Re[2]: OpenGL+GLUT the future

[Ian Crawford <icrawfor@uoguelph.ca>]

> Glut is built on top of Mesa/OpenGL. If Mesa/OpenGL has glide support
> (yes it does) then GLUT has glide support.

I'm not so sure about that.  I mean, glut manages the hardware side of
things, right?  Using X as an example, GLUT opens a window and
processes all the events.  Now, you are right in that I don't know
much about GLUT and I'm sure I under estimate it, but wouldn't the
ability to interface with the 3dfx card (through glide) be a job for
GLUT?  If what you're saying is correct, I should be able to take a
program out of the red book and compile it _as is_ and when I say "go"
it will switch to full screen mode and draw pretty things using my
3dfx card.

Now, having Mesa/OpenGL run through glide is a requirement, but I
suspect there is more on the GLUT side of things.

I might be wrong here.  I don't claim to be an expert in this field.
However, assuming I'm not wrong, my original question remains: is
there a glide version of GLUT?

> GLUT is GPLed. And it's author is expanding its library to support
> game programming.

Cool.  Out of curiosity, is there an official GLUT homepage?  Some
where on SGI's site, maybe?

> Not really, a Messa/GLUT/GUI can be faster than X11, simply because it
> can take advantage of the hardware in a way X11 cannot.

Could you expand on this?  This probably isn't for the list, so you
can reply directly to me.

> There is a large scope for improvement in the software renderer, and
> geometry engine.

Again, I agree.  I'm sure Mesa will overtake SGI's OpenGL in the
future.  However, since PCs in general are slow/cheap (I am comparing
this to Alpha, SGI, and other RISC architectures), if you really want
great performance, you need hardware acceleration.  Can GLUT allow me
(a lowly PC user) to use that?

> Do you think the current high end cards are optimised for OpenGL,
> direct3d or penguinplayGraphics?

There are high end cards of all kinds out there.  Lately, I've looked
into 2D/3D cards with OpenGL acceleration since that's what I've been
developing for at work.  There are some great cards out there, but
they can be pricey (often >$1000).  As for 3dfx cards (high end in the
gaming world), the only interface to them is through glide, so
anything built on top of glide will run well on those cards.

Now, let me try and bring this back to the intention of the thread.
I'm going to try to be objective here.

GLUT has a lot going for it.  It is simple, portable, and relatively
fast.  It already runs on a variety of platforms and has built in
support for keyboard, mouse, and joystick events (on all platforms?).
 These are all things that a game SDK needs.

What can't you do with GLUT? (I might stick my foot in my mouth here.
Please be gentle if I do.)

GLUT lacks is the ability to give up control of the window/screen (in
whole or in part) to be used by a different display mechanism (cut
scenes and an a general purpose OpenGL widget come to mind).

GLUT has a predefined "main loop" (I think...) than cannot be
overridden (to allow you to handle events differently, etc.).  This
may just add another rule to the design of your program, but I'm sure
this is, somehow, limiting.

Now, you mentioned that GLUT is distributed under the GPL.  Perhaps,
we can take advantage of this and use the GLUT code.  We could fill in
the blanks and end up with something _like_ GLUT (i.e., a portable
framework for 3D graphics), only better suited for our purpose.

Well, food for thought.  It's late and I'm going to bed.

Ian