer games are more focussed on accurate physics."It's graphics programming... there are no hidden bugs... or physics.
Er superbouncing is a geometry exploit. Not a feature. And my point about the graphics programming one, where if something looks right, it is right being false... I was trying to articulate (and failing miserably) the fact that in games, graphics should be more than visual cues, but actually representing what's going on, if at all possible. I brought up Halo 2's battle rifle because the animations for reloading are longer than it actually takes to reload, so people just cancel the animation after a certain frame of it by quickly double-switching weapons or hitting the melee button. This effectively cancels the animation for the reload because the double-switch or melee is much faster and you're allowed to switch weapons in the midst of reloading. And while it cancels the longer animation, you still have a fully reloaded weapon. This is why you can't just look at a graphic and tell if it's right. Little glitches like this completely change the game (and in my opinion, for the worse)
That's a pretty nice collection of books there, so you gotta give the guy credit for that. That said though, as the saying goes...Those who can, do. Those who can't, teach.
@JJVHWhile KevinJ is mostly wrong, not all games have to care about pushing the latest graphics. Casual games, even 3D causal games, definately don't. And there are plenty of advantages to using a "slower" higher-level language. There are some pretty decent Java/C# game engines out there that can push pretty decent graphics through opengl/directx pipelines. You probably wouldn't use them for next-gen games, nor for heavy simulation, but there's no reason to say that all games MUST be C++. These days, all the real graphics work occurs on the card, so as long as you're able to push data fast enough (quite possible, even with bytecode languages), that's not a huge issue. Again, simulation, such as physics, is where the problem lies, but if all you're doing is basic collision detection on a few bodies...Did a lot of research and testing about a year ago on the feasability of non-C++ game engines for a project, and we eventually ended up deciding on Java. Yes, it's a 3D game. Yes, the performance is fine. As has already been stated, a real game programmer uses whatever is best for the project :P That's *usually* C++...but that's overkill for the casual market, that's for sure :P
Oh my, what a little storm in a teacup I've stumbled across here.Guys. It's mostly humor. It's designed to make students think, not to be a world-changing inscribed on stone set of commandments from on high. Apparently it offends some people's delicate sensibilities though. Some of you are definitely sensitive about the issues I've raised. Which means my purpose in writing it has mostly been fulfilled.How often have I heard "Those that can do, those that can't, teach"? Where originality is concerned, you've hit an all-time low there. I'm sorry, but I'm underwhelmed. How about "Those that can, do, those that can't, teach, and those that can't even do that leave pissy little comments on a wiki."The experts in game programming here who think teaching isn't a profession are more than welcome to come guest lecture in my class one evening. Just shoot me an email and I'll set you up. Who's up for the challenge?
hockeyAug 4, 2006
Discipline dude, simple discipline. You've got to literally will yourself to do it and see your projects through.
nuvemAug 4, 2006
"Come on. What if there are hidden bugs. Moreover newer games are more focussed on accurate physics."It's graphics programming... there are no hidden bugs... or physics.
nerdyninjaAug 4, 2006
Er superbouncing is a geometry exploit. Not a feature. And my point about the graphics programming one, where if something looks right, it is right being false... I was trying to articulate (and failing miserably) the fact that in games, graphics should be more than visual cues, but actually representing what's going on, if at all possible. I brought up Halo 2's battle rifle because the animations for reloading are longer than it actually takes to reload, so people just cancel the animation after a certain frame of it by quickly double-switching weapons or hitting the melee button. This effectively cancels the animation for the reload because the double-switch or melee is much faster and you're allowed to switch weapons in the midst of reloading. And while it cancels the longer animation, you still have a fully reloaded weapon. This is why you can't just look at a graphic and tell if it's right. Little glitches like this completely change the game (and in my opinion, for the worse)
m0n3yAug 4, 2006
Actually, his list was last updated on December 19, 2002. So no exaggeration is needed... Just another case of those who can't do, teach...
redbaritoneAug 4, 2006
And according to him, those who can't teach sell Amway.
violentvinylAug 4, 2006
That's a pretty nice collection of books there, so you gotta give the guy credit for that. That said though, as the saying goes...Those who can, do. Those who can't, teach.
squeeselfAug 4, 2006
@JJVHWhile KevinJ is mostly wrong, not all games have to care about pushing the latest graphics. Casual games, even 3D causal games, definately don't. And there are plenty of advantages to using a "slower" higher-level language. There are some pretty decent Java/C# game engines out there that can push pretty decent graphics through opengl/directx pipelines. You probably wouldn't use them for next-gen games, nor for heavy simulation, but there's no reason to say that all games MUST be C++. These days, all the real graphics work occurs on the card, so as long as you're able to push data fast enough (quite possible, even with bytecode languages), that's not a huge issue. Again, simulation, such as physics, is where the problem lies, but if all you're doing is basic collision detection on a few bodies...Did a lot of research and testing about a year ago on the feasability of non-C++ game engines for a project, and we eventually ended up deciding on Java. Yes, it's a 3D game. Yes, the performance is fine. As has already been stated, a real game programmer uses whatever is best for the project :P That's *usually* C++...but that's overkill for the casual market, that's for sure :P
ianparberryMay 24, 2007
Oh my, what a little storm in a teacup I've stumbled across here.Guys. It's mostly humor. It's designed to make students think, not to be a world-changing inscribed on stone set of commandments from on high. Apparently it offends some people's delicate sensibilities though. Some of you are definitely sensitive about the issues I've raised. Which means my purpose in writing it has mostly been fulfilled.How often have I heard "Those that can do, those that can't, teach"? Where originality is concerned, you've hit an all-time low there. I'm sorry, but I'm underwhelmed. How about "Those that can, do, those that can't, teach, and those that can't even do that leave pissy little comments on a wiki."The experts in game programming here who think teaching isn't a profession are more than welcome to come guest lecture in my class one evening. Just shoot me an email and I'll set you up. Who's up for the challenge?