Very Computer

Help! Please tell me how I'm wrong!

I am running Microsoft's new OpenGL 1.1 under Windows NT 4.0 and
cannot for the life of me gain access to glBlendEquation and without
this function, my project is at a stand still.

I got this implementation from:

         ftp://ftp.microsoft.com/Softlib/Opengl95.exe

It contained newer larger versions of the .dlls and .libs than
NT 4.0 (???) and a new set of includes.

Checking the includes, they did not contain 1.1isms that
the 1.1 OpenGL include contains on our SGIs.

The readme included with the above did say:

Quote:>The Windows NT 4.0 and Windows 95 releases of OpenGL includes new functionality
>and performance enhancements.  These include:
>1) A complete implementation of OpenGL 1.1.  OpenGL 1.1 contains several
>functions, including vertex array, polygon offset, logic ops, and several new
>functions for handling textures.  The vertex array and texture calls are
>particularly significant, as they may enable order of magnitude performance
>improvements in some applications.

Please Help!

Perry Kivolowitz

| Help! Please tell me how I'm wrong!
|
| I am running Microsoft's new OpenGL 1.1 under Windows NT 4.0 and
| cannot for the life of me gain access to glBlendEquation and without
| this function, my project is at a stand still.
|
| I got this implementation from:
|
|        ftp://ftp.microsoft.com/Softlib/Opengl95.exe

(actually ftp://ftp.microsoft.com/Softlib/MSLFILES/Opengl95.exe)

| It contained newer larger versions of the .dlls and .libs than
| NT 4.0 (???) and a new set of includes.
|
| Checking the includes, they did not contain 1.1isms that
| the 1.1 OpenGL include contains on our SGIs.

There is currently no OpenGL 1.1 release for SGI.  Our current
offerings are 1.0 plus extensions which provide a superset of 1.1
functionality.

On SGI there is glBlendEquationEXT, but this extension is not part of
the OpenGL 1.1 specification, and is not supported in the Microsoft
release.

| But, as I have said, I cannot find any evidence of glBlendFunc's
| whereabouts.

I assume you meant BlendEquation, since BlendFunc is part of OpenGL
1.0 and is supported by the MS release.

| Either I'm wrong - which I'm hoping (and that someone will explain to me
| how to find this function) or Microsoft's claim of completeness is
| significantly overstated.

Not all of the extensions provided by SGI and other vendors became
part of the OpenGL 1.1 spec.  As far as I know the Microsoft release
is 1.1 compliant (I haven't verified this personally).

If you are interested in the GL_LOGIC_OP mode for RGB, you can instead
use glEnable(GL_COLOR_LOGIC_OP).  If you want a different mode, you'll
have to find another way to achieve the same functionality.

--
Michael I. Gold     http://www.veryComputer.com/;  415/933-1709

And my mama cried,  "Nanook a no no! Don't be a * eskimo! Save your
money, don't go to the show!"  Well I turned around and I said, "Ho! Ho!"

      Except for the historical information
   contained herein, the matters discussed
   in this communication are forward looking
   statements that involve risks and
   uncertainties that could cause actual
   results to differ materially from those
   projected. Have a nice day.

__________________________________________________

Quote:> On SGI there is glBlendEquationEXT, but this extension is not part of
> the OpenGL 1.1 specification, and is not supported in the Microsoft
> release.

For what its worth, it was indeed my error in looking for glBlendEquation
in NT OpenGL in the first place. I have since learned, as Michael points
out, that it isn't part the 1.1 spec. Groan.

--
Perry Kivolowitz                        The opinions expressed herein are
Founder, Elastic Reality, Inc.          my own and do not represent those
Avid Technology, Inc.                   of Avid Technology, Inc.

| ...
| And therein lies my problem. For my application, I critically needed
| to set glBlendEquation to GL_MAX. Without it, my application requires
| dividing the intensity of a brunch by a factor related to its size
| to account for repeated addition over a small region. The size
| sensitive division limits the size of the brushes I can use too
| severely.
|             .... Unless someone can suggest a workaround, I'd say this
| makes it pretty darned hard to use OpenGL under Windows for many types
| of operations, like paint with soft edged brushes. If there is a way
| around this, I'd love to hear about it.

I'm not sure I understand all your issues, so I apologize if this
reply isn't relevant.

Does the GL_MAX blend equation really do what you want?  For example,
if you passed a white brush with dark gray edges over a light gray
background, you'd get aliased brush strokes.  The white part of the
brush would win at the brush center, the background would win at the
brush edges, but you wouldn't get a soft-edged effect, and the stroke
width would appear to vary with the background gray level.

Most OpenGL-based painting programs that I've seen use ordinary
arithmetic blending, and treat the effects of repeated addition as a
feature.  The more you move a brush over an area, the more the brush
color covers the underlying image.  But it sounds like that's not the
effect you want.

One thing you might try is using the stencil buffer.  For example, you
could clear the stencil buffer to zero, enable framebuffer updates
only where the stencil buffer is zero, and set the stencil buffer to
one for each pixel that's updated.  That would allow you to pass a
blended brush over an area, and guarantee that the brush affects each
pixel no more than once.  (There would be artifacts where brush
strokes intersect, though; the first stroke would appear to overlap
later strokes.  Using more stencil bits might let you work around
this if it's important.)

Allen

Quote:>...my application requires
>dividing the intensity of a brunch by a factor related to ...

        :-),
        -P.

--
********************* (Note new snail-mail address.) **********************
* Peter S. Shenkin, Chemistry, Columbia U., 3000 Broadway, Mail Code 3153,*

MacroModel WWW page: http://www.cc.columbia.edu/cu/chemistry/mmod/mmod.html

I'm sorry that the 1.1 standard doesn't have the extensions you need for your
product.  What I'd like to suggest to you **and any other ISV's** looking at
the Microsoft platform is that you get us in the loop as early as possible with
your technical needs.

We have worked with key ISV's to supply Microsoft-specific OpenGL extensions,
and we'd appreciate hearing more about additional such functional requirements.
Now is a particularly bad time to respond to such requests, since we've just
released new implementations for both NT and Windows95, but we'd like to hear
from you.

In addition to helping us implement proprietary extensions, this feedback helps
us understand which proposed extensions are worth promoting into core
functionality in the next specification of OpenGL, a move which helps make the
features you need pervasive and implementation-independent.  This is
particularly the case with extensions which other vendors have already
implemented.

Thanks!

Steve Wright
OpenGL Program Manager
Microsoft Corporation