plib-users

Re: [Plib-users] Matrix rotation question

[Steve B. <sjb...@ai...>]

"Curtis L. Olson" wrote:
> 
> It sounds like you may be experiencing "gimble lock".  If so, you
> should take a look at the definitive document on the subject:
> 
>     http://web2.airmail.net/sjbaker1/eulers_are_evil.html

I agree - this has all the symptoms of that.  When you are pitched up
by 90 degrees, roll and heading end up rotating about the same axis.

Better go learn some quaternions!

----------------------------- Steve Baker -------------------------------
HomeMail : <sjb...@ai...>   WorkMail: <sj...@li...>
HomePage : http://web2.airmail.net/sjbaker1
Projects : http://plib.sf.net     http://tuxaqfh.sf.net  http://tuxkart.sf.net
           http://agtoys.sf.net   http://prettypoly.sf.net
           http://freeglut.sf.net http://toobular.sf.net

Re: [Plib-users] Installing plib-1.4.1

[Lars T. <to...@cs...>]

On Wed, 11 Jul 2001, Sam Stickland wrote:

> 
> ----- Original Message -----
> From: "Lars Tornow" <to...@cs...>
> To: <pli...@li...>
> Sent: Wednesday, July 11, 2001 8:47 PM
> Subject: [Plib-users] Installing plib-1.4.1
> 
> 
> > Hallo,
> >
> > I've got some problems installing plib version 1.4.1. The compiler says...
> >
> > c++ -DPACKAGE=\"plib\" -DVERSION=\"1.4.1\" -DHAVE_LIBDL=1 -DHAVE_LIBGL=1
> > -DHAVE_LIBGLU=1 -DHAVE_LIBGLUT=1 -DHAVE_GL_GL_H=1 -DHAVE_GL_GLU_H=1
> > -DLINUX_JOYSTICK_IS_PRESENT=1 -DGLUT_IS_PRESENT=1  -I. -I.
> > -I../../src/util  -I/usr/local/include  -g -O2 -O6 -Wall -c netSocket.cxx
> > netSocket.cxx: In method `int netSocket::accept(netAddress *)':
> > netSocket.cxx:217: passing `int *' as argument 3 of `accept(int, sockaddr
> > *, socklen_t *)' changes signedness
> > netSocket.cxx: In method `int netSocket::recvfrom(void *, int, int,
> > netAddress *)':
> > netSocket.cxx:254: passing `int *' as argument 6 of `recvfrom(int, void *,
> > unsigned int, int, sockaddr *, socklen_t *)' changes signedness
> > make: *** [netSocket.o] Error 1
> >
> > This happens only in version 1.4.0 and 1.4.1. Plib-1.2.0 compiles fine
> > (but I need 1.4.1 :) for FlightGear-0.7.8; Thanks to booth teams - great
> > work) I have gcc-2.95.2 and a normal SuSE 7.0.
> >
> > Any guesses what goes wrong ?
> 
> Yes - on some platforms socklen_t is an int, and on others an unsigned int.  It appears in these cases 'int' has been written rather
> than socklen_t. So this code will work on some setups, but complain on others.
> 
> I don't have time right now to fix this myself, but I've sent this mail to the developers list so hopefully one of them can pick
> this up.
> 
> Sam
> 

hello, 

I've found a quick solution for this:

an expicit cast to unsigned int

int
netSocket::accept ( netAddress* addr )
{
  assert ( handle != -1 ) ;
  socklen_t addr_len = sizeof(netAddress) ;
  return ::accept(handle,(sockaddr*)addr,&(unsigned int)addr_len);
}

I know it doesn't solve the problem on the root, but ... 
for me it works for now :)

bye lars.

Re: [Plib-users] Matrix rotation question

[Curtis L. O. <cu...@me...>]

It sounds like you may be experiencing "gimble lock".  If so, you
should take a look at the definitive document on the subject:

    http://web2.airmail.net/sjbaker1/eulers_are_evil.html

Regards,

Curt.


McEvoy, Nick writes:
> Sorry matrix operations are not my strong point ... can somebody please
> point out my stupid mistake !?
> 
> I'm developing a space sim with PLIB ... and I have a problem where my
> spaceship behaves in a crazy manner when its heading is directly 'UP' and
> tries to roll to the left or right.
> 
> // Say my initial my position is as follows: (ie. I'm pointing directly up
> ... NOTE: this only happens when pointing directory UP !!!)
> sgCoord mPosition;
> sgSetCoord(&mPosition, 0.0f, 0.0f, 0.0f, 0.0f, 90.0f, 0.0f);
> 
> // Say my direction change is as follows: (ie. I'm rolling to the left)
> sgVec3 mDirChange;
> sgSetVec3(mDirChange, 0.0f, 0.0f, -0.01f);
> 
> // I have a loop where the direction change is applied as follows to update
> the hpr:
> sgMat4 POS, ROT;
> sgMakeCoordMat4(POS, &mPosition);
> sgMakeRotMat4(ROT, mDirChange);
> sgPreMultMat4(POS, ROT);
> sgSetCoord(&mPosition, POS);
> 
> // Why does this loop toggle the hpr part of my mPosition as follows:
> (causing the model to FREAK OUT flipping between ~0 and 90)
> hpr:	0, 90, 0
> hpr:	0, 90, 90.01
> hpr:	0, 90, 6.18986e-006
> hpr:	0, 90, 90.01
> hpr:	0, 90, 1.38196e-005
> hpr:	0, 90, 90.01
> hpr:	0, 90, 1.38196e-005
> etc ...
> 
> Any help appreciated,
> Nick
> mailto:nic...@as...
> http://members.ozemail.com.au/~ndmcevoy/
> 
> 
> _______________________________________________
> plib-users mailing list
> pli...@li...
> http://lists.sourceforge.net/lists/listinfo/plib-users

-- 
Curtis Olson   Human Factors Research Lab       FlightGear Project
Twin Cities    cu...@hf...                cu...@fl...
Minnesota      http://www.menet.umn.edu/~curt   http://www.flightgear.org

[Plib-users] Matrix rotation question

[McEvoy, N. <nic...@ds...>]

Sorry matrix operations are not my strong point ... can somebody please
point out my stupid mistake !?

I'm developing a space sim with PLIB ... and I have a problem where my
spaceship behaves in a crazy manner when its heading is directly 'UP' and
tries to roll to the left or right.

// Say my initial my position is as follows: (ie. I'm pointing directly up
... NOTE: this only happens when pointing directory UP !!!)
sgCoord mPosition;
sgSetCoord(&mPosition, 0.0f, 0.0f, 0.0f, 0.0f, 90.0f, 0.0f);

// Say my direction change is as follows: (ie. I'm rolling to the left)
sgVec3 mDirChange;
sgSetVec3(mDirChange, 0.0f, 0.0f, -0.01f);

// I have a loop where the direction change is applied as follows to update
the hpr:
sgMat4 POS, ROT;
sgMakeCoordMat4(POS, &mPosition);
sgMakeRotMat4(ROT, mDirChange);
sgPreMultMat4(POS, ROT);
sgSetCoord(&mPosition, POS);

// Why does this loop toggle the hpr part of my mPosition as follows:
(causing the model to FREAK OUT flipping between ~0 and 90)
hpr:	0, 90, 0
hpr:	0, 90, 90.01
hpr:	0, 90, 6.18986e-006
hpr:	0, 90, 90.01
hpr:	0, 90, 1.38196e-005
hpr:	0, 90, 90.01
hpr:	0, 90, 1.38196e-005
etc ...

Any help appreciated,
Nick
mailto:nic...@as...
http://members.ozemail.com.au/~ndmcevoy/

Re: [Plib-users] Installing plib-1.4.1

[Lars T. <to...@cs...>]

On Wed, 11 Jul 2001, Sam Stickland wrote:

> 
> ----- Original Message -----
> From: "Lars Tornow" <to...@cs...>
> To: <pli...@li...>
> Sent: Wednesday, July 11, 2001 8:47 PM
> Subject: [Plib-users] Installing plib-1.4.1
> 
> 
> > Hallo,
> >
> > I've got some problems installing plib version 1.4.1. The compiler says...
> >
> > c++ -DPACKAGE=\"plib\" -DVERSION=\"1.4.1\" -DHAVE_LIBDL=1 -DHAVE_LIBGL=1
> > -DHAVE_LIBGLU=1 -DHAVE_LIBGLUT=1 -DHAVE_GL_GL_H=1 -DHAVE_GL_GLU_H=1
> > -DLINUX_JOYSTICK_IS_PRESENT=1 -DGLUT_IS_PRESENT=1  -I. -I.
> > -I../../src/util  -I/usr/local/include  -g -O2 -O6 -Wall -c netSocket.cxx
> > netSocket.cxx: In method `int netSocket::accept(netAddress *)':
> > netSocket.cxx:217: passing `int *' as argument 3 of `accept(int, sockaddr
> > *, socklen_t *)' changes signedness
> > netSocket.cxx: In method `int netSocket::recvfrom(void *, int, int,
> > netAddress *)':
> > netSocket.cxx:254: passing `int *' as argument 6 of `recvfrom(int, void *,
> > unsigned int, int, sockaddr *, socklen_t *)' changes signedness
> > make: *** [netSocket.o] Error 1
> >
> > This happens only in version 1.4.0 and 1.4.1. Plib-1.2.0 compiles fine
> > (but I need 1.4.1 :) for FlightGear-0.7.8; Thanks to booth teams - great
> > work) I have gcc-2.95.2 and a normal SuSE 7.0.
> >
> > Any guesses what goes wrong ?
> 
> Yes - on some platforms socklen_t is an int, and on others an unsigned int.  It appears in these cases 'int' has been written rather
> than socklen_t. So this code will work on some setups, but complain on others.
> 
> I don't have time right now to fix this myself, but I've sent this mail to the developers list so hopefully one of them can pick
> this up.
> 
> Sam
> 

Hello, thanks for the quick answer. Looks like I have the wrong
environment :) Maybe I find some time to look at the problem myself, now
that I know what the problem is.

bye lars.

---------------------------------------------------------
          Real Programmers don't document - 
If it was hard to write, it should be hard to understand. 
---------------------------------------------------------

Re: [Plib-users] Installing plib-1.4.1

[Steve B. <sjb...@ai...>]

Sam Stickland wrote:

> Yes - on some platforms socklen_t is an int, and on others an unsigned int.  It appears in these cases 'int' has been written rather
> than socklen_t. So this code will work on some setups, but complain on others.

Yes - but on yet others, socklen_t is not defined.

There is no way to avoid an ifdef here.  :-(

----------------------------- Steve Baker -------------------------------
HomeMail : <sjb...@ai...>   WorkMail: <sj...@li...>
HomePage : http://web2.airmail.net/sjbaker1
Projects : http://plib.sf.net     http://tuxaqfh.sf.net  http://tuxkart.sf.net
           http://agtoys.sf.net   http://prettypoly.sf.net
           http://freeglut.sf.net http://toobular.sf.net

Re: [Plib-users] Installing plib-1.4.1

[Sam S. <sa...@sp...>]

----- Original Message -----
From: "Lars Tornow" <to...@cs...>
To: <pli...@li...>
Sent: Wednesday, July 11, 2001 8:47 PM
Subject: [Plib-users] Installing plib-1.4.1


> Hallo,
>
> I've got some problems installing plib version 1.4.1. The compiler says...
>
> c++ -DPACKAGE=\"plib\" -DVERSION=\"1.4.1\" -DHAVE_LIBDL=1 -DHAVE_LIBGL=1
> -DHAVE_LIBGLU=1 -DHAVE_LIBGLUT=1 -DHAVE_GL_GL_H=1 -DHAVE_GL_GLU_H=1
> -DLINUX_JOYSTICK_IS_PRESENT=1 -DGLUT_IS_PRESENT=1  -I. -I.
> -I../../src/util  -I/usr/local/include  -g -O2 -O6 -Wall -c netSocket.cxx
> netSocket.cxx: In method `int netSocket::accept(netAddress *)':
> netSocket.cxx:217: passing `int *' as argument 3 of `accept(int, sockaddr
> *, socklen_t *)' changes signedness
> netSocket.cxx: In method `int netSocket::recvfrom(void *, int, int,
> netAddress *)':
> netSocket.cxx:254: passing `int *' as argument 6 of `recvfrom(int, void *,
> unsigned int, int, sockaddr *, socklen_t *)' changes signedness
> make: *** [netSocket.o] Error 1
>
> This happens only in version 1.4.0 and 1.4.1. Plib-1.2.0 compiles fine
> (but I need 1.4.1 :) for FlightGear-0.7.8; Thanks to booth teams - great
> work) I have gcc-2.95.2 and a normal SuSE 7.0.
>
> Any guesses what goes wrong ?

Yes - on some platforms socklen_t is an int, and on others an unsigned int.  It appears in these cases 'int' has been written rather
than socklen_t. So this code will work on some setups, but complain on others.

I don't have time right now to fix this myself, but I've sent this mail to the developers list so hopefully one of them can pick
this up.

Sam

[Plib-users] Installing plib-1.4.1

[Lars T. <to...@cs...>]

Hallo,

I've got some problems installing plib version 1.4.1. The compiler says...

c++ -DPACKAGE=\"plib\" -DVERSION=\"1.4.1\" -DHAVE_LIBDL=1 -DHAVE_LIBGL=1
-DHAVE_LIBGLU=1 -DHAVE_LIBGLUT=1 -DHAVE_GL_GL_H=1 -DHAVE_GL_GLU_H=1
-DLINUX_JOYSTICK_IS_PRESENT=1 -DGLUT_IS_PRESENT=1  -I. -I.
-I../../src/util  -I/usr/local/include  -g -O2 -O6 -Wall -c netSocket.cxx
netSocket.cxx: In method `int netSocket::accept(netAddress *)':
netSocket.cxx:217: passing `int *' as argument 3 of `accept(int, sockaddr
*, socklen_t *)' changes signedness
netSocket.cxx: In method `int netSocket::recvfrom(void *, int, int,
netAddress *)':
netSocket.cxx:254: passing `int *' as argument 6 of `recvfrom(int, void *,
unsigned int, int, sockaddr *, socklen_t *)' changes signedness
make: *** [netSocket.o] Error 1

This happens only in version 1.4.0 and 1.4.1. Plib-1.2.0 compiles fine
(but I need 1.4.1 :) for FlightGear-0.7.8; Thanks to booth teams - great
work) I have gcc-2.95.2 and a normal SuSE 7.0.

Any guesses what goes wrong ?

Thanks in advance, bye lars.

Re: [Plib-users] Flight-Dyanmic-Simulator

[Steve B. <sjb...@ai...>]

Michael Wessels wrote:

> I like to announce my 'Flight-Dynamic-Simulator', which uses Plib for
> the graphics. You will find the simulator on the web-site
> 'www.flight-dynamic-simulator.de'

I'd be happy to put a link to FDS on the PLIB home page - but there is
a problem.  You are only shipping the '.EXE' file for FDS and that's
in violation of the LGPL license for PLIB.

You must either release the sources, or the '.OBJ' files needed to
rebuild the program.

You *MUST* do this immediately - or cease distributing FDS.  THIS
IS A VERY SERIOUS MATTER. (Particularly because you are selling
the full version!)  Once you have done that, you should make an
effort to reach people you've already shipped to so that they have
the opportunity to download the files that they are entitled to.

Please read the file 'LICENSE' at the top of the PLIB directory.

  "If you link a program with the library, you must provide
   complete object files to the recipients so that they can
   relink them with the library, after making changes to the
   library and recompiling it.  And you must show them these
   terms so they know their rights."

I take this matter very seriously.

----------------------------- Steve Baker -------------------------------
HomeMail : <sjb...@ai...>   WorkMail: <sj...@li...>
HomePage : http://web2.airmail.net/sjbaker1
Projects : http://plib.sf.net     http://tuxaqfh.sf.net  http://tuxkart.sf.net
           http://agtoys.sf.net   http://prettypoly.sf.net
           http://freeglut.sf.net http://toobular.sf.net

Re: [Plib-users] Flight-Dyanmic-Simulator

[Steve B. <sjb...@ai...>]

ada...@ho... wrote:
> 
> On Sat, 30 Jun 2001, Michael Wessels wrote:
> 
> > Hi all,
> >
> > I like to announce my 'Flight-Dynamic-Simulator', which uses Plib for
> > the graphics. You will find the simulator on the web-site
> > 'www.flight-dynamic-simulator.de'
> >
> > With best regards
> >
> > Michael
> 
> I can't do that.  Netscape says that site doesn't exist.  And want's
> me to try again.  What should I try next time?

It worked OK for me....try again.

----------------------------- Steve Baker -------------------------------
HomeMail : <sjb...@ai...>   WorkMail: <sj...@li...>
HomePage : http://web2.airmail.net/sjbaker1
Projects : http://plib.sf.net     http://tuxaqfh.sf.net  http://tuxkart.sf.net
           http://agtoys.sf.net   http://prettypoly.sf.net
           http://freeglut.sf.net http://toobular.sf.net

Re: [Plib-users] Flight-Dyanmic-Simulator

[<ada...@ho...>]

On Sat, 30 Jun 2001, Michael Wessels wrote:

> Hi all,
>
> I like to announce my 'Flight-Dynamic-Simulator', which uses Plib for
> the graphics. You will find the simulator on the web-site
> 'www.flight-dynamic-simulator.de'
>
> With best regards
>
> Michael

I can't do that.  Netscape says that site doesn't exist.  And want's
me to try again.  What should I try next time?



~~
     ``This is the most beautiful coin I have ever seen.  It has amazing
     detail, and that 3D, holographic image makes it look like it's from
     some futuristic, sci-fi movie.''
               -- An actual, American, TV commercial

         /||  Adam Seyfarth    <mailto:ada...@ho...  ||\
        <=AB||  <http://tuxaqfh.sf.net> <http://tuxkart.sf.net>  ||=BB>
         \||  <http://plib.sf.net>        <http://vim.sf.net>  ||/

[Plib-users] Flight-Dyanmic-Simulator

[Michael W. <michael.wessels@z.zgs.de>]

Hi all,

I like to announce my 'Flight-Dynamic-Simulator', which uses Plib for
the graphics. You will find the simulator on the web-site
'www.flight-dynamic-simulator.de'

With best regards

Michael

Re: [Plib-users] js_demo->Btns

[Steve B. <sjb...@ai...>]

Adam wrote:

> linux.  Here's my example (sorry, different PROMPT)
> 
> <1> ./js_demo                    /work/adam/plib_examples-1.4.0/src/js
> 
> <adam presses A,B,C and D all at once>
> 
> Joystick test program.
> ~~~~~~~~~~~~~~~~~~~~~~
> Joystick 1 not detected
> +---------------JS.0-----------------+---------------JS.1-----------------+
> | Btns Ax:0 Ax:1                     |       ~~~ Not Detected ~~~         |
> +------------------------------------+------------------------------------+
> | 0000 +0.0 +0.0   .    .    .    .  |  .   .   .   .   .   .   .   .   . |
> 
> (note: I tar'd it in /work/adam)...where's the 'f'?  If it's just
> something that's not displaying right, (ie, *working* right, but not
> *displaying* right) then I'm happy.  But it's haunting.....

Hmmm - I wonder if there is something electrically 'funny' - like the
wire providing voltage to the joystick to the PC is kinda skimpy and
when you try to pull up four input pins at once, the voltage drops enough
to prevent the port from seeing it as 1-bits.

That's rather a long-shot - but I simply can't imagine how printf could
be broken in such an obvious way.

----------------------------- Steve Baker -------------------------------
HomeMail : <sjb...@ai...>   WorkMail: <sj...@li...>
HomePage : http://web2.airmail.net/sjbaker1
Projects : http://plib.sf.net     http://tuxaqfh.sf.net  http://tuxkart.sf.net
           http://agtoys.sf.net   http://prettypoly.sf.net
           http://freeglut.sf.net http://toobular.sf.net

Re: [Plib-users] js_demo->Btns

[Adam <ada...@ho...>]

On Fri, 29 Jun 2001, Steve Baker wrote:

> Adam wrote:
> >
> > Hi,
> >     Has anyone noticed this before?  When you do the supplied
> > js_demo program, making one of those hex-numbers=3D'f' then it prints=
 as
> > '0'.  I don't think it's printf's fault. (it has worked for me) If
> > this hasn't been fixed already, I'll give it a try.
>
> So are you saying that if I press four joystick buttons down at once
> (A,B,C,D for example) then those four bits will make an 'f' in hex and
> that doesn't get printed out correctly?
>
> Well, that particular test works just fine for me...
>
> src/js> ./js_demo
>
> <steve presses A,B,C and D all at once>
>
> Joystick test program.
> ~~~~~~~~~~~~~~~~~~~~~~
> Joystick 1 not detected
> +---------------JS.0-----------------+---------------JS.1--------------=
---+
> | Btns Ax:0 Ax:1                     |       ~~~ Not Detected ~~~      =
   |
> +------------------------------------+---------------------------------=
---+
> | 000f +0.0 +1.0   .    .    .    .  |  .   .   .   .   .   .   .   .  =
 . |
>
> ...no problem!
>
> What OS are you using?

linux.  Here's my example (sorry, different PROMPT)

<1> ./js_demo                    /work/adam/plib_examples-1.4.0/src/js

<adam presses A,B,C and D all at once>

Joystick test program.
~~~~~~~~~~~~~~~~~~~~~~
Joystick 1 not detected
+---------------JS.0-----------------+---------------JS.1----------------=
-+
| Btns Ax:0 Ax:1                     |       ~~~ Not Detected ~~~        =
 |
+------------------------------------+-----------------------------------=
-+
| 0000 +0.0 +0.0   .    .    .    .  |  .   .   .   .   .   .   .   .   .=
 |

(note: I tar'd it in /work/adam)...where's the 'f'?  If it's just
something that's not displaying right, (ie, *working* right, but not
*displaying* right) then I'm happy.  But it's haunting.....



--=20
     There's no easy quick way out, we're gonna have to live through our
     whole lives, win, lose, or draw.
               -- Walt Kelly

         /||  name..............................Adam Seyfarth  ||\
        /=AB||  email...........<mailto:ada...@ho...>  ||=BB\
        \=AB||  <http://tuxaqfh.sf.net> <http://tuxkart.sf.net>  ||=BB/
         \||  <http://plib.sf.net>        <http://vim.sf.net>  ||/

Re: [Plib-users] js_demo->Btns

[Steve B. <sjb...@ai...>]

Adam wrote:
> 
> Hi,
>     Has anyone noticed this before?  When you do the supplied
> js_demo program, making one of those hex-numbers='f' then it prints as
> '0'.  I don't think it's printf's fault. (it has worked for me) If
> this hasn't been fixed already, I'll give it a try.

So are you saying that if I press four joystick buttons down at once
(A,B,C,D for example) then those four bits will make an 'f' in hex and
that doesn't get printed out correctly?

Well, that particular test works just fine for me...

src/js> ./js_demo

<steve presses A,B,C and D all at once>

Joystick test program.
~~~~~~~~~~~~~~~~~~~~~~
Joystick 1 not detected
+---------------JS.0-----------------+---------------JS.1-----------------+
| Btns Ax:0 Ax:1                     |       ~~~ Not Detected ~~~         |
+------------------------------------+------------------------------------+
| 000f +0.0 +1.0   .    .    .    .  |  .   .   .   .   .   .   .   .   . | 

...no problem!

What OS are you using?

----------------------------- Steve Baker -------------------------------
HomeMail : <sjb...@ai...>   WorkMail: <sj...@li...>
HomePage : http://web2.airmail.net/sjbaker1
Projects : http://plib.sf.net     http://tuxaqfh.sf.net  http://tuxkart.sf.net
           http://agtoys.sf.net   http://prettypoly.sf.net
           http://freeglut.sf.net http://toobular.sf.net

[Plib-users] js_demo->Btns

[Adam <ada...@ho...>]

Hi,
    Has anyone noticed this before?  When you do the supplied
js_demo program, making one of those hex-numbers=3D'f' then it prints as
'0'.  I don't think it's printf's fault. (it has worked for me) If
this hasn't been fixed already, I'll give it a try.



--=20
     There comes a time in the affairs of a man when he has to take
     the bull by the tail and face the situation.
               -- W.C. Fields

         /||  name..............................Adam Seyfarth  ||\
        /=AB||  email...........<mailto:ada...@ho...>  ||=BB\
        \=AB||  <http://tuxaqfh.sf.net> <http://tuxkart.sf.net>  ||=BB/
         \||  <http://plib.sf.net>        <http://vim.sf.net>  ||/

Re: [Plib-users] Toying with Ogg

[Cameron M. <li...@to...>]

* sjb...@ai... [2001.06.29 02:26]:
> Cameron Moore wrote:
> > 
> > * sjb...@ai... [2001.06.29 00:20]:
> > > Cameron Moore wrote:
> > >
> > > > I'm playing around with Ogg Vorbis at the moment, and I'm stuck trying
> > > > to figure out how to use it with plib.  My first goal is to use
> > > > slSample( Uchar *buff, int leng ) to play the stream.
> > >
> > > Isn't that going to be a rather large file though?
> > 
> > If by 'file' you mean the .ogg file, then why does it have to be large?
> > What are you getting at?
> 
> The input '.ogg' files are highly compressed - right - but the output is
> just a regular audio stream - so it's gonna be *HUGE* - if you have figured
> out how to get Ogg to generate (say) 1/10th second of audio into a buffer,
> then return to the calling program - then you can make it stream into
> PLIB/SL - but if it insists on unpacking the entire track into a buffer,
> or a file on disk then it's going to be REALLY big.

Ahh...I see what you mean.  And yes, libvorbisfile decodes the streams
in sections (which you define the size of), so you could decode a
section and send it to plib.

> > > But I think you really need to find a way for Ogg's output to continuously
> > > churn out bytes and feed them into SL...that's *hard* to do...but 2.4 Megabytes
> > > per minute of music is a heck of a lot of memory to burn.
> > 
> > That's probably more work than I want to do, but we'll see.  8-)  And
> > about the mem usage, how is ogg harder on memory than wav or au?
> 
> Oh - it's not harder - it should be exactly the same...but we don't normally
> ship in-game music as WAV or AU...we use MOD - or something similar that
> is *MUCH* more compact.

Okay, I wasn't familiar with MOD.

> > IIRC, Q3A uses MP3s for some of it's intro music and doesn't seem to
> > have any problems with it.
> 
> Well, it ships on CD-ROM for starters!  But in any case, it's unpacking
> the MP3 into raw uncompressed audio in realtime...so it never has to
> hold than a tiny fraction of a second of audio at one time...just
> before the hardware has finished playing that small chunk, you go back to
> the MP3 (or Ogg or MOD) decoder and tell it to generate the next tiny
> chunk.

Okay, this is starting to make more sense to me.  Thanks for clearing
some of this up.
-- 
Cameron Moore

Re: [Plib-users] Toying with Ogg

[Steve B. <sjb...@ai...>]

Cameron Moore wrote:
> 
> * sjb...@ai... [2001.06.29 00:20]:
> > Cameron Moore wrote:
> >
> > > I'm playing around with Ogg Vorbis at the moment, and I'm stuck trying
> > > to figure out how to use it with plib.  My first goal is to use
> > > slSample( Uchar *buff, int leng ) to play the stream.
> >
> > Isn't that going to be a rather large file though?
> 
> If by 'file' you mean the .ogg file, then why does it have to be large?
> What are you getting at?

The input '.ogg' files are highly compressed - right - but the output is
just a regular audio stream - so it's gonna be *HUGE* - if you have figured
out how to get Ogg to generate (say) 1/10th second of audio into a buffer,
then return to the calling program - then you can make it stream into
PLIB/SL - but if it insists on unpacking the entire track into a buffer,
or a file on disk then it's going to be REALLY big.

> > But I think you really need to find a way for Ogg's output to continuously
> > churn out bytes and feed them into SL...that's *hard* to do...but 2.4 Megabytes
> > per minute of music is a heck of a lot of memory to burn.
> 
> That's probably more work than I want to do, but we'll see.  8-)  And
> about the mem usage, how is ogg harder on memory than wav or au?

Oh - it's not harder - it should be exactly the same...but we don't normally
ship in-game music as WAV or AU...we use MOD - or something similar that
is *MUCH* more compact.

> IIRC, Q3A uses MP3s for some of it's intro music and doesn't seem to
> have any problems with it.

Well, it ships on CD-ROM for starters!  But in any case, it's unpacking
the MP3 into raw uncompressed audio in realtime...so it never has to
hold than a tiny fraction of a second of audio at one time...just
before the hardware has finished playing that small chunk, you go back to
the MP3 (or Ogg or MOD) decoder and tell it to generate the next tiny
chunk.

However, if you have to call the decoder just once and have it decode
the entire track - then you'll have HUGE memory or disk buffers to
contend with....that's all I'm trying to warn you about.

----------------------------- Steve Baker -------------------------------
HomeMail : <sjb...@ai...>   WorkMail: <sj...@li...>
HomePage : http://web2.airmail.net/sjbaker1
Projects : http://plib.sf.net     http://tuxaqfh.sf.net  http://tuxkart.sf.net
           http://agtoys.sf.net   http://prettypoly.sf.net
           http://freeglut.sf.net http://toobular.sf.net

Re: [Plib-users] Toying with Ogg

[Cameron M. <li...@to...>]

* sjb...@ai... [2001.06.29 00:20]:
> Cameron Moore wrote:
> 
> > I'm playing around with Ogg Vorbis at the moment, and I'm stuck trying
> > to figure out how to use it with plib.  My first goal is to use
> > slSample( Uchar *buff, int leng ) to play the stream.
> 
> Isn't that going to be a rather large file though?

If by 'file' you mean the .ogg file, then why does it have to be large?
What are you getting at?

> > The problem for this newbie is that libvorbisfile converts the .ogg
> > stream into a char[] buffer (PCM) but plib wants a Uchar* buffer.  Is
> > there a trick to converting between these datatypes?
> 
> Well, it depends on how Ogg translates that into a voltage.
> 
> If it's signed char, then probably  +127 is the max positive, zero is
> zero and -128 is max negative...as you'd expect.
> 
> SL is using 255 is max positive, 128 is zero volts and 0 is max negative,
> so, looking at the bit patterns:
> 
>    volts    Ogg          SL
>     +max    01111111     11111111
>             00111111     10111111
>     zero    00000000     10000000
>             11000000     01000000
>     -max    10000000     00000000
> 
> ...so you just need to flip the topmost bit of each byte...
> 
>    slByte = oggByte ^ 0x80 ;
> 
> Tadaaa!

You rock.

> But I think you really need to find a way for Ogg's output to continuously
> churn out bytes and feed them into SL...that's *hard* to do...but 2.4 Megabytes
> per minute of music is a heck of a lot of memory to burn.

That's probably more work than I want to do, but we'll see.  8-)  And
about the mem usage, how is ogg harder on memory than wav or au?
IIRC, Q3A uses MP3s for some of it's intro music and doesn't seem to
have any problems with it.
-- 
Cameron Moore

Re: [Plib-users] Toying with Ogg

[Steve B. <sjb...@ai...>]

Cameron Moore wrote:

> I'm playing around with Ogg Vorbis at the moment, and I'm stuck trying
> to figure out how to use it with plib.  My first goal is to use
> slSample( Uchar *buff, int leng ) to play the stream.

Isn't that going to be a rather large file though?

> The problem for this newbie is that libvorbisfile converts the .ogg
> stream into a char[] buffer (PCM) but plib wants a Uchar* buffer.  Is
> there a trick to converting between these datatypes?

Well, it depends on how Ogg translates that into a voltage.

If it's signed char, then probably  +127 is the max positive, zero is
zero and -128 is max negative...as you'd expect.

SL is using 255 is max positive, 128 is zero volts and 0 is max negative,
so, looking at the bit patterns:

   volts    Ogg          SL
    +max    01111111     11111111
            00111111     10111111
    zero    00000000     10000000
            11000000     01000000
    -max    10000000     00000000

...so you just need to flip the topmost bit of each byte...

   slByte = oggByte ^ 0x80 ;

Tadaaa!

But I think you really need to find a way for Ogg's output to continuously
churn out bytes and feed them into SL...that's *hard* to do...but 2.4 Megabytes
per minute of music is a heck of a lot of memory to burn.

----------------------------- Steve Baker -------------------------------
HomeMail : <sjb...@ai...>   WorkMail: <sj...@li...>
HomePage : http://web2.airmail.net/sjbaker1
Projects : http://plib.sf.net     http://tuxaqfh.sf.net  http://tuxkart.sf.net
           http://agtoys.sf.net   http://prettypoly.sf.net
           http://freeglut.sf.net http://toobular.sf.net

[Plib-users] Toying with Ogg

[Cameron M. <li...@to...>]

Hello all,
I'm playing around with Ogg Vorbis at the moment, and I'm stuck trying
to figure out how to use it with plib.  My first goal is to use
slSample( Uchar *buff, int leng ) to play the stream.

The problem for this newbie is that libvorbisfile converts the .ogg
stream into a char[] buffer (PCM) but plib wants a Uchar* buffer.  Is
there a trick to converting between these datatypes?

The first person to implement LoadOggFile() gets to be my hero for a
week (unless your name is Steve Baker, then your just more of a hero
than last week).  ;-)
-- 
Cameron Moore

Re: [Plib-users] transparent cube

[Sam S. <sa...@sp...>]

----- Original Message -----
From: "Steve Baker" <sjb...@ai...>
To: <pli...@li...>
Sent: Wednesday, June 27, 2001 4:17 PM
Subject: Re: [Plib-users] transparent cube


> Marc Escher wrote:
> >
> > Thanks, that does the job well enough.
>
> It's only "right" when all the translucent surfaces are the same colour
> though.
>
> If you use textured alpha for things like trees, the above hack is a
> disaster.  That's why IT MUST NOT BE COMMITTED INTO PLIB.  Be very careful
> not to make this change in the directory where you do your PLIB CVS checkin's.

I use two directories for plib - once checked out via SSH and one checked out anonymously.  I use the anonymous version for day to
day use, and it contains all my hacks and stuff.

If I want to commit a change back into CVS I do an update on both versions (the SSH version should show no modified files), and then
run a diff over both directories.  I have a nice graphical merge tool that allows me to see the differences between the files and
copy over the required changes. And then I test and commit the SSH version.

This also has the advantage that I get to review all the changes again before doing a commit (not that I've done that many :) ).

Sam

Re: [Plib-users] transparent cube

[Steve B. <sjb...@ai...>]

Marc Escher wrote:
> 
> At 09:36 AM 6/26/01 -0700, you wrote:
> >We recently had a discussion about this on plib-devel.  Read this
> >
> >http://web2.airmail.net/sjbaker1/alpha_sorting.html
> >
> >For my game, I simply disable Z-write translucent polygons.  Here's my hack
> >for ssgDList.cxx
> >
> >void _ssgDrawDList ()
> >{
> >   glDepthMask ( false ) ;
> >
> >   for ( int i = 0 ; i < next_dlist ; i++ )
> >     dlist [ i ] . draw () ;
> >
> >   next_dlist = 0 ;
> >
> >   glDepthMask ( true ) ;
> >}
> 
> Thanks, that does the job well enough.

It's only "right" when all the translucent surfaces are the same colour
though.

If you use textured alpha for things like trees, the above hack is a
disaster.  That's why IT MUST NOT BE COMMITTED INTO PLIB.  Be very careful
not to make this change in the directory where you do your PLIB CVS checkin's.

----------------------------- Steve Baker -------------------------------
HomeMail : <sjb...@ai...>   WorkMail: <sj...@li...>
HomePage : http://web2.airmail.net/sjbaker1
Projects : http://plib.sf.net     http://tuxaqfh.sf.net  http://tuxkart.sf.net
           http://agtoys.sf.net   http://prettypoly.sf.net
           http://freeglut.sf.net http://toobular.sf.net

Re: [Plib-users] transparent cube

[Marc E. <ma...@fa...>]

At 10:01 PM 6/26/01 -0500, you wrote:
>Marc Escher wrote:
>
> > I still have some problems with transparencies :-( . From the doc and your
> > previews answers PLIB renders translucent polygon in no particular order.
>
>That's true.
>
> > Unfortunately I have to display some objects that globally act like
> > transparent cubes,  with a correct rendering of the faces in the back. Is
> > there a way to force the drawing order of leaves in ssg ? Or any way around
> > to get the right result ?
>
>We have had a contribution of an alpha-polygon sorter (well - I think it's
>actually a transparent *OBJECT* sorter)...I just havn't gotten around to
>checking it into the CVS yet.
>
>It won't be in a 1.4.x release - we're putting it into 1.5.xx - so it'll
>be *formally* released in the 1.6.0 release in a few months time.

Well, I suppose I'll have to wait, and go on with the simple "DepthMask" hack.

Thanks, anyway.


Marc.


___________Escher Marc__________Alternet Fabric, sarl_____

rue de Langallerie, 6          Tel     : ++41-21-351.10.20
CH-1003 Lausanne            Fax    : ++41-21-351.10.20
                                         mailto : ma...@fa...
http://www.fabric.ch          smsto  : http://www.flashsms.com/marc
_____________________________________________________
--- Pingouins dans les champs, hiver mechant. ---

RE: [Plib-users] transparent cube

[Marc E. <ma...@fa...>]

At 09:36 AM 6/26/01 -0700, you wrote:
>We recently had a discussion about this on plib-devel.  Read this
>
>http://web2.airmail.net/sjbaker1/alpha_sorting.html
>
>For my game, I simply disable Z-write translucent polygons.  Here's my hack
>for ssgDList.cxx
>
>void _ssgDrawDList ()
>{
>   glDepthMask ( false ) ;
>
>   for ( int i = 0 ; i < next_dlist ; i++ )
>     dlist [ i ] . draw () ;
>
>   next_dlist = 0 ;
>
>   glDepthMask ( true ) ;
>}


Thanks, that does the job well enough.


Marc.


___________Escher Marc__________Alternet Fabric, sarl_____

rue de Langallerie, 6          Tel     : ++41-21-351.10.20
CH-1003 Lausanne            Fax    : ++41-21-351.10.20
                                         mailto : ma...@fa...
http://www.fabric.ch          smsto  : http://www.flashsms.com/marc
_____________________________________________________
--- Pingouins dans les champs, hiver mechant. ---