Suggestion for future tutorials

Good point, Protozone. I find the best way to learn is to have a piece of code that does just one thing. Then you can tweak it and watch the results immediately. It would be a bit like, say, a chemistry experiment where you must change one variable at a time to understand its effect. If the tuts are too complex your energy is dissipated by too many factors to master at once.

Just my two cents worth... And don't get me wrong, NeHe, your tuts are by far the best on the web. Keep up the good work!

P.S. Suggestion for a next tut: collision detection? I know everyone says its "easy" and stuff can be found about it all over the web, but I haven't yet found a straightforward and simple way of not passing through walls in Tutorial 23. Naveed's was good but the user would just get stuck when he hit a wall, we need an algorithm that will make him slide along the wall when he hits it. After that, but related to collision detection, would come gravity I think: how do you keep the user's feet on the ground?

I found some good non-commercial open-source stuff here concerning collision detection, for those savvy enough to dissect the code, I'd love to have some feedback:

Read about it here:
http://www.cs.unc.edu/~geom/OBB/OBBT.html

Download it here:
ftp://ftp.cs.unc.edu/pub/users/manocha/CODE/Proximity/RAPID201_MSVC.tar.gz

Edited by - Keermalec on November 9, 2000 7:26:13 AM

Well, i don''t fully agree.

Part of learning to program is to learn how to read code and
extract what type of info is important and what it not. I mean
to really learn how to be a good OpenGL programmer you must learn
how to read the code that others have written. Tutorials just
teach you the basic tricks of how to achive a certain effect, but
what they dont teach you is how to put it all together into a
more complex program where you include many of those tricks.
And then we have all theese things that a tutorial simply could
not cover, the only way to learn those is to by a book on the
matter or examine someone elses code and then that ability to
read others code will become invaluable.

NeHe''s tutorials are still very simple even though he has
unrelevant code in them. I think he has found a fine balance
between simple and advnaced. The first 15 tuts are the learning
platform to get you aquainted with OpenGL, then he increase
difficulty gradualy, mostly because there is no EASY way to
explain the thing the tuts cover.

Now you say, yeah yeah, thats easy for a pro to say. Well i am
definitly NOT a pre. I have coded in OpenGL for about 4 weeks now
and is pretty new to C++ as well. I am no Genius either, just a
simple average guy who loves to code. I read some books in the
beginning that was aimed towards beginners, but believe me when i
say that they were much harder to read than NeHe''s tutorials.
What i learned from the books in 3 weeks taught me less than
what NeHe''s tutorials taught me in a week. Now after reading
the tuts i went back to the books, and it was alot easier to
understand them now than before.

Hmmm, when i start to type i have a hard time to stop, hehe.
I will end this reply now, or could continue for hours

Sorry for my bad spelling and gramma, not a native english
speaking guy.

Keep up the good work NeHe, love your site.

Allmight.

PS. I will drop you a demo one fo theese days nehe wich is my
absolute first ever. DS.

Ok well I don''t have a tutorial but if you want to implement simple collision detection to keep someone in a maze check out the code to my 2dracer.zip on my site. It uses a point in polygon algorithm to determine if the user is hitting a wall. basically I draw an n-sided polygon and if the users input has moved them outside the polygon then I don''t allow that movement. Find it in the projects section if memory serves.

http://www.CornflakeZone.com
//-- Modelling and animation in every dimension --//

The point of nehe''s tutz is not teach stuff in an easy way

but more to teach cool looking stuff and stuff everyone want to do in the most simple way ... Of course, for a beginner, casting shadow is not hard to understand, that''s why there are the first tutz

This is an interesting and difficult topic that I have thought long and hard about. In the beginning I wanted to explain things in such a simple way that just reading through the tutorial once would explain just about everything you needed to know to accomplish the one effect talked about.

When lesson 10 was released it was a hard choice on my part. Post the code, which was not a gradual increase from lesson 9, or not post the code, and deny people the chance to see what the author had written.

As you may have noticed, lesson 10 was not an easy tutorial to learn from. After posting, I was flooded with mail asking "why are there so few comments?". So I tried to add more comments. Then I was told "too hard".

But on the other hand, I had a group of people that breezed through the first few lessons asking "why didn''t you add this feature or that feature to lesson 10", basically asking why the tutorial wasn''t a complete 3D engine with collision and the works.

I have always attempted to gradually introduce new effects, commands, etc. But there are times where the code I''m posting covers more than one topic. So I try to explain it all. Looking back at the first 25 tutorials I''ve found alot of mistakes on my part, alot of sections that I thought sounded clear at the time, but now ... I would like to rewrite the entire site, but it''s just too much work, and by doing that I''d end up spending the next year fixing things.

By now I feel that if you''ve been reading the tutorials, and figuring out what it is I''m trying to say, you should have a pretty good understanding of the basics. Spotting my errors should be easy, and tutorials should be more advanced.

If I posted news today saying "one effect per tutorial" I''d be flooded with people saying "man, I want MORE". If I posted news stating that I was going to cram more into each tutorial I''d get mail saying "slow down". So for now, I''ll try to keep the amount you have to learn each tutorial at a reasonable level and hope that the majority of you are happy.

I think after I''ve posted the next few tutorials (shadows, volumetric fog, map files, and a few others), I will rewrite the main code in a lesson 1 style tutorial. Nothing but the base code. Split into files, cleaned up, packed with features (bmp and .tga loading, ati fixes, etc). Then from there we''ll move on. Eventually I''d like to turn the tutorials into a project that we can all work on together. A 3D engine built by the visitors. Teaching us teamwork, combining our tutorials into a real project, etc. Long ways away I''m sure But stuff I do think about.

Hi,
It''s a great idea to create a 3d "NeHe''s Visitors" engine
i just want to say that tutorials are for almost quite simply to understand, but that''s that some stuffs are heavier to understand ,(but i''m beginner , since 4 months i beginn! sorry for my english!)
it could be useful for beginner like me, that someone create OpenGL exercizes like in school !
ciao and keep it real guy!
al