Just start now. You never know until you have tried. Just do GDNet forums a break, don't try and write Quake.

--------------------------
Carpe Diem

Edited by - DeltaVee on July 5, 2000 3:18:05 PM

I remember my first attempt at 3D. I made some blocks out of points and I could walk around them. Then I got stuck when I tried to figure out a way to fill in the faces. At that point I went out and bought "Cutting Edge 3D Game Programming with C++" by John De Goes. After going through that book I could actually write my own polygon rasterizer w/ clipping, texturing, and Gouraud shading.

A few months ago, when I started looking into D3D, I was like, "this is so easy!", because it just takes care of all those details for you. Honestly, I don''t know if it''s even worth learning the details of polygon rasterization. I mean, considering the profusion of 3D hardware, I wonder if software programmers will ever need to be familiar with that part of the pipeline again.

Yeah, I would definitely not try writing a quake caliber project. I would be pleased with something very simple...walking around a block for example. Particle Engines Totally impress me more than anything ( I could stare at WitchLords Particle Engine Demo for hours ). I would love to be able to do something like that ( maybe in a few months I will be able to?) What I think I plan on doing is to continue working on 2d stuff, becuase I don''t want to totally ignore that...and also on the side start reading on 3d Maths (I found a few good sites) and 3d programming, and then see if I can make a smooth transition..how does that sound..? or do you think it will be overkill? well, thanks for the replies again .. you are all awesome.. keep them coming

a usefull link!
http://www.geocities.com/SiliconValley/Park/9784/tut.html

Floyd

That''s 100% true, you will never need to draw your own triangles ever again, but getting a perpective corrected texture mapper finished and up to a reasonable speed is fantastic for learning about optimizing algorithms, and interpolation, which you may well need elsewhere in a typical game.
And of course, you will find every other 3d programmer *can* do software engines, so they''ll always have that edge.

Floyd` - wow that site looks really really useful! thanks. Simon_brown, I think I understand what you are saying . I agree that I think it is a really good idea, and even needed, to understand how 3d programming works, and even be able to apply it. But once you know it, it is better probably to use an API..am I correct?

hrmm..is it just me...or are there quite a few 404''s on that page . Especially for the basic stuff. Hrmmm

>>But once you know it, it is better probably to use an API..am I correct?

Absolutely. Actually D3D (and most likely OpenGL too, i''ve haven''t used it) can do a lot of your 3d engine for you if you want it to. And if you want to use Hardware T&L then you *have* to let D3D transform and light your triangles. So basically, the overlap between a software and hardware engine is getting smaller.
Having said that I would still make sure you are comfortable at least with the *general concepts* of 3d programming, if not the detail and implementation, before tackling your chosen API.

simon_brown - that is exactly what I am going to do...I will try implementing some things in general 3d and get nice and comfortable with it and then I will choose an API and see how it feels. If I still feel a little lost then I will go back and learn more straight 3d. thanks. Any more comments or suggestions for me or for anyone else?