plib-users

[Plib-users] ssgaShape subclass -- how to handle ssgaType*?

[Daniel S. <djs...@uc...>]

I have been subclassing ssgaShapes to customize the shapes I am drawing.

I need to implement ssgaShape::getTypeName() -- should I do this with my 
own name or one of the existing ssg shape names? What are the 
implications of using my own name?

In a similar vein, my shapes implement a "Type" function, like this

#define _SSGA_TYPE_RECT       0x00200000
inline int ssgaTypeRect  () { return _SSGA_TYPE_RECT   | ssgaTypeShape ();}

This simply follows the example of the shapes defined in ssg.

I'm not comfortable doing this, as it may be incompatible with future 
additions to the ssg shape collection. At the same time, I don't see 
where this type is used after I initialize the 'type' member in my 
class's constructor. Can anyone educate me on this?

Is there a cleaner way to do this, so that I won't have trouble if/when 
ssg adds new shape def's?

Dan

Re: [Plib-users] Texture remains enabled after first use

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

Daniel Sperka wrote:

> Sounds like the best thing is to create a dummy SimpleState and use it 
> for ALL objects in the tree.  If I use texture anywhere, create a 
> distinct SimpleState and use it for those objects that need it, and make 
> certain that the dummy state explicitly disables GL_TEXTURE_2D. Same 
> goes for other GL states.

Yes - that's certainly a reasonable approach.

What I do in my games is to use the ssgLoaderOptions::setCreateStateCallback()
to have the loaders call my application whenever a new ssgState is needed.
I use the texture filename to look up the full desired GL state in a text
file that I load separately on startup.  All polygons without texture (and
there are VERY few of those in my applications) are given a 1x1 pure white
texture instead - so SSG doesn't have to turn texture on and off.

I also use that to set things like the Alpha clamp value that AC3D doesn't
give you a way to set in the model.

That text file also tells my application things about the polygons
(such as their 'solidity', their coefficient of friction and their
temperature) that SSG doesn't need to know - but which matter in the game.
It's really easy to tag my fire texture as being hot and insubstantial,
and my ice texture as being cold, hard and slippery.

There is also a default entry in the table for polygons with textures
that aren't listed in the file.

I also (in my son's first serious game effort) swap out some polygons
for particle systems based on their texture.  So I paint a texture
which is red with the words "IMAGINE A FIRE HERE" written on it.  You
can model a triangle with that texture on it - and when that model
loads, the game's loader will replace it with an ssgaFire of similar
size.  Now you can light fires in your game level just by dropping
in these special polygons in AC3D (or whatever).

Someday, it would be nice to make that feature be a part of ssgAux,
but I don't have the time right now.

---------------------------- Steve Baker -------------------------
HomeEmail: <sjb...@ai...>    WorkEmail: <sj...@li...>
HomePage : http://www.sjbaker.org
Projects : http://plib.sf.net    http://tuxaqfh.sf.net
            http://tuxkart.sf.net http://prettypoly.sf.net
-----BEGIN GEEK CODE BLOCK-----
GCS d-- s:+ a+ C++++$ UL+++$ P--- L++++$ E--- W+++ N o+ K? w--- !O M-
V-- PS++ PE- Y-- PGP-- t+ 5 X R+++ tv b++ DI++ D G+ e++ h--(-) r+++ y++++
-----END GEEK CODE BLOCK-----

Re: [Plib-users] Texture remains enabled after first use

[Daniel S. <djs...@uc...>]

Thanks, Steve.

Sounds like the best thing is to create a dummy SimpleState and use it 
for ALL objects in the tree.  If I use texture anywhere, create a 
distinct SimpleState and use it for those objects that need it, and make 
certain that the dummy state explicitly disables GL_TEXTURE_2D. Same 
goes for other GL states.

Dan

Re: [Plib-users] Texture remains enabled after first use

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

Daniel Sperka wrote:

> If I explicitly use a state (for the non-textured shape) that calls
>    rect_state -> disable            ( GL_TEXTURE_2D ) ;
> then the problem disappears also.

That is the correct solution.

The way ssgState works (and in retrospect, this may not have been so smart)
is that each attribute (like texture being enabled) has THREE states: Enabled,
Disabled and Don't Care.

By default, everything is Don't Care.  So if you don't explicitly disable
or enable texture, you'll get whatever state was last applied.

This is a good thing for *most* state elements because very often you don't
care what state the graphic pipe is in for many of the states and SSG can
save time by not making lots of redundant state change calls.

> Is this by design?

Yes.

> If so, is this true of other enable/disable states?

Yes.

---------------------------- Steve Baker -------------------------
HomeEmail: <sjb...@ai...>    WorkEmail: <sj...@li...>
HomePage : http://www.sjbaker.org
Projects : http://plib.sf.net    http://tuxaqfh.sf.net
            http://tuxkart.sf.net http://prettypoly.sf.net
-----BEGIN GEEK CODE BLOCK-----
GCS d-- s:+ a+ C++++$ UL+++$ P--- L++++$ E--- W+++ N o+ K? w--- !O M-
V-- PS++ PE- Y-- PGP-- t+ 5 X R+++ tv b++ DI++ D G+ e++ h--(-) r+++ y++++
-----END GEEK CODE BLOCK-----

[Plib-users] Texture remains enabled after first use

[Daniel S. <djs...@uc...>]

Hi -

I have a simple app which draws a textured shape, then draws a 
non-textured (solid color) shape. Problem is that the non-textured shape 
uses the same texture that the textured shape did.

Full disclosure: My non-textured shape is a custom class derived from 
ssgaShape. It implements a rectangle in this test phase.

If I reverse the order that the shapes are added to the root, the 
problem disappears.

If I explicitly use a state (for the non-textured shape) that calls
    rect_state -> disable            ( GL_TEXTURE_2D ) ;
then the problem disappears also.


It seems that when a leaf which uses texture is rendered, the GL state 
GL_TEXTURE_2D remains enabled until it is explicitly disabled.

Is this by design?

If so, is this true of other enable/disable states?


Thanks,

Dan

Re: [Plib-users] Model infinite plane

[<jni...@ya...>]

doesn't the ssga ocean wave class use an infinite
plane?

that is if you fly over, the "ocean" plane never ends.

check it out and see if that helps.


--- Daniel Sperka <djs...@uc...> wrote:

> Hi -
> 
> I'm looking for suggestions as to an efficient way
> to model an infinite 
> plane. My thought is to subclass VtxTable and use
> the pre-cull callback 
> to compute the intersection of the plane with the
> frustum, and place 
> those points into the VtxTable prior to returning.
> 
> This is the general case -- i.e. where the plane
> itself may not be 
> aligned with the frustum in any particular way.
> 
> Dan
> 
> 
>
-------------------------------------------------------
> SF email is sponsored by - The IT Product Guide
> Read honest & candid reviews on hundreds of IT
> Products from real users.
> Discover which products truly live up to the hype.
> Start reading now. 
> http://productguide.itmanagersjournal.com/
> _______________________________________________
> plib-users mailing list
> pli...@li...
>
https://lists.sourceforge.net/lists/listinfo/plib-users
> 

[Plib-users] Model infinite plane

[Daniel S. <djs...@uc...>]

Hi -

I'm looking for suggestions as to an efficient way to model an infinite 
plane. My thought is to subclass VtxTable and use the pre-cull callback 
to compute the intersection of the plane with the frustum, and place 
those points into the VtxTable prior to returning.

This is the general case -- i.e. where the plane itself may not be 
aligned with the frustum in any particular way.

Dan

Re: [Plib-users] Skinning

[Arnt K. <ar...@c2...>]

On Sun, 26 Dec 2004 21:45:34 -0600, Steve wrote in message 
<41C...@ai...>:

> Antoine Rennuit wrote:
> 
> > Is there a support for skinning in plib ? Maybe an attempt?
> 
> I'm not quite sure what you mean by that - but whatever it is,
> it doesn't sound like anything we'd be doing.

.."show us!", is a better approach, unless you're telling him you 
do _not_ want any skinning at all in Plib, regardless of whatever 
he means by "skinning", and whatever he feels might merit it, 
he would the one to show us, and really, regardless of what 
you meant to say, you said "No" to that "show us."

..that said, there's talk about a need for new kindsa 3d model 
loaders over at flightgear(.org)-devel, .blender and .ac3d etc,
itching volonteers are always welcome to "show us."  ;-)

-- 
..med vennlig hilsen = with Kind Regards from Arnt... ;-)
...with a number of polar bear hunters in his ancestry...
  Scenarios always come in sets of three: 
  best case, worst case, and just in case.

Re: [Plib-users] Skinning

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

Antoine Rennuit wrote:

> Is there a support for skinning in plib ? Maybe an attempt?

I'm not quite sure what you mean by that - but whatever it is,
it doesn't sound like anything we'd be doing.

---------------------------- Steve Baker -------------------------
HomeEmail: <sjb...@ai...>    WorkEmail: <sj...@li...>
HomePage : http://www.sjbaker.org
Projects : http://plib.sf.net    http://tuxaqfh.sf.net
            http://tuxkart.sf.net http://prettypoly.sf.net
-----BEGIN GEEK CODE BLOCK-----
GCS d-- s:+ a+ C++++$ UL+++$ P--- L++++$ E--- W+++ N o+ K? w--- !O M-
V-- PS++ PE- Y-- PGP-- t+ 5 X R+++ tv b++ DI++ D G+ e++ h--(-) r+++ y++++
-----END GEEK CODE BLOCK-----

[Plib-users] Skinning

[Antoine R. <ant...@la...>]

Hello,

 

Is there a support for skinning in plib ? Maybe an attempt?

Re: [Plib-users] PLIB (pui) + SDL?

[Slowhand_47 <slo...@gm...>]

Am Wed, 22 Dec 2004 23:34:05 +0100 schrieb Alberto BARSELLA <is...@fr...>:

> (MapUI.o is my object file, the source, MapUI.cpp only contains an 
> #include <plib/pu.h> and a call to puInit(), for now).
> I could not find anything similar by searching with google, but a lot of 
> places seem to indicate that plib should work find with SDL... what am I 
> doing wrong?

Does your file contain "#include <SDL.h>" *before* you include
pu.h? Have you set the #define PU_USING_SDL?

I'm currently integrating PUI into an SDL application, so I can
assure you it works. I had some trouble compiling the SDL-example
from the PLIB-CVS (only showed a black screen), but in my
application it works fine.

In your MapUI.cpp you should have lines like this:

#include <SDL.h>
#include <plib/pu.h>

SDL.h must be included before pu.h! Then, make sure that
your CFLAGS and LFLAGS contain all necessary stuff for SDL
(sdl-config --cflags and sdl-config --libs). Add a compiler
switch which defines the above mentioned macro. For gcc this
would be: -DPU_USING_SDL
Last, make sure to link plibpu, plibul and plibfnt to your
application.

Kind regards,
Jan

-- 
Jan Reucker
email: jan dot reucker at web dot de
web:   http://www.reucker-online.de

[Plib-users] PLIB (pui) + SDL?

[Alberto B. <is...@fr...>]

Hello,
	I'm trying to use PUI inside an SDL+openGL application, but my 
compilation fails with:

MapUI.o(.gnu.linkonce.t._Z9puInitSDLv+0x15): In function `puInitSDL()':
/usr/lib/gcc/i586-mandrake-linux-gnu/3.4.1/../../../../include/c++/3.4.1/bits/stl_algobase.h:156: 
undefined reference to `puSetWindowFuncs(int (*)(), void (*)(int), void 
(*)(int*, int*), void (*)(int, int))'
MapUI.o(.gnu.linkonce.t._Z9puInitSDLv+0x1d): In function 
`std::basic_ostream<char, std::char_traits<char> >& operator<< 
<int>(std::basic_ostream<char, std::char_traits<char> >&, _xypoint<int> 
const&)':
/usr/include/plib/puSDL.h:51: undefined reference to `puRealInit()'

(MapUI.o is my object file, the source, MapUI.cpp only contains an 
#include <plib/pu.h> and a call to puInit(), for now).
I could not find anything similar by searching with google, but a lot of 
places seem to indicate that plib should work find with SDL... what am I 
doing wrong?
(I'm running Mandrake10.1, i586 version, I tried both the plib-devel 
package provided by mandrake and I recompiled plib myself).

Thanks in advance,
Alberto
-- 
Alberto BARSELLA
** Beliefs are dangerous. Beliefs allow the mind to stop functioning.
A non-functioning mind is clinically dead.  Believe in nothing... **

Re: [Plib-users] Sky and Pw

[Benjamin E. T. <be...@cs...>]

so, after about an hour of reading up on glxChooseVisual and playing
around with the pw sourcecode, i looked back at the sample files i was
using, and found the fault. so, hope i didn't waste anyone else's time
too...

-ben

On Tue, 21 Dec 2004, Benjamin E. Taitelbaum wrote:

> Have any of you been able to get the sky example working with pw instead
> of glut? When I use pw, I'm not able to get the colors working. I can post
> examples, but it's probably easier just to make the changes to the example
> source file.
>
> thanks,
> ben
>
>
>
> -------------------------------------------------------
> SF email is sponsored by - The IT Product Guide
> Read honest & candid reviews on hundreds of IT Products from real users.
> Discover which products truly live up to the hype. Start reading now.
> http://productguide.itmanagersjournal.com/
> _______________________________________________
> plib-users mailing list
> pli...@li...
> https://lists.sourceforge.net/lists/listinfo/plib-users
>

[Plib-users] Sky and Pw

[Benjamin E. T. <be...@cs...>]

Have any of you been able to get the sky example working with pw instead
of glut? When I use pw, I'm not able to get the colors working. I can post
examples, but it's probably easier just to make the changes to the example
source file.

thanks,
ben

Re: [Plib-users] Graphic driver corruption in PUI Menus ?

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

Richard Andrews wrote:

>>p.s. thanks for helping me make up my mind between nvidia and ati ;)
> 
> I've heard people complaining about NVidia cards enough to make me think twice 
> about leaping into bed with NVidia.

Where I work, we stress graphics cards and their drivers pretty severely - and
there is no shadow of a doubt that nVidia's drivers are head-and-shoulders better
than ATI's.  Not without *any* problems - just generally much better.

ATI's hardware is sometimes faster than nVidia - sometimes slower - it depends in
minute detail on what exactly you are doing.

> Had I known earlier I may well have gone 
> for an NVidia card (the linux favourite). But, I know the ATI now and on the 
> whole have been quite happy with it. One day I might lash out and buy an 
> GeForce of whatever but not yet. Besides ATI is a brain expanding 
> challenge ;-)

This problem you are seeing looks similar to something we saw *years* ago
with the Mesa drivers.  The texture is somehow being stored incorrectly.

Look closely inside the button in the bottom-left corner of the screen.
It's supposed to say "Start Game" - with each letter being a separate
polygon.

You can actually see that the letters 't' and 'a' are coming up with
the same pattern in the two places that each letter is drawn.

Could you run the PLIB FNT library test program and get us a snapshot
of that?

---------------------------- Steve Baker -------------------------
HomeEmail: <sjb...@ai...>    WorkEmail: <sj...@li...>
HomePage : http://www.sjbaker.org
Projects : http://plib.sf.net    http://tuxaqfh.sf.net
            http://tuxkart.sf.net http://prettypoly.sf.net
-----BEGIN GEEK CODE BLOCK-----
GCS d-- s:+ a+ C++++$ UL+++$ P--- L++++$ E--- W+++ N o+ K? w--- !O M-
V-- PS++ PE- Y-- PGP-- t+ 5 X R+++ tv b++ DI++ D G+ e++ h--(-) r+++ y++++
-----END GEEK CODE BLOCK-----

Re: [Plib-users] Graphic driver corruption in PUI Menus ?

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

Benjamin E. Taitelbaum wrote:

> p.s. thanks for helping me make up my mind between nvidia and ati ;)

That's really no contest for Linux users.  ATI's drivers have a long
history of sucking in one way or another.  If you have a choice, nVidia
is a better bet.

Having said that though, ATI's problems are generally a lot more subtle
than this.  Anything that would stop the PLIB font library from drawing
text correctly would be a pretty major problem for almost anything.
FNT doesn't use any clever OpenGL trickery at all.  It's all just a
bunch of glBegin/glTexCoord/glVertex/glEnd blocks.

---------------------------- Steve Baker -------------------------
HomeEmail: <sjb...@ai...>    WorkEmail: <sj...@li...>
HomePage : http://www.sjbaker.org
Projects : http://plib.sf.net    http://tuxaqfh.sf.net
            http://tuxkart.sf.net http://prettypoly.sf.net
-----BEGIN GEEK CODE BLOCK-----
GCS d-- s:+ a+ C++++$ UL+++$ P--- L++++$ E--- W+++ N o+ K? w--- !O M-
V-- PS++ PE- Y-- PGP-- t+ 5 X R+++ tv b++ DI++ D G+ e++ h--(-) r+++ y++++
-----END GEEK CODE BLOCK-----

Re: [Plib-users] Graphic driver corruption in PUI Menus ?

[Benjamin E. T. <be...@cs...>]

> System:
> Intel 3,06
> 1024 MB RAM
> Suse 9.1 Pro
> Hercules 3D Prophet (Radeon 9700 Pro Chipset) 128 MB
> ATI drivers (from SuSe supplementary) 4.14.06

I'm assuming you mean the 3.14.06 drivers? or does SuSe do some weird
renumbering?

Try running fglxgears (it may just be called glxgears) and see if it
works. if it doesn't, then it's not plib at all (which is what a lot of us
seem to expect).

* the ati drivers are built for specific versions of X, however they don't
have separate numbers -- this can be really confusing... my advice would
be to try installing the drivers by hand:

1) run X --config  to see your version
2) go to http://www.ati.com/support/driver.html and download the correct
linux driver.
3) install the driver as per the instructions on the site. this doesn't
look too easy, and i've never tried it, so good luck...

If you're using xorg instead of xfree, i believe the version for XFree86
4.3.0 will work with xorg 6.7.0 (but not 6.8.0).


good luck,
ben

p.s. thanks for helping me make up my mind between nvidia and ati ;)

Re: [Plib-users] Graphic driver corruption in PUI Menus ?

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

Richard Andrews wrote:
> Hi,
> 
> Having just updated my ATI Radeon drivers to 4.14.06 I have found problems 
> with menus on apps that use the plib libraries. For example Tuxkart and 
> Flightgear both show garbled unreadable text on the menubar as well as drop 
> downs.

So it sounds like textures are somehow being screwed up.  It's hard to see
why that would be - maybe you could email me a screen-shot?

Do other (non-PLIB) OpenGL applications work correctly?

---------------------------- Steve Baker -------------------------
HomeEmail: <sjb...@ai...>    WorkEmail: <sj...@li...>
HomePage : http://www.sjbaker.org
Projects : http://plib.sf.net    http://tuxaqfh.sf.net
            http://tuxkart.sf.net http://prettypoly.sf.net
-----BEGIN GEEK CODE BLOCK-----
GCS d-- s:+ a+ C++++$ UL+++$ P--- L++++$ E--- W+++ N o+ K? w--- !O M-
V-- PS++ PE- Y-- PGP-- t+ 5 X R+++ tv b++ DI++ D G+ e++ h--(-) r+++ y++++
-----END GEEK CODE BLOCK-----

Re: [Plib-users] Graphic driver corruption in PUI Menus ?

[Richard A. <ric...@ot...>]

On Sunday 12 December 2004 15:42, Arnt Karlsen wrote:
> ..chk the X docs for how to give the chip set or whatever-it-was
> ID in your x(|XF86C|)config(-4) or commandline.
Hmmm.. I think I will leave that until I get a weekend where I am not trying 
to get an assignment finished. Choosing between failing a Uni module and 
getting fgfs running is tricky but I think I have to go for the assignment 
this time ;-)

> to 1xAGP, I see 4 to 5 fps in FlightGear without a scenery, 
Ouch. I get reasonable framerates but the scenery flickers like mad and of 
course the menus look like they're written in klingon.

>bad scenery path put me _on_ da big ass drink.  ;-)
Sounds painful :-)

I'll let you know if I have any luck when I finally get a chance to tinker. In 
the meantime see you on the Flightgear lists Arnt!

Cheers

Rich

Re: [Plib-users] Graphic driver corruption in PUI Menus ?

[Arnt K. <ar...@c2...>]

On Sun, 12 Dec 2004 13:45:01 +0100, Richard wrote in message 
<200...@ot...>:

> Hej Amt,
> 
> On Sunday 12 December 2004 12:53, Arnt Karlsen wrote:
> > ..what happens if you lie to your box and tell it "you're an 9200
> > pro " and use DRI?
> The card is recognised by the system automagically and the XF86Config
> has it entered as a "ATI Graphics Adapter" using fglrx and DRI. There
> doesn't seem to be an entry for overriding the actual card used i.e.
> seems to be a generic driver for a wide range of ATI cards.
> 
> If there is a way to 'simply' override the device found, I could give
> it a go. Only worry I have is that by changing anything like that I
> will make the system unstable etc. Still reasonably new at all this...

..chk the X docs for how to give the chip set or whatever-it-was 
ID in your x(|XF86C|)config(-4) or commandline.

..playing with my new shiny 9250, googling adviced me it would be seen
as a 9200 pro if it could't do 8xAGP, on an 5 year old AMD k6-2 450MHz
on a Epox MVP3G2 with 384MB ram, which can be pushed all the way up 
to 1xAGP, I see 4 to 5 fps in FlightGear without a scenery, bad scenery
path put me _on_ da big ass drink.  ;-)   

..the google advice was "specify your chip to get 8xAGP."  ;-)

..in the default KSFO scenery I get 1 to 2 fps.  In Armagetron at the
same window size (1280x1024) , I see 40-54 fps with a default setup. 

..I will try more of these games to learn more on its performance.

..the 9250 is a 128 bit unit with 128MB, the standard 15pin vga, dvi 
and tv outputs, and is built by Sapphire.

-- 
..med vennlig hilsen = with Kind Regards from Arnt... ;-)
...with a number of polar bear hunters in his ancestry...
  Scenarios always come in sets of three: 
  best case, worst case, and just in case.

Re: [Plib-users] Graphic driver corruption in PUI Menus ?

[Richard A. <ric...@ot...>]

Hej Amt,

On Sunday 12 December 2004 12:53, Arnt Karlsen wrote:
> ..what happens if you lie to your box and tell it "you're an 9200 pro "
> and use DRI?
The card is recognised by the system automagically and the XF86Config has it 
entered as a "ATI Graphics Adapter" using fglrx and DRI. There doesn't seem 
to be an entry for overriding the actual card used i.e. seems to be a generic 
driver for a wide range of ATI cards.

If there is a way to 'simply' override the device found, I could give it a go. 
Only worry I have is that by changing anything like that I will make the 
system unstable etc. Still reasonably new at all this...

Thanks 

Rich

Re: [Plib-users] Graphic driver corruption in PUI Menus ?

[Arnt K. <ar...@c2...>]

On Sun, 12 Dec 2004 10:33:56 +0100, Richard wrote in message 
<200...@ot...>:

> Hi,
> 
> Having just updated my ATI Radeon drivers to 4.14.06 I have found
> problems with menus on apps that use the plib libraries. For example
> Tuxkart and Flightgear both show garbled unreadable text on the
> menubar as well as drop downs. 
> 
> The problem was not there before the updated drivers were installed so
> I do suspect that there lies the source of the problem. I have also
> upgraded plib to 1.83 (from SuSe's 1.6) but with unfortunately no
> effect.
> 
> Is anyone else experiencing this or is there anything I can try to
> eliminate the problem? The loss of flightgear is sorely felt.
> 
> System:
> Intel 3,06
> 1024 MB RAM
> Suse 9.1 Pro 
> Hercules 3D Prophet (Radeon 9700 Pro Chipset) 128 MB 
> ATI drivers (from SuSe supplementary) 4.14.06
> 
> Using XF86Config-4 configured with fglrxconfig and not an XF86Config
> as generated by Sax2.

..what happens if you lie to your box and tell it "you're an 9200 pro "
and use DRI? 

-- 
..med vennlig hilsen = with Kind Regards from Arnt... ;-)
...with a number of polar bear hunters in his ancestry...
  Scenarios always come in sets of three: 
  best case, worst case, and just in case.

[Plib-users] Graphic driver corruption in PUI Menus ?

[Richard A. <ric...@ot...>]

Hi,

Having just updated my ATI Radeon drivers to 4.14.06 I have found problems 
with menus on apps that use the plib libraries. For example Tuxkart and 
Flightgear both show garbled unreadable text on the menubar as well as drop 
downs. 

The problem was not there before the updated drivers were installed so I do 
suspect that there lies the source of the problem. I have also upgraded plib 
to 1.83 (from SuSe's 1.6) but with unfortunately no effect.

Is anyone else experiencing this or is there anything I can try to eliminate 
the problem? The loss of flightgear is sorely felt.

System:
Intel 3,06
1024 MB RAM
Suse 9.1 Pro 
Hercules 3D Prophet (Radeon 9700 Pro Chipset) 128 MB 
ATI drivers (from SuSe supplementary) 4.14.06

Using XF86Config-4 configured with fglrxconfig and not an XF86Config as 
generated by Sax2.

Thanks for any help

Richard Andrews

[Plib-users] cross-compiling with mingw32

[Benjamin E. T. <be...@cs...>]

hi,

I've just started working with plib, and thought others might find this
useful:

To cross compile a windows plib binary from linux:

1) Install the mingw32 (also called xmingw32) tools. These can be emerged,
apt-gotten, yummed, or whatever your favorite way of installing is.

2) compile plib using the mingw32-g++ compiler. The following should work:
bash$ ./configure  --build=i686-linux --target=i386-mingw32 --host=i386-mingw32 --prefix=/usr/local/i386-mingw32

(where prefix is the location of mingw. it might be /opt/xmingw for example)

3) I've been using the following makefile, and it has worked fine so far,
although I might have left something out, since I haven't tested all the
plib functions. You'll have to change it slightly to fit your system and
project/object names, but this should serve as a good starting point.


best,
ben


## Makefile
## ========

# replace with the object files for your project
OBJS=ui.o gfx.o util.o

## i might have missed some, or there might be too many,
## depending on your project
PLIBS=  -lplibpu  -lplibpw \
	-lplibsg -lplibssg \
	-lplibfnt -lplibul \
	-lplibssgaux

## linux (maybe windows or cygwin?)
## make sure the cross-compiling section is commented out
CC=g++
FLAGS= -g
INCS=
LIBS= -lGL -lGLU -lm
OUT_EXT=

## cross compiling from linux
## uncomment this section to cross-compile
## no need to comment the previous section since these
## variables will overwrite the former ones
##
## change the following variables to fit your system. The
## uncommented ones are for fc2,
## and the commented ones are for gentoo
MINGW_BIN=/usr/local/bin#/opt/xmingw/bin
MINGW_BASE=/usr/local/i386-mingw32#/opt/xmingw/i386-mingw32msvc
CC=${MINGW_BIN}/i386-mingw32-g++#${MINGW_BIN}/i386-mingw32msvc-g++
FLAGS=-DWIN32 -g
INCS=-I${MINGW_BASE}/include
LIBS=-L${MINGW_BASE}/lib -lopengl32 -lm -lgdi32 -lwinmm
PLIBS:=-L${MINGW_BASE}/lib ${PLIBS}
OUT_EXT=.exe

CarToon: ${OBJS}
	${CC} ${FLAGS}  ${OBJS} \
	${PLIBS} ${LIBS} cartoon.cc \
	-o CarToon${OUT_EXT}

%.o: %.cc %.h
	${CC} ${FLAGS} ${INCS} -c $<

clean:
	rm *.o ; rm *_test ; rm *.exe ; rm *~ ; rm CarToon

## end of Makefile

p.s. sorry if i'm posting this twice. i didn't see it go through the first
time...

[Plib-users] Auto-calibration working in windows, not in Linux

[Julien M. <dp...@ya...>]

Hello,

I'm playing the game Torcs, and i've two racing wheels (1 Logitech Formula
Force GP, and one cheap APM wheel). I've exactly the same problem with
both.

In windows they're working correctly (probably using the calibration
values from the control panel?), in Linux often the movement isn't
detected at all.

I'm using the CVS version of Torcs, and the 1.8.3 of PLIB. I checked with
js_demo, it always returns 0 for the axis (buttons are OK). so i modified
the js_demo source to replace the read()'s by rawRead's, i get values like
30 - 400 - 900. Something strange is that those values are different from
those i get with jscalibrator (it uses the libjsw library IIRC but i
thought the values displayed were also those given verbatim by the OS?)

I then modified it putting back the read's, and at the beginning i've put
setMin/MaxRange and setCenter and hardcoded the values for my wheel. This
way i got correct values -1,0,1 with read.

In windows there is a centralized standard location for the calibration
values of the control panel (registry?), but i think there is no such
thing in Linux.

Is there a plan to use something similar in Linux? (jscalibrator writes a
~/.joystick, or maybe we can use a system-wide file like /etc/joystick?)

Regards,

D.plomat


	

	
		
Découvrez le nouveau Yahoo! Mail : 250 Mo d'espace de stockage pour vos mails ! 
Créez votre Yahoo! Mail sur http://fr.mail.yahoo.com/ 
 
Avec Yahoo! faites un don et soutenez le Téléthon en cliquant sur http://www.telethon.fr/030-Don/10-10_Don.asp