Mail Archives: djgpp/1998/08/03/13:24:12
*********** REPLY SEPARATOR ***********
On 98-08-02, at 16:04, Arthur wrote:
>> >> I think that most 3D API's have everything needed to write a good 3d game.
>> >> However I may be wrong... :)
>> >
>> >YOU WHAT?! Call yourself a programmer?
>> >
>> >Other things that are needed include talant, and a brain. These are
>> things that about
>> >half the modern day 3D programmers are without.
>>
>> What I wrote concerned only graphics. It's obvious that a game must be based on
>> a really good idea, programmer must be good at maths, physics, etc.
>> I realize it's not very ambitious.
>>
>> >Trouble about APIs is that the more powerful they are, the less you have
>> to program.
>> >If you can do 3D in software, not using any APIs *then* you're a programmer. And a
>> >game does NOT have to be 3D to be good.
>>
>> Writing a win32 game (that uses 3D hardware) without any standard 3D API
>> is like writing
>> DJGPP DOS program without libc.
>> Have you ever seen a game (that uses 3D hardware) which doesn't use 3D
>> API? (like DirectX, Glide,
>> OpenGL) I haven't. And I'm sure there will ever be no such game. Besides,
>> it's almost impossible
>> to write a faster 3D API than OpenGL, DirectX, Glide.
>
>Read the paragraph that I wrote above. It is indeed almost impossible to write a
>hardware-based 3D system without an API; but I said "3D in software, not using any
>APIs"
Ok, but don't you think it's a waste of time? Would you spend a month working on
something that works from -100% up to +5% faster than DirectX? (unless you're
going to use a technique which is not supported in any API, and should work
much faster than API's equivalent.)
>You can beat software APIs using your own routines, if you're good (and several demo
>crews are). This is why I said that you're a good programmer if you can do that.
>Using an API requires much less talant.
I agree of course, but to do that, you have to be very, very good.
Pawel Kowalski
- Raw text -