gl4java-usergroup

Re: [gl4java-usergroup] Linux & Display Lists

[Jean-Yves B. <jea...@wa...>]

Hello,

Be carefull:
parameter of glGenLists(param) is the number of display list you want to create.
The return value is an id used to call and delete the display list.
In you code, it seems to be the same parameter.

I think this code would be better:

                nName = gl.glGenLists (1);          // generate one new list
                gl.glNewList (nName, gl.GL_COMPILE);    // start of display list definition

                ...<opengl calls>...                            // I'm currently just using lines & quads

                gl.glEnd();                                             // end of display list definition

                ...

                gl.glCallList (nName);                          // draw display list


I can't say about linux, but on macintosh and windows (as you said), it runs very well.

regards.

Jean-Yves.

Thorsten Roemer a *crit :

> Hi *,
>
> did someone tried to use Display Lists under Linux ?
> On Windows it works, on Linux there's just a black window ?!?!?!
>
> - Red Hat 6.2
> - Kernel 2.2.14
> - XFree 4.01
> - NVidea GeForce2 Driver 0.94
>
> It's something like that:
>
>                 int nName = 1;
>                 gl.glDeleteLists (nName, 1);                    // delete old list
>                 nName = gl.glGenLists (nName);          // generate one new list
>                 gl.glNewList (nName, gl.GL_COMPILE);    // start of display list definition
>
>                 ...<opengl calls>...                            // I'm currently just using lines & quads
>
>                 gl.glEnd();                                             // end of display list definition
>
>                 ...
>
>                 gl.glCallList (nName);                          // draw display list
>
> Any ideas's, comments ???
>
> regards, Thorsten
>
> --
> Thorsten Roemer
> IVISTAR Kommunikationssysteme AG
> Ehrenbergstr. 19 / 10245 Berlin
> http://www.ivistar.de
>
> _______________________________________________
> gl4java-usergroup mailing list
> gl4...@li...
> http://lists.sourceforge.net/mailman/listinfo/gl4java-usergroup

--
--------------------------------------
Jean-Yves BRUD
POLYQUARK - Ingenierie & Creation 3D
Palahou - 31330 LARRA    France
Tel: 05.62.79.03.33  Fax: 05.62.79.03.38
Mail: jea...@wa...
--------------------------------------

[gl4java-usergroup] Linux & Display Lists

[Thorsten R. <tho...@iv...>]

Hi *,

did someone tried to use Display Lists under Linux ?
On Windows it works, on Linux there's just a black window ?!?!?!

- Red Hat 6.2
- Kernel 2.2.14
- XFree 4.01
- NVidea GeForce2 Driver 0.94

It's something like that:

		int nName = 1;
		gl.glDeleteLists (nName, 1);			// delete old list
		nName = gl.glGenLists (nName);		// generate one new list
		gl.glNewList (nName, gl.GL_COMPILE);	// start of display list definition

		...<opengl calls>...				// I'm currently just using lines & quads
				
		gl.glEnd();						// end of display list definition

		...

		gl.glCallList (nName);				// draw display list

Any ideas's, comments ???

regards, Thorsten

--
Thorsten Roemer
IVISTAR Kommunikationssysteme AG
Ehrenbergstr. 19 / 10245 Berlin
http://www.ivistar.de

Re: [gl4java-usergroup] Mac Port: Fetching Function Pointer Dynamically (FindSymbol, ...)

[Jean-Yves B. <jea...@wa...>]

Hello Swen,

I am not sure I can help you in your Mac port but I try to compile the
previous release of GL4 (2.3.1.0) made by gerard.

To do this, I used MetroWerks Code warrior, and I rebuilt the extension.

I made this to find a bug I noticed in July about reading
the FRONT buffer that appears only on mac platform.

In the package made by gerard, there where several Metrowerks projects:
- one to build the extension,
- one to build the zip file,
- one to build some demos.

If I can help more.

Sven Goethel a *crit :

> It works !
>
> :-)
>
> Using the OpenGLLibrary !
>
> I have to clean up things, then I really know
> in detail ... how it works :-))
>
> E.g:
>         - using c2pstrcpy (carbon stuff)
>         for converting the c string to an pascal string !!
>
>         - seeking all symbols out of the lib ...
>
> TODO:
>         Now I have to check the Privilege mechanism
>         to get prviliges for the security manager to
>         install and ...
>
>         Adding the new visual capabilities like
>         accumulator bits, etc (new since version 2.3.1 ...)
>
> Later ...
>
> Sven
> --
> mailto:sgo...@ja...
> www   : http://www.jausoft.com ; pgp: http://www.jausoft.com/gpg/
> voice : +49-521-2399440, +49-170-2115963; fax: +49-521-2399442
> _______________________________________________
> gl4java-usergroup mailing list
> gl4...@li...
> http://lists.sourceforge.net/mailman/listinfo/gl4java-usergroup

--
--------------------------------------
Jean-Yves BRUD
POLYQUARK - Ingenierie & Creation 3D
Palahou - 31330 LARRA    France
Tel: 05.62.79.03.33  Fax: 05.62.79.03.38
Mail: jea...@wa...
--------------------------------------

Re: [gl4java-usergroup] Mac Port: Fetching Function Pointer Dynamically (FindSymbol, ...)

[Sven G. <sgo...@ja...>]

It works !

:-)

Using the OpenGLLibrary !

I have to clean up things, then I really know
in detail ... how it works :-))

E.g:
	- using c2pstrcpy (carbon stuff)
	for converting the c string to an pascal string !!

	- seeking all symbols out of the lib ...

TODO:
	Now I have to check the Privilege mechanism
	to get prviliges for the security manager to
	install and ...

	Adding the new visual capabilities like
	accumulator bits, etc (new since version 2.3.1 ...)

Later ...

Sven
-- 
mailto:sgo...@ja...
www   : http://www.jausoft.com ; pgp: http://www.jausoft.com/gpg/
voice : +49-521-2399440, +49-170-2115963; fax: +49-521-2399442

Re: [gl4java-usergroup] Mac Port: Fetching Function Pointer Dynamically (FindSymbol, ...)

[Sven G. <sgo...@ja...>]

... still no success:

GetSharedLibrary returns 0 -> OK !
FindSymbol returns -2802 -> cfragNoSymbolErr, symbol not found
	tried with GetSharedLibrary:
		OpenGLLibrary
		OpenGLEngine
		OpenGLRenderer
		OpenGLRendererATI

here is the source:

#ifdef macintosh
        #include <agl.h>
        #include <string.h>
        #include <CodeFragments.h>
        #include <Errors.h>

        #define fragNoErr 0
#endif  

...

#ifdef macintosh
        Str255
errName;                                                                
Ptr glLibMainAddr = 0;
        CFragConnectionID glLibConnectId = 0;
        CFragSymbolClass glLibSymClass = 0;
        OSErr returnError=fragNoErr;
        static char buffer[256];

    returnError =
                GetSharedLibrary("\pOpenGLRenderer",
                               kPowerPCCFragArch,
                               kReferenceCFrag,
                               &glLibConnectId,
                               &glLibMainAddr,
                   errName);

        printf("GetSharedLibrary errMsg: %s\n", errName);
        fflush(NULL);

        if (returnError != fragNoErr)
        {
                printf ("GetSharedLibrary Err(%d): Ahhh!  Didn't find
LIBRARY !\n",
                        returnError);
                fflush(NULL);
        }

        if (returnError == fragNoErr)
        {
                sprintf(buffer, "\p%s", func);

                returnError =
                  FindSymbol (glLibConnectId, buffer, &funcPtr, &
glLibSymClass );

                if (returnError != fragNoErr)
                {
                  printf ("GetSharedLibrary Err(%d): Ahhh!  Didn't find
SYMBOL: %s !\n",
                        returnError, func);
                  fflush(NULL);
                }
                returnError = fragNoErr; // fall back to ok mode ...
        }
        if (returnError == fragNoErr && glLibConnectId!=NULL)
        {
                returnError = CloseConnection(&glLibConnectId);

                if (returnError != fragNoErr)
                {
                  printf ("GetSharedLibrary Err(%d): Ahhh!  Didn't close
LIBRARY !\n",
                        returnError);
                  fflush(NULL);
                }
        }

        lmethod=2;

#endif

		
-- 
mailto:sgo...@ja...
www   : http://www.jausoft.com ; pgp: http://www.jausoft.com/gpg/
voice : +49-521-2399440, +49-170-2115963; fax: +49-521-2399442

Re: [gl4java-usergroup] Errors

[Virgil W. <vir...@ra...>]

>...
> Sounds fine, if all platforms can handle it - I guess so !
> A nice extension to the GLContext class ?!
>
> So please - just send your code if it is done, piece for piece.
> Or will you hold things proprietary ? :o)

Once I get the interface worked out the way I like it (and make sure it
works on unix), I'd be happy to submit stuff for inclusion.

Some stuff might not work on all platforms, though.  For instance, I'm not
sure the screen resolution changing stuff will work everywhere (under X on
unix for instance).  Somebody else with more experience might know if you
can do this.  As such, I doubt that the gl4java project would be interested
in such things.  I suppose I could always put the platform specific
utilities together into their own package.  That way, folks could use them
if they wanted, but not expect platform specific functionality to be
available on all platforms.

>
>
>...
> Puhh ...
> Well, this looks very native .. may be not interesting.
> But all the parts, like timing funcs etc. :-), are portable ..

As soon as you can plug a force feedback joystick into HPUX (and have it
work), I"ll start to worry about the native code to control the joystick.
This stuff is all specific to my project, though, so it has no impact
(actual or intended) on gl4java.  Since the topic kinda came up, though,
I'll point out that I'm not using java because I want my software to run
"everywhere".  I'm using java because I enjoy using the language.  As you
might expect, my target audience for a game is 98/ME/2000/NT/linux  (in that
order).  Ignore linux, and I should be using c++/direct3d.  But, I like
opengl and java better.  That's why I'm playing with gl4java.  Soon, my
senses will probably come around and I'll write the opengl rendering code in
C, and only use java for the game logic.   But for now, it's nice to
prototype everything in java.

>
> Sorry for being that non diplomatics and shrill,
> just want to force you to show me something :o)
>

No problem.  gl4java is a great project/product.  I have nothing but the
highest regards for everyone involved with the project.

Virgil

Re: [gl4java-usergroup] Errors

[Sven G. <sgo...@ja...>]

Virgil Wall wrote:
> 
> > Really ?
> > I thought that even 50ms precision should be enough
> > -> about 20 fps (more or less :-) !
> 
> The visual quality of using the more precise clock is pretty apparent in
> some situations.  The human eye is pretty good at catching "glitches" in
> anticipated  motion.  Pretty much any commercial game uses more accurate
> timing.
> 
> >
> > Well - anyway, if you really have to be that precisous,
> > i wonder, how you handle the _very_ unprecise Thread.sleep() call ?
> > Because you are guaranteed to be woke up in the right time ...
> 
> Because I time the sleep too.  What matters is how much time elapsed since
> the last update, Thread.sleep/Thead.yield and all.  The only reason to
> sleep/yield at all is to prevent thread starvation.  So naturally, you have
> to take a time stamp, sleep a little, take another timestamp and repeat.
>

well, it must be interesting ...
may be some kind of stereo capabilities will need this quality also.
 
> >
> > > (For the curious, the Magician Java opengl crap has a percision timer
> > > call... it's pretty each to do, but one might argue that it has nothing
> to
> > > do with opengl)
> > >
> >
> > Ok ... so why don't you offer your native function
> > to GL4Java ?
> 
> Because I'd have to make it work on windows, unix and mac:) (windows and
> unix are no problem,btw)
> Also, if I'm going to go to all of this trouble, then I'd probably offer
> more than just a few timing calls.  For instance, I use native calls to
> change the screen resolution and make the fullscreen window a top level
> window  (so it would be possible to do full screen things at some resolution
> other than what the user's desktop is set to).
> 

Sounds fine, if all platforms can handle it - I guess so !
A nice extension to the GLContext class ?!

So please - just send your code if it is done, piece for piece.
Or will you hold things proprietary ? :o)

> >
> > If the implementation uses the Unix like time functions,
> > I guess we can add it for ALL OS/platforms !
> > I love to add it into the next release, if you are fast :-) enough ..
> > Currently I am working for:
> > - Linux*, Solaris-, SGI/Irix*, HP-UX-
> > - MacOS-
> > - Win32*
> > ports of GL4Java !
> > (* well done, - some platform specific problems, but under work)
> >
> > Well, please do not make OS depended stuff - if possible.
> > Your application should run on all of our GL4Java platforms, right ?
> 
> Nope :) to be honest, I could care less if anyone on HPUX can play this game
> or not.  If I'm going to care about HPUX then I'm also going to care about
> AIX... which isn't in your list.  

Ooops ...
AIX should work also :-)
AIX was my official 1st Unix port of GL4Java,
but now, i do not have access to this machine, to produce binaries ...

All Unix machines with X11R6 (and hopefully gcc), should work !
Just need some time and machine access ...

> By the time I'm done, I'm also probably
> going to be using DirectInput on windows (so I can poll the keyboard
> directly and use a force feedback joystick).  I might also use
> directsound....  So, a native timing function is the least of my concern.
>

Puhh ...
Well, this looks very native .. may be not interesting.
But all the parts, like timing funcs etc. :-), are portable ..
 
> > So just try to minimize special native things,
> 
> I hear you.... and I do just as any person might (its much easier to
> manage).
> 
> > and offer them (the native atoms) to the gl4java project - no problem !
> 
> Virgil

Sorry for being that non diplomatics and shrill, 
just want to force you to show me something :o)

Yours, Sven
-- 
mailto:sgo...@ja...
www   : http://www.jausoft.com ; pgp: http://www.jausoft.com/gpg/
voice : +49-521-2399440, +49-170-2115963; fax: +49-521-2399442

Re: [gl4java-usergroup] Errors

[Virgil W. <vir...@ra...>]

> Really ?
> I thought that even 50ms precision should be enough
> -> about 20 fps (more or less :-) !

The visual quality of using the more precise clock is pretty apparent in
some situations.  The human eye is pretty good at catching "glitches" in
anticipated  motion.  Pretty much any commercial game uses more accurate
timing.

>
> Well - anyway, if you really have to be that precisous,
> i wonder, how you handle the _very_ unprecise Thread.sleep() call ?
> Because you are guaranteed to be woke up in the right time ...

Because I time the sleep too.  What matters is how much time elapsed since
the last update, Thread.sleep/Thead.yield and all.  The only reason to
sleep/yield at all is to prevent thread starvation.  So naturally, you have
to take a time stamp, sleep a little, take another timestamp and repeat.

>
> > (For the curious, the Magician Java opengl crap has a percision timer
> > call... it's pretty each to do, but one might argue that it has nothing
to
> > do with opengl)
> >
>
> Ok ... so why don't you offer your native function
> to GL4Java ?

Because I'd have to make it work on windows, unix and mac:) (windows and
unix are no problem,btw)
Also, if I'm going to go to all of this trouble, then I'd probably offer
more than just a few timing calls.  For instance, I use native calls to
change the screen resolution and make the fullscreen window a top level
window  (so it would be possible to do full screen things at some resolution
other than what the user's desktop is set to).

>
> If the implementation uses the Unix like time functions,
> I guess we can add it for ALL OS/platforms !
> I love to add it into the next release, if you are fast :-) enough ..
> Currently I am working for:
> - Linux*, Solaris-, SGI/Irix*, HP-UX-
> - MacOS-
> - Win32*
> ports of GL4Java !
> (* well done, - some platform specific problems, but under work)
>
> Well, please do not make OS depended stuff - if possible.
> Your application should run on all of our GL4Java platforms, right ?

Nope :) to be honest, I could care less if anyone on HPUX can play this game
or not.  If I'm going to care about HPUX then I'm also going to care about
AIX... which isn't in your list.  By the time I'm done, I'm also probably
going to be using DirectInput on windows (so I can poll the keyboard
directly and use a force feedback joystick).  I might also use
directsound....  So, a native timing function is the least of my concern.


> So just try to minimize special native things,

I hear you.... and I do just as any person might (its much easier to
manage).

> and offer them (the native atoms) to the gl4java project - no problem !

Virgil

Re: [gl4java-usergroup] Errors

[Sven G. <sgo...@ja...>]

Virgil Wall wrote:
> 
> > Dear Virgil,
> >
> > I want to encourage you to use GLAnimCanvas,
> > or hte way GLAnimCanvas uses the animation update methods !
> >
> > I see it in many demos, that you use the threaded sDisplay call for
> > default !
> > This only works within a very good native multithreading JVM,
> > and does steals a lot of ressources ...
> >
> > You can see, that the GLAnimCanvas demos does have the option to
> > refresh directly (useRepaint... method), but it is disabled for default.
> >
> > I think, that the repaint method should be fast enough for us ... >= 20
> > fps :-)
> >
> > Of course, the direct way gives us quiet the same fps, like the native
> > application, but we do pay responsiveness of the JVM for it !
> >
> > Please check the PERFORMANCE mail from me to the mailinglist,
> > and the PerformanceLog's within the demos/MiscDemos directory !
> >
> > Yours, Sven
> 
> Well, the reason I have to call sDisplay in a loop is that I need to very
> accurately track how long it takes to do the display.
> GLAnuimCanvas tracks frames per second by using System.currentTimeMillis().
> This only yields timing accurate to around 16ms (on a good day).  This isn't
> really that good (for a game anyway).  Since I'm writing a game, I want very
> high frame rates and I need to know very precisely how long each update
> takes (so I can move game objects at the right rate).  To solve this
> problem, I make a call into a jni library I've written to more accurately
> calculate time.  (Using the system clock on windows to achieve near
> microsecond accuracy).
> Also, I do a Thread.sleep() once per loop (in order to set a maximum fps).
> 
> In summary, I pretty much do what GLAnimCanvas does, I just need to be able
> to very precisely track how long each update takes (and using
> System.currentTimeMillis() isn't good enough)
>

Really ?
I thought that even 50ms precision should be enough 
-> about 20 fps (more or less :-) !

Well - anyway, if you really have to be that precisous,
i wonder, how you handle the _very_ unprecise Thread.sleep() call ?
Because you are guaranteed to be woke up in the right time ...

> (For the curious, the Magician Java opengl crap has a percision timer
> call... it's pretty each to do, but one might argue that it has nothing to
> do with opengl)
> 

Ok ... so why don't you offer your native function
to GL4Java ?

If the implementation uses the Unix like time functions,
I guess we can add it for ALL OS/platforms !
I love to add it into the next release, if you are fast :-) enough ..
Currently I am working for:
	- Linux*, Solaris-, SGI/Irix*, HP-UX-
	- MacOS-
	- Win32*
ports of GL4Java !
(* well done, - some platform specific problems, but under work)

Well, please do not make OS depended stuff - if possible.
Your application should run on all of our GL4Java platforms, right ?

So just try to minimize special native things,
and offer them (the native atoms) to the gl4java project - no problem !


> Virgil

Sven
-- 
mailto:sgo...@ja...
www   : http://www.jausoft.com ; pgp: http://www.jausoft.com/gpg/
voice : +49-521-2399440, +49-170-2115963; fax: +49-521-2399442

Re: [gl4java-usergroup] Errors

[Virgil W. <vir...@ra...>]

> Dear Virgil,
>
> I want to encourage you to use GLAnimCanvas,
> or hte way GLAnimCanvas uses the animation update methods !
>
> I see it in many demos, that you use the threaded sDisplay call for
> default !
> This only works within a very good native multithreading JVM,
> and does steals a lot of ressources ...
>
> You can see, that the GLAnimCanvas demos does have the option to
> refresh directly (useRepaint... method), but it is disabled for default.
>
> I think, that the repaint method should be fast enough for us ... >= 20
> fps :-)
>
> Of course, the direct way gives us quiet the same fps, like the native
> application, but we do pay responsiveness of the JVM for it !
>
> Please check the PERFORMANCE mail from me to the mailinglist,
> and the PerformanceLog's within the demos/MiscDemos directory !
>
> Yours, Sven

Well, the reason I have to call sDisplay in a loop is that I need to very
accurately track how long it takes to do the display.
GLAnuimCanvas tracks frames per second by using System.currentTimeMillis().
This only yields timing accurate to around 16ms (on a good day).  This isn't
really that good (for a game anyway).  Since I'm writing a game, I want very
high frame rates and I need to know very precisely how long each update
takes (so I can move game objects at the right rate).  To solve this
problem, I make a call into a jni library I've written to more accurately
calculate time.  (Using the system clock on windows to achieve near
microsecond accuracy).
Also, I do a Thread.sleep() once per loop (in order to set a maximum fps).

In summary, I pretty much do what GLAnimCanvas does, I just need to be able
to very precisely track how long each update takes (and using
System.currentTimeMillis() isn't good enough)

(For the curious, the Magician Java opengl crap has a percision timer
call... it's pretty each to do, but one might argue that it has nothing to
do with opengl)


Virgil

Re: [gl4java-usergroup] Errors

[Virgil W. <vir...@ra...>]

> Dear Virgil,
>
> I want to encourage you to use GLAnimCanvas,
> or hte way GLAnimCanvas uses the animation update methods !
>
> I see it in many demos, that you use the threaded sDisplay call for
> default !
> This only works within a very good native multithreading JVM,
> and does steals a lot of ressources ...
>
> You can see, that the GLAnimCanvas demos does have the option to
> refresh directly (useRepaint... method), but it is disabled for default.
>
> I think, that the repaint method should be fast enough for us ... >= 20
> fps :-)
>
> Of course, the direct way gives us quiet the same fps, like the native
> application, but we do pay responsiveness of the JVM for it !
>
> Please check the PERFORMANCE mail from me to the mailinglist,
> and the PerformanceLog's within the demos/MiscDemos directory !
>
> Yours, Sven

Well, the reason I have to call sDisplay in a loop is that I need to very
accurately track how long it takes to do the display.
GLAnuimCanvas tracks frames per second by using System.currentTimeMillis().
This only yields timing accurate to around 16ms (on a good day).  This isn't
really that good (for a game anyway).  Since I'm writing a game, I want very
high frame rates and I need to know very precisely how long each update
takes (so I can move game objects at the right rate).  To solve this
problem, I make a call into a jni library I've written to more accurately
calculate time.  (Using the system clock on windows to achieve near
microsecond accuracy).
Also, I do a Thread.sleep() once per loop (in order to set a maximum fps).

In summary, I pretty much do what GLAnimCanvas does, I just need to be able
to very precisely track how long each update takes (and using
System.currentTimeMillis() isn't good enough)

(For the curious, the Magician Java opengl crap has a percision timer
call... it's pretty each to do, but one might argue that it has nothing to
do with opengl)


Virgil

[gl4java-usergroup] Mac Port: Fetching Function Pointer Dynamically (FindSymbol, ...)

[Sven G. <sgo...@ja...>]

No success ...

This piece of code,
should give us the function-pointer (funcPtr) !

Purpose:
	The function ptr is achieved dynamically out of the
	OpenGL implementation ...

But: It does not work ...

Ideas ?


#include <CodeFragments.h>

void * LIBAPIENTRY getGLProcAddressHelper(const char * func, int *
method,
                                          int debug, int verbose )
{
        void * funcPtr=NULL;
        int lmethod; ...
  

#ifdef __mac_os
        /* funcPtr = GetProcAddress(func); */
        static char buffer[256];
        Ptr glLibMainAddr;
        CFragConnectionID glLibConnectId;
        CFragSymbolClass glLibSymClass;
                       GetSharedLibrary ("OpenGLLibrary",
                                          kCompiledCFragArch
/*kPowerPCCFragArch */,
                                          kStubLibraryCFrag,
                                          &glLibConnectId,
                                          &glLibMainAddr,
                                          buffer);

        printf("GetSharedLibrary errMsg: %s\n", buffer);
        fflush(NULL);

        FindSymbol      (glLibConnectId, func, &funcPtr, & glLibSymClass
);

        /* With this peace, or without,
           just getting NULL ! :-(
	
        if(glLibSymClass == kTVectorCFragSymbol)
        {
                printf("should be found: symbol class equals
kTVectSymbol\n");
        } else {
                printf("should be NOT found: symbol class NOT equals
kTVectSymbol\n");
                funcPtr = NULL;
        }
        */
        fflush(NULL);

        CloseConnection(&glLibConnectId);

        lmethod=2;

#endif                

-- 
mailto:sgo...@ja...
www   : http://www.jausoft.com ; pgp: http://www.jausoft.com/gpg/
voice : +49-521-2399440, +49-170-2115963; fax: +49-521-2399442

Re: [gl4java-usergroup] MAC PORT

[Sven G. <sgo...@ja...>]

aNt wrote:
> 
> > Subject: [gl4java-usergroup] MAC PORT
> >
> > MAC PORT (MacOS 9.0 PPC)
> > ========================
> >
> > I got it :-) !
> > I have created a shared library,
> > which is installed as an extension ...
> > It works.
> 
> nice one, huray :)
> 
> > Now I have still the problem, how I can query the functions (pointers)
> > of the OpenGL functions, located within the OpenGL shared library !
> 
> they all should be in the apple sdk for opengl:
> http://www.apple.com/opengl/
> 
> should all be there.
> 

Nono.

The problem is, that I DYNAMICALLY fetch the function pointer out of the 
GL library, for all GL functions.
This is done, because not all GL functions are implemented in some 
implementations !
So, there are no more dummy nope functions, and no more DLL's 
for special GL implementations !

For Windows, I use:
	wglGetProcAddress
	
	if not succesfull, I use:

		LoadLibrary
		GetProcAddress
		FreeLibrary

For Unix, I use:

	dlopen
	dlsym
	dlclose

	if on of the following exist (queried by mechanism above), I will use:

	glXGetProcAddress, or glXGetProcAddressARB, or glXGetProcAddressEXT

	otherwise, using the dlsym mechanism above !

So - how is it done for the MAC ?


Yours, Sven
-- 
mailto:sgo...@ja...
www   : http://www.jausoft.com ; pgp: http://www.jausoft.com/gpg/
voice : +49-521-2399440, +49-170-2115963; fax: +49-521-2399442

[gl4java-usergroup] MAC PORT

[Sven G. <sgo...@ja...>]

MAC PORT (MacOS 9.0 PPC)
========================

I got it :-) !
I have created a shared library,
which is installed as an extension ...
It works.

Now I have still the problem, how I can query the functions (pointers)
of the OpenGL functions, located within the OpenGL shared library !

HELP !
=======

QUESTIONS !
=============

Should I create a FAT binary ?

Should I only support PPC ?

Whats about MacOSX ?
Does my MacOS 9.0 PPC shared library runs under MacOSX ?

...

-----------------------------

Please answer !

Thanxs, Sven
-- 
mailto:sgo...@ja...
www   : http://www.jausoft.com ; pgp: http://www.jausoft.com/gpg/
voice : +49-521-2399440, +49-170-2115963; fax: +49-521-2399442

Re: [gl4java-usergroup] Errors

[Sven G. <sgo...@ja...>]

Virgil Wall wrote:

> Ok, I think I've got it figured out.  Version 2.3.1.1 was letting me get
> away with all sorts of tom foolery.  For starters, the paint method of
> GLCanvas is being called, which is in turn calling my display.  All of this
> is happening before I've set up appropriate data structures ,etc.  My test
> application is written to basically do this kind of thing (rewritten for
> simplicity):
> 
> public void run(){
>     setupdata();
>     while(true){
>         modifydata();
>         canvas.sDisplay();
>     }
> }
> 

Dear Virgil,

I want to encourage you to use GLAnimCanvas,
or hte way GLAnimCanvas uses the animation update methods !

I see it in many demos, that you use the threaded sDisplay call for
default !
This only works within a very good native multithreading JVM,
and does steals a lot of ressources ...

You can see, that the GLAnimCanvas demos does have the option to
refresh directly (useRepaint... method), but it is disabled for default.

I think, that the repaint method should be fast enough for us ... >= 20
fps :-)

Of course, the direct way gives us quiet the same fps, like the native
application, but we do pay responsiveness of the JVM for it !

Please check the PERFORMANCE mail from me to the mailinglist,
and the PerformanceLog's within the demos/MiscDemos directory !

Yours, Sven
-- 
mailto:sgo...@ja...
www   : http://www.jausoft.com ; pgp: http://www.jausoft.com/gpg/
voice : +49-521-2399440, +49-170-2115963; fax: +49-521-2399442

Re: [gl4java-usergroup] Errors

[Virgil W. <vir...@ra...>]

> > In addition, after switching to version 2.4.1.0, I see these error
messages
> > printed out when my app runs:
> > GL ERROR : invalid operation
> > GL ERROR : 1282 == 0x502
> >
> Nobody will be able to help you until you send more detailed info: a piece
of
> code where is occurs and precise command after this error is signaled (yes
it
> means putting gljCheckGL() after each OpenGL command in this chunk of code
to
> spot the bug).
>

Ok, I think I've got it figured out.  Version 2.3.1.1 was letting me get
away with all sorts of tom foolery.  For starters, the paint method of
GLCanvas is being called, which is in turn calling my display.  All of this
is happening before I've set up appropriate data structures ,etc.  My test
application is written to basically do this kind of thing (rewritten for
simplicity):

public void run(){
    setupdata();
    while(true){
        modifydata();
        canvas.sDisplay();
    }
}

My display method in canvas draws things based on data structures that it
expects will already have been initialized in setupdata().  However, the
paint method of GLCanvas is getting called before I start the above
thread... and paint is calling display (So I'm trying to draw things based
on uninitialized data structures... hence the errors).  The paint method of
GLCanvas is getting called before I Start my thread since I am creating a
frame (and showing it) which holds a GLCanvas before I start the above
thread.

Additionally, I used to have a quad that had a texture mapped to it, but no
bindtexture had been called first. (Silliness, basically)

2.3.1.1 was lets me get away with these two errors without chastising me
(not sure why).  2.4.1.0 doesn't (which is a good thing :)

Thanks,
Virgil

> Max

Re: [gl4java-usergroup] Screen flash and mouse click

[Virgil W. <vir...@ra...>]

> Dear Virgil,
>
> the classes GLCanvas (and therefor GLAnimCanvas) in the package
> gl4java.awt,
> adds itself as a mouselistener, to repaint if:
> mouseClicked
> ...
>
> So, if you use GLCanvas or (especially) GLAnimCanvas,
> just overwrite mouseClicked with a no operation implementation,
> or whatever you want to do.
>
> I made this as an easy solution for the incomplete window refresh,
> if an "own window" is used and it is shown again after hidden by another
> window.



Yep, that did the trick.  Guess I've been out of touch for too long. Looking
at the 2.3.1.1 source I still have I see that the GLCanvas used to not add
itself as a mouselistener.  I was calling sDisplay myself so I had not
bothered overriding update.  So, every mouse click was calling repaint,
which caused an eventual update, etc.

Overriding mouseClicked (and update and paint for completeness) got rid of
the flash.

Thanks Sven and A11W,

Virgil


>
> Yours, Sven
> --
> mailto:sgo...@ja...
> www   : http://www.jausoft.com ; pgp: http://www.jausoft.com/gpg/
> voice : +49-521-2399440, +49-170-2115963; fax: +49-521-2399442
>

Re: [gl4java-usergroup] Screen flash and mouse click

[Virgil W. <vir...@ra...>]

> Dear Virgil,
>
> the classes GLCanvas (and therefor GLAnimCanvas) in the package
> gl4java.awt,
> adds itself as a mouselistener, to repaint if:
> mouseClicked
> ...
>
> So, if you use GLCanvas or (especially) GLAnimCanvas,
> just overwrite mouseClicked with a no operation implementation,
> or whatever you want to do.
>
> I made this as an easy solution for the incomplete window refresh,
> if an "own window" is used and it is shown again after hidden by another
> window.



Yep, that did the trick.  Guess I've been out of touch for too long. Looking
at the 2.3.1.1 source I still have I see that the GLCanvas used to not add
itself as a mouselistener.  I was calling sDisplay myself so I had not
bothered overriding update.  So, every mouse click was calling repaint,
which caused an eventual update, etc.

Overriding mouseClicked (and update and paint for completeness) got rid of
the flash.

Thanks Sven and A11W,

Virgil


>
> Yours, Sven
> --
> mailto:sgo...@ja...
> www   : http://www.jausoft.com ; pgp: http://www.jausoft.com/gpg/
> voice : +49-521-2399440, +49-170-2115963; fax: +49-521-2399442
>

Re: [gl4java-usergroup] Screen flash and mouse click

[Sven G. <sgo...@ja...>]

Virgil Wall wrote:
> 
> I have not spent much time investigating this, however I have been using
> gl4java 2.3.1.1 and just recently switched to version 2.4.1.0.
> After switching to 2.4.1.0 if I click with the mouse pointer rapidly in a
> test app that is drawing an animation, I see a screen flash.  The flash is
> worse if the window is made larger.  It's also easier to notice if the
> window has a light colored background (a black background makes it hard to
> detect).
> 
> When I run the same test app under 2.3.1.1, no flash occurs.
> 
> Anyone else noticed this behavior?
> 
> (Note that my test application is a game, that's why I'm clicking rapidly.
> Also, note that it doesn't matter if I have a mouseListener registered or
> not)
> 
> Thanks,
> Virgil Wall
> 

Dear Virgil,

the classes GLCanvas (and therefor GLAnimCanvas) in the package
gl4java.awt,
adds itself as a mouselistener, to repaint if:
	mouseClicked
	...

So, if you use GLCanvas or (especially) GLAnimCanvas, 
just overwrite mouseClicked with a no operation implementation,
or whatever you want to do.

I made this as an easy solution for the incomplete window refresh,
if an "own window" is used and it is shown again after hidden by another
window.

Yours, Sven
-- 
mailto:sgo...@ja...
www   : http://www.jausoft.com ; pgp: http://www.jausoft.com/gpg/
voice : +49-521-2399440, +49-170-2115963; fax: +49-521-2399442

Re: [gl4java-usergroup] Errors

[Sven G. <sgo...@ja...>]

Max Gilead wrote:
> 
> Virgil Wall wrote:
> 
> > In addition, after switching to version 2.4.1.0, I see these error messages
> > printed out when my app runs:
> > GL ERROR : invalid operation
> > GL ERROR : 1282 == 0x502
> >
> > I don't see these when using version 2.3.1.1.
> >
> > Any idea what they mean?
> 
> Nobody will be able to help you until you send more detailed info: a piece of
> code where is occurs and precise command after this error is signaled (yes it
> means putting gljCheckGL() after each OpenGL command in this chunk of code to
> spot the bug).
> 

... where glGetError is allowed ... !!!

gljCheckGL() exists within the default implementation of:
	- GLAnimCanvas:
		- init
		- display

gljCheckGL calls the OpenGL command glGetError,
so it is restricted like glGetError is restricted !

The GL ERROR number which is printed is the error,
glGetError receives ..
You may want to add a Exception e.printStackTrace(...),
if gljCheckError returns true !

Yours, Sven


-- 
mailto:sgo...@ja...
www   : http://www.jausoft.com ; pgp: http://www.jausoft.com/gpg/
voice : +49-521-2399440, +49-170-2115963; fax: +49-521-2399442

Re: [gl4java-usergroup] Errors

[Max G. <gi...@li...>]

Virgil Wall wrote:

> In addition, after switching to version 2.4.1.0, I see these error messages
> printed out when my app runs:
> GL ERROR : invalid operation
> GL ERROR : 1282 == 0x502
>
> I don't see these when using version 2.3.1.1.
>
> Any idea what they mean?

Nobody will be able to help you until you send more detailed info: a piece of
code where is occurs and precise command after this error is signaled (yes it
means putting gljCheckGL() after each OpenGL command in this chunk of code to
spot the bug).

Max


--
Max Gilead (gi...@li...) http://3d.linart.krakow.pl/OfficinaArtificialis
-----------------------------------------------------------------------------
-- Duck season! Rabbit season! Duck season! Rabbit season! ELMER SEASON!!! --

Re: [gl4java-usergroup] An Application

[Sven G. <sgo...@ja...>]

Dear Martin,

this sounds great !

I have generated the SGI and Solaris port on your machine either,
but I have some problems with the Solaris port.

Can you check the symbols.mak file and/or the way I do generate the
native libs ?

It looks like, that the SGI port works, 
but somebody should test it lokally ....

I need some help with the Mac !
I have downloaded the free Mac compiler & dev environment.
But it looks very time consuming, to do a makefile ...
Do you have some examples ?

I have a Q:
	If we use an own created window, e.g. because the visual of the
	native java window does not fit our needs,
	the window refresh command is not send,
	if another overlapping window disappears.

	So I have added a mouse-click event to redraw the OpenGL component.
	The overlapped part is not refreshed automatically ...

Do you use the dynamically GL function achievemend (like GL4Java does
today),
or do you use statically binding ?

And so on ....

thanxs in advance, 

sven
-- 
mailto:sgo...@ja...
www   : http://www.jausoft.com ; pgp: http://www.jausoft.com/gpg/
voice : +49-521-2399440, +49-170-2115963; fax: +49-521-2399442

Re: [gl4java-usergroup] Errors

[Artiste on t. W. <a1...@So...>]

Virgil Wall wrote:
> 
> In addition, after switching to version 2.4.1.0, I see these error messages
> printed out when my app runs:
> GL ERROR : invalid operation
> GL ERROR : 1282 == 0x502
> 
> I don't see these when using version 2.3.1.1.
> 
> Any idea what they mean?
> 
> Thanks,
> Virgil Wall

I am making a game too (http://perso.club-internet.fr/b_lamy).

I think the click calls a repaint (through AWT refrshing event - not
mouse
listener), in a different thread that your game's one (the AWT event
thread). If your game calls a rendering at the same time, there may be a
synchronization problem, leading to openGL error and bad rendering (the
flash).

This is just an idea, but i think you should disable AWT rendering in
your game's canvas, if your game is rendering 3D permanently, by adding
:

  public void update(Graphics g) {  }
  public void repaint(      ) {  }
  public void repaint(long l) {  }
  public void render() { sDisplay();	}

in your canvas's class. To draw the 3D, you'll use the new render()
method.

A11W

[gl4java-usergroup] An Application

[Martin O. <ma...@ig...>]

Sven

    We have just released the 2.0 version of WebWinds. The 3D rendering
interface between OpenGl and the
various platforms was written with a lot of help from the gl4java group.
However the code has been
severely customized to our particular application as the 3D is just a
part of  a 160,000 line java application.
You can download from http://webwinds.jpl.nasa.gov  (its free) . We have
tested on Mac, Sgi, Sun, Windows
and  Linux and the OpenGl run just fine.

thanks for the help

martin

[gl4java-usergroup] Re: gl4java-usergroup digest, Vol 1 #120 - 2 msgs

[Dev B. <pat...@nc...>]

Hey, has it finally been determined where Ziemski and the Mac port have gone
to? Also, quick question: If I have the source for a program using gl4java
written on a Mac can I recompile it on a PC provided gl4java is installed on
the machine? I'm wondering if the syntax and functionality are the same
between ports.

thanks,

Dev
-- 
Dev Brown
Columbia, SC
USA
And remember kids -- What kind of wood doesn't float?
Natalie Wood!