openvrml-develop — OpenVRML developers mailing list

Re: [openvrml-develop] Does OpenVrml ......

[Leopold P. <le...@wo...>]

Hi,

thank's for the answer. You have mentioned that you have tested lookat and 
Xj3D and lookat wins performance wise.

Here, there are some people that proposing to develop this project in java. 
It will win portability, so we could use it in Irix, Solaris, Linux and 
Windoze. But, I'm a bit afraid, because I think that by now, java has a lot 
of limitations about the performance. So, if I would like to make a simulator 
of a robotic arm, with a lot of calculations (paths, kinematic inverse, etc) 
and after, I have to do all the work of representing it using a (more or 
less) realistic scene, java maybe is not an option, and I need to go 
necessary to C++/openvrml. It will be great, if someone have 

Do you have tested some complex scene in java and in lookat? there's a lot of 
difference?

Best regards,

Leo


A Dijous 07 Març 2002 23:28, Nigel Stewart va escriure:
> Hi,
>
> You may also like to try Xj3D, the Java-based VRML and X3D
> open source browser from the Web3D group.  I'm using both
> lookat and Xj3D at the moment - lookat wins performance wise.
> I'm doing a simple robot using OrientationInterpolators
> routed to Transform nodes in the robot kinematic heirachy,
> works nicely apart from the well known ROUTE-TO-PROTO bug
> in OpenVRML.  (Therefore not using PROTO's)  I will probably
> be setting this as an assignment for our Web3D students.
>
> You may be interested in this link:
> http://goanna.cs.rmit.edu.au/~nigels/Web3D/
>
> Cheers,
>
> Nigel
>
> > Hi,
> >
> > thank's a lot for the mail. I have read it with with a lot of care.
> >
> > I'm using linux, and I would like to do a Open source program. So, any
> > propietary or Windows program it's no a way to me.
> >
> > I have looked the link of Sandy Ressler, and I think that it's very
> > interesting to me. One problem that I found always with lookat it's that
> > I cannot view a lot of vrml files. For example some of the Sandy
> > Ressler's webs.
> >
> > Otherwise, I begin to rethink my master thesis, and maybe I use a java3d
> > to view the scenes. I don't know.

Re: [openvrml-develop] Does OpenVrml ......

[Nigel S. <ni...@ni...>]

Hi,

You may also like to try Xj3D, the Java-based VRML and X3D
open source browser from the Web3D group.  I'm using both 
lookat and Xj3D at the moment - lookat wins performance wise.
I'm doing a simple robot using OrientationInterpolators
routed to Transform nodes in the robot kinematic heirachy,
works nicely apart from the well known ROUTE-TO-PROTO bug
in OpenVRML.  (Therefore not using PROTO's)  I will probably
be setting this as an assignment for our Web3D students.

You may be interested in this link:
http://goanna.cs.rmit.edu.au/~nigels/Web3D/

Cheers,

Nigel

> Hi,
> 
> thank's a lot for the mail. I have read it with with a lot of care.
> 
> I'm using linux, and I would like to do a Open source program. So, any
> propietary or Windows program it's no a way to me.
> 
> I have looked the link of Sandy Ressler, and I think that it's very
> interesting to me. One problem that I found always with lookat it's that I
> cannot view a lot of vrml files. For example some of the Sandy Ressler's webs.
> 
> Otherwise, I begin to rethink my master thesis, and maybe I use a java3d to
> view the scenes. I don't know.

--
                     Nigel Stewart (ni...@ni...)    
                     http://www.nigels.com/

Re: [openvrml-develop] openvrml on aix

[Braden M. <br...@en...>]

On Wed, 2002-03-06 at 11:30, Chr...@GE... wrote:
> hi
> 
> i have a problem compiling openvrml 0.11.2 on aix. i tried several
> compilers and ./configure runs through without any "problems". then i try
> to make the source. but everytime it breaks on compiling CharScanner.cpp
> with the following error message.
> 
> <schnipp ----------------------------------------------------------
> 
> c++ -DHAVE_CONFIG_H -I. -I. -I../.. -g -O2 -c CharScanner.cpp  -DPIC -o
> .libs/CharScanner.lo
> CharScanner.cpp: In method `bool
> antlr::CharScannerLiteralsLess::operator ()(const string &, const string
> &) const':
> CharScanner.cpp:91: implicit declaration of function `int
> antlr::stricmp(...)'
> make: *** [CharScanner.lo] Error 1
> 
> ----------------------------------------------------------------------schnapp>
> 
> i dont have experience with c++ or c anyway, so maybe someone can help me
> with aix as platform. or whats really needed by openvrml to compile.

Try adding "-DANTLR_REALLY_NO_STRCASECMP" to CPPFLAGS when you
configure.

-- 
Braden McDaniel                           e-mail: <br...@en...>
<http://endoframe.com>                    Jabber: <br...@ja...>

[openvrml-develop] openvrml on aix

[<Chr...@GE...>]

hi

i have a problem compiling openvrml 0.11.2 on aix. i tried several
compilers and ./configure runs through without any "problems". then i try
to make the source. but everytime it breaks on compiling CharScanner.cpp
with the following error message.

<schnipp ----------------------------------------------------------

c++ -DHAVE_CONFIG_H -I. -I. -I../.. -g -O2 -c CharScanner.cpp  -DPIC -o
.libs/CharScanner.lo
CharScanner.cpp: In method `bool
antlr::CharScannerLiteralsLess::operator ()(const string &, const string
&) const':
CharScanner.cpp:91: implicit declaration of function `int
antlr::stricmp(...)'
make: *** [CharScanner.lo] Error 1

----------------------------------------------------------------------schnapp>

i dont have experience with c++ or c anyway, so maybe someone can help me
with aix as platform. or whats really needed by openvrml to compile.

fya


Christian Setzer
GE CompuNet Muenchen
Enterprise Computing Solutions
Hoerselbergstrasse 7, 81677 Muenchen, Germany
Phone: +49 (0)89 / 382-47742, , Mobile: +49 (0) 171 1284997
E-Mail : Chr...@GE...
Visit us on the Internet: http://www.gecits-eu.com


This email is confidential. If you are not the intended recipient,
you must not disclose or use the information contained in it.
If you have received this mail in error, please tell us
immediately by return email and delete the document.

Re: [openvrml-develop] Does OpenVrml ......

[Leopold P. A. <le...@wo...>]

Hi,

thank's a lot for the mail. I have read it with with a lot of care.

I'm using linux, and I would like to do a Open source program. So, any=20
propietary or Windows program it's no a way to me.

I have looked the link of Sandy Ressler, and I think that it's very=20
interesting to me. One problem that I found always with lookat it's that =
I=20
cannot view a lot of vrml files. For example some of the Sandy Ressler's =
webs.

Otherwise, I begin to rethink my master thesis, and maybe I use a java3d =
to=20
view the scenes. I don't know.

Than's ,

best regards,

Leo


A Divendres 01 Mar=E7 2002 18:13, v=E0reu escriure:
> Hello leopold,
>
> Getting the candidate node to modify will in my opinion be a lot easier=
 if
> you name the node something anthropomorphic. Better yet, something
> anthropomorphic that conforms to the H-anim specification.
>
> So, you really need a 3-D modeler that exports to VRML using the DEF
> keyword. I believe Lightwave 3D does that. Carrara and Amapi do not. It=
 may
> not be a problem if you have a small robotic system with on a dozen to =
a
> hundred nodes. You can hand edit those systems. You can also get Spazz3=
D
> (VRML authoring tool) and rename the nodes. Then re-export to VRML.
>
> Oops, I suppose that I should verify that openvrml supports the DEF
> keyword. I'm being lazy today. Anyone want to answer yes or no to the D=
EF
> quiesion?
>
> This is a sort of standards type problem. Solved for Avatars in H-anim =
spec
> but for general scenes not solved. That means that everyone has to get
> together and agree on naming for a wide variety of core objects like
> architecture, communications devices, vehicles, aircraft, computer
> equipment, tools,.....
>
> Then the artists and the modeling systems know how to export using stan=
dard
> names for the DEF keyword.
>
> This does not solve your problem of course. However, since Bradon's pos=
t
> seemed to indicate a "doable" solution I will solve the problem for the
> Macintosh. This will be on my agenda for Siggraph and probably Macworld=
 NY.
> However, if you are on Linux  it may not help you.
>
> So, you might want to contact Sandy Ressler. He made a Robolab interfac=
e to
> Blaxxun I believe using the EAI to at least change the viewpoint. Of co=
urse
> this is not our favorite open source solution...:)
>
>  http://ovrt.nist.gov/papers.html is the link to Sandy's paper.
>
> John F. Richardson
>
> >Hi,
> >
> >first off all I have to say that I'm not senior programmer, I'm just
> >beginning.
> >
> >I'm doing my master thesis, and I'm doing a robot simulator. I don't k=
now
> > if my questions are similar than John Richardson have made before, bu=
t
> > maybe.
> >
> >I would like to use OpenVrml to show the robots movements, just all. M=
y
> >question is how I have to do it?, or in which way I have to work.
> >
> >For example, I have a vrml file with a robot arm with their links. I c=
an
> > make vrml scene and show the file, ok. In my simulator, for example, =
I
> > can calculate a path in the space and I have the join values of every
> > link of the arm in the time.
> > time 0s join values of the links =3D[0,10,15,20,15,20]
> > time 0.1s join values =3D[0,12,16,22,18]
> >  time 0,2s join values .....
> >
> >So, it's possible to change the position of every link in the scene in
> > this way to simulate a movement? or really I have a big confusion of =
the
> > openvrml libs, and I need another kind of libs?
> >
> >Leo
> >
> >Pd I'm sorry for my horrible English....
> >
> >--
> >Leopold Palomo Avellaneda
> >
> >Linux User 152692
> >Catalonia

--=20
Leopold Palomo Avellaneda

Linux User 152692=20
Catalonia

[openvrml-develop] Rearchitecture - initial checkin

[Braden M. <br...@en...>]

I have checked in the rearchitecture code I've been working on. The
branch tag is "OpenVRML-REARCH_BRANCH". Much work still needs to be done
just to get it Basically Working:

        * Changes are not reflected in the Java binding code. The
          rearchitected code should compile, but only --without-jdk.
        * Though the code will compile, it will not link. There remain a
          number of missing method definitions.
        * In addition to the missing method definitions, some methods
          are defined, but their implementation needs to be filled in.
          These are marked with Doxygen @todo tags, so they should be
          readily visible in the generated documentation.
        * Code to support PROTOs is completely disabled in this branch.
          Once the code that is there now is brought to a working state,
          PROTO support needs to be redone.

Though this code is very unfinished, the nature of the rearchitecture
should be clear. I encourage OpenVRML users to examine the code to get a
feel for the direction. Feel free to comment and ask questions.

In order to avoid a Merge Nightmare when this code is finally ready to
merge back into the main branch, I'd like any code checked in on the
trunk also to be checked in on the rearchitecture branch. Please
coordinate with me to do this if you do not want to do it yourself.

If anyone is interested in helping me out on the rearchitecture branch,
I could use a hand. Please mail me.

-- 
Braden McDaniel                           e-mail: <br...@en...>
<http://endoframe.com>                    Jabber: <br...@ja...>

Re: [openvrml-develop] Grants for openvrml development

[Braden M. <br...@en...>]

On Fri, 2002-03-01 at 12:20, John Richardson wrote:
> Hello,
> 
> I assume that everyone on the list really likes to produce open source 24
> hours a day. However, some "day jobs" and other social activities similar
> to laundry and comic book reading may interfere.
> 
> I was wondering if it would be advisable for the group to start a thread
> related to gathering information and applying for real grants to
> specifically enhance openvrml, generate virtual environments, produce
> prodigious test viewers,......
> 
> It would help if the Grants contained humongous amounts of dollars, euros,
> pounds, yen,......

A very interesting idea. I'm not sure where to start, though. If you or
someone else has leads on RFPs that could be addressed by extending
OpenVRML, that could be an exciting opportunity.

-- 
Braden McDaniel                           e-mail: <br...@en...>
<http://endoframe.com>                    Jabber: <br...@ja...>

Re: [openvrml-develop] Re: OpenVRML

[Thomas F. <tf...@us...>]

The required libs for Win32 can be found at 
http://openvrml.org/dist/win32_required_libs.zip

This zipfile includes js32.dll and js32.lib which should be all you need 
to get the js stuff running on Windows.

Later,
Tom

Christopher M Morley wrote:

>Hi John, I'm afraid I don't have a working version of openvrml with js either
>at the moment, but I am sure some one on the the developers list can help.
>
>There is no limit to the number of nodes, assuming you have
>enough memory.
>
>Chris
>
>----- Original Message -----
>From: "John Wallis" <jw...@ix...>
>To: <cm...@ve...>
>Sent: Friday, March 01, 2002 1:09 AM
>Subject: OpenVRML
>
>
>Hello Mr Morley,
>
>1.
>I'm interested to see what Open VRML can do, but
>I just can't seem to get it to work.
>
>It is virtually impossible to find a js32.dll that compatible.
>And the link that OpenVRML suggests are all dead.
>
>Can you help me, perhaps send me a js32.dll per attachment?
>
>2.
>Is there an upper limit on how many nodes a VRML world can have?
>
>Thank You
>  John Wallis
>
>
>***********************************************
>* John Wallis                                 *
>*                                             *
>* Institut für Angewandte Mathematik          *
>* Universität Heidelberg                      *
>* Im Neuenheimer Feld 294                     *
>* 69120 Heidelberg                            *
>*                                             *
>* Tel. : +49 6221 54 4986                     *
>* Tel. : +49 6221 54 5043                     *
>* Email: jw...@ix...     *
>***********************************************
>
>

[openvrml-develop] Re: OpenVRML

[Christopher M M. <cm...@ve...>]

Hi John, I'm afraid I don't have a working version of openvrml with js ei=
ther
at the moment, but I am sure some one on the the developers list can help.

There is no limit to the number of nodes, assuming you have
enough memory.

Chris

----- Original Message -----
From: "John Wallis" <jw...@ix...>
To: <cm...@ve...>
Sent: Friday, March 01, 2002 1:09 AM
Subject: OpenVRML


Hello Mr Morley,

1.
I'm interested to see what Open VRML can do, but
I just can't seem to get it to work.

It is virtually impossible to find a js32.dll that compatible.
And the link that OpenVRML suggests are all dead.

Can you help me, perhaps send me a js32.dll per attachment?

2.
Is there an upper limit on how many nodes a VRML world can have?

Thank You
  John Wallis


***********************************************
* John Wallis                                 *
*                                             *
* Institut f=FCr Angewandte Mathematik          *
* Universit=E4t Heidelberg                      *
* Im Neuenheimer Feld 294                     *
* 69120 Heidelberg                            *
*                                             *
* Tel. : +49 6221 54 4986                     *
* Tel. : +49 6221 54 5043                     *
* Email: jw...@ix...     *
***********************************************

[openvrml-develop] Grants for openvrml development

[John R. <ric...@sp...>]

Hello,

I assume that everyone on the list really likes to produce open source 24
hours a day. However, some "day jobs" and other social activities similar
to laundry and comic book reading may interfere.

I was wondering if it would be advisable for the group to start a thread
related to gathering information and applying for real grants to
specifically enhance openvrml, generate virtual environments, produce
prodigious test viewers,......

It would help if the Grants contained humongous amounts of dollars, euros,
pounds, yen,......

John F. Richardson

Re: [openvrml-develop] Does OpenVrml ......

[John R. <ric...@sp...>]

Hello leopold,

Getting the candidate node to modify will in my opinion be a lot easier if
you name the node something anthropomorphic. Better yet, something
anthropomorphic that conforms to the H-anim specification.

So, you really need a 3-D modeler that exports to VRML using the DEF
keyword. I believe Lightwave 3D does that. Carrara and Amapi do not. It may
not be a problem if you have a small robotic system with on a dozen to a
hundred nodes. You can hand edit those systems. You can also get Spazz3D
(VRML authoring tool) and rename the nodes. Then re-export to VRML.

Oops, I suppose that I should verify that openvrml supports the DEF
keyword. I'm being lazy today. Anyone want to answer yes or no to the DEF
quiesion?

This is a sort of standards type problem. Solved for Avatars in H-anim spec
but for general scenes not solved. That means that everyone has to get
together and agree on naming for a wide variety of core objects like
architecture, communications devices, vehicles, aircraft, computer
equipment, tools,.....

Then the artists and the modeling systems know how to export using standard
names for the DEF keyword.

This does not solve your problem of course. However, since Bradon's post
seemed to indicate a "doable" solution I will solve the problem for the
Macintosh. This will be on my agenda for Siggraph and probably Macworld NY.
However, if you are on Linux  it may not help you.

So, you might want to contact Sandy Ressler. He made a Robolab interface to
Blaxxun I believe using the EAI to at least change the viewpoint. Of course
this is not our favorite open source solution...:)

 http://ovrt.nist.gov/papers.html is the link to Sandy's paper.

John F. Richardson

>Hi,
>
>first off all I have to say that I'm not senior programmer, I'm just
>beginning.
>
>I'm doing my master thesis, and I'm doing a robot simulator. I don't know if
>my questions are similar than John Richardson have made before, but maybe.
>
>I would like to use OpenVrml to show the robots movements, just all. My
>question is how I have to do it?, or in which way I have to work.
>
>For example, I have a vrml file with a robot arm with their links. I can make
>vrml scene and show the file, ok. In my simulator, for example, I can
>calculate a path in the space and I have the join values of every link of the
>arm in the time.
> time 0s join values of the links =[0,10,15,20,15,20]
> time 0.1s join values =[0,12,16,22,18]
>  time 0,2s join values .....
>
>So, it's possible to change the position of every link in the scene in this
>way to simulate a movement? or really I have a big confusion of the openvrml
>libs, and I need another kind of libs?
>
>Leo
>
>Pd I'm sorry for my horrible English....
>
>--
>Leopold Palomo Avellaneda
>
>Linux User 152692
>Catalonia

[openvrml-develop] Does OpenVrml ......

[Leopold P. A. <le...@wo...>]

Hi,

first off all I have to say that I'm not senior programmer, I'm just
beginning.

I'm doing my master thesis, and I'm doing a robot simulator. I don't know if
my questions are similar than John Richardson have made before, but maybe.

I would like to use OpenVrml to show the robots movements, just all. My
question is how I have to do it?, or in which way I have to work.

For example, I have a vrml file with a robot arm with their links. I can make
vrml scene and show the file, ok. In my simulator, for example, I can
calculate a path in the space and I have the join values of every link of the
arm in the time.
 time 0s join values of the links =[0,10,15,20,15,20]
 time 0.1s join values =[0,12,16,22,18]
  time 0,2s join values .....

So, it's possible to change the position of every link in the scene in this
way to simulate a movement? or really I have a big confusion of the openvrml
libs, and I need another kind of libs?

Leo

Pd I'm sorry for my horrible English....

--
Leopold Palomo Avellaneda

Linux User 152692
Catalonia

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

-- 
Leopold Palomo Avellaneda

Linux User 152692 
Catalonia

Re: [openvrml-develop] libmozjs ...

[Braden M. <br...@en...>]

On Thu, 2002-02-28 at 12:25, Leopold Palomo Avellaneda wrote:
> Hi list,
> 
> I would like to compile openvrml with libmozjs support, and I'm not be able 
> to do it. I'm using mandrake 8.1 and I have installed in /usr/lib:
> 
> -rw-r--r--    1 root     root       567616 feb 22 12:23 /usr/lib/libjs.a
> -rwxr-xr-x    1 root     root       107456 oct  1 20:54 /usr/lib/libjsj.so*
>  lrwxrwxrwx    1 root     root           10 feb 25 09:32 /usr/lib/libjs.so -> 
> libjs.so.0
> lrwxrwxrwx    1 root     root           14 feb 22 12:49 /usr/lib/libjs.so.0 
> -> libjs.so.0.1.0
> -rw-r--r--    1 root     root       445700 feb 22 12:23 
> /usr/lib/libjs.so.0.1.0
> -rwxr-xr-x    1 root     root       551832 oct  1 20:54 /usr/lib/libmozjs.so*
> 
> I have tried to run configure with:
> --with-mozjs=/usr/lib  or
> --with-mozjs=/usr/ or
> 
> 
> and always I have the same error that didn't found the libmozjs.
> 
> I don't what to do to compile with javascript support. 
> Please, can anyone tell me what to do?

The autoconf test checks for a header file as well as the library. Check
the contents of your config.log to see exactly how the test is failing.

-- 
Braden McDaniel                           e-mail: <br...@en...>
<http://endoframe.com>                    Jabber: <br...@ja...>

Re: [openvrml-develop] Does OpenVRML 0.11.x allow for adding/modifying fields

[Braden M. <br...@en...>]

On Wed, 2002-02-27 at 15:06, John Richardson wrote:
> Hello,
> 
> Does OpenVRML 0.11.x allow for adding/modifying fields?

Not truly generically, but for the most part yes.

This will be improving dramatically.

> I want to run a C++
> or Java or FORTRAN simulation and then based upon the results of some
> algorithm executing in the simulation direct OpenVRML to change a nodes
> position, add or subtract a node and change the value of a nodes field.

Probably doable.

> I presume that the easiest way would be to subclass ViewerOpenGL and then
> modify the scene graph. Sort of like a direct hook to OpenVRML rather than
> through something like the EAI.

Yes.

> So, I also suppose that I should ask if OpenVRML 0.11.x supports the EAI?

No.

> What about the Script node?

We're pretty good in that department. There are holes: in JavaScript,
VrmlMatrix is missing and directOutput isn't properly supported. Tom
Flynn has been working on Java support for our next release.

> I suppose that there would have to be some method getting the target node
> for modification. Any ideas?

You can get the root nodes from VrmlScene and basically walk the scene
graph until you find what you need.

-- 
Braden McDaniel                           e-mail: <br...@en...>
<http://endoframe.com>                    Jabber: <br...@ja...>

[openvrml-develop] libmozjs ...

[Leopold P. A. <le...@wo...>]

Hi list,

I would like to compile openvrml with libmozjs support, and I'm not be ab=
le=20
to do it. I'm using mandrake 8.1 and I have installed in /usr/lib:

-rw-r--r-- =A0 =A01 root =A0 =A0 root =A0 =A0 =A0 567616 feb 22 12:23 /us=
r/lib/libjs.a
-rwxr-xr-x =A0 =A01 root =A0 =A0 root =A0 =A0 =A0 107456 oct =A01 20:54 /=
usr/lib/libjsj.so*
=A0lrwxrwxrwx =A0 =A01 root =A0 =A0 root =A0 =A0 =A0 =A0 =A0 10 feb 25 09=
:32 /usr/lib/libjs.so ->=20
libjs.so.0
lrwxrwxrwx =A0 =A01 root =A0 =A0 root =A0 =A0 =A0 =A0 =A0 14 feb 22 12:49=
 /usr/lib/libjs.so.0=20
-> libjs.so.0.1.0
-rw-r--r-- =A0 =A01 root =A0 =A0 root =A0 =A0 =A0 445700 feb 22 12:23=20
/usr/lib/libjs.so.0.1.0
-rwxr-xr-x =A0 =A01 root =A0 =A0 root =A0 =A0 =A0 551832 oct =A01 20:54 /=
usr/lib/libmozjs.so*

I have tried to run configure with:
--with-mozjs=3D/usr/lib =A0or
--with-mozjs=3D/usr/ or


and always I have the same error that didn't found the libmozjs.

I don't what to do to compile with javascript support.=20
Please, can anyone tell me what to do?

Best regards,

--=20
Leopold Palomo Avellaneda

Linux User 152692=20
Catalonia

[openvrml-develop] Does OpenVRML 0.11.x allow for adding/modifying fields

[John R. <ric...@sp...>]

Hello,

Does OpenVRML 0.11.x allow for adding/modifying fields? I want to run a C++
or Java or FORTRAN simulation and then based upon the results of some
algorithm executing in the simulation direct OpenVRML to change a nodes
position, add or subtract a node and change the value of a nodes field.

I presume that the easiest way would be to subclass ViewerOpenGL and then
modify the scene graph. Sort of like a direct hook to OpenVRML rather than
through something like the EAI.

So, I also suppose that I should ask if OpenVRML 0.11.x supports the EAI?
What about the Script node?

I suppose that there would have to be some method getting the target node
for modification. Any ideas?

Any answers and thoughts would be appreciated.

John F. Richardson

FW: [openvrml-develop] Boudning boxes, setting viewpoints & perspectives

[Gregory H. <gha...@ya...>]

Hi - this message provides more information on the specific task that
I'm having trouble accomplishing with OpenVRML.  Any additional help
would be appreciated!
=20
Thanks for all this help, <helpful list subscriber's name removed>. I am
not working with just one part described in a single VRML file - these
parts are uploaded to me as VRML files and I need to create the six
images from those files.  I don't have control over the authoring of the
VRML, unfortunately.  They are created by a CAD tool in most cases and
do not contain any viewpoint information.=20
=20
I do, however, have the ability to modify my class that's derived from
ViewerOpenGL (or even ViewerOpenGL if I have to).  Here's the
pseudo-code of what I want to do:
=20
          1. Receive Uploaded File
=20
          2. Compute Bounding Box of the "part" described in the file
=20
          3. Compute six points that give me views of the object that
result in a nice edge-to-edge "fit to window" for the "part" given a
particular window size
=20
          4. Create renderings from each of these six points, into an
OpenGL rendering context that writes to a DIB (bitmap)
=20
          5. Compress the six bitmaps into JPG's
=20
          6. Return the six JPG's to the user
=20
I can do all of this with ease except numbers 2 and 3, which require me
to understand a little bit more about the OpenVRML library/toolkit.  Do
you think it's possible to solve this problem using the OpenVRML kit?
Thanks again for your help - it really is appreciated.
=20
// greg //
=20
-----Original Message-----
From: <helpful list subscriber>
Sent: Tuesday, February 19, 2002 6:55 PM
To: Gregory Hassett
Subject: Re: [openvrml-develop] Boudning boxes, setting viewpoints &
perspectives
=20
Hello,


The easiest way is to use a VRML authoring tool to incorporate the
viewpoints into your VRML file. Then take screenshots of the resulting
image in WinLookat/MacLookat/Lookat or in a commercial browser.

However, you seem to want to actually modify the ViewerOpenGL. I only
have the Mac source but there is a HandleKey method I believe in
ViewerOpenGL. This is what would process the next and previous viewpoint
commands dynamically. You still need the viewpoints in the VRML file so
you still need an authoring tool. HandleKey will probably just tell you
what key is mapped by the OpenVRML libraries to viewpoint navigation
since there does not seem to be a user manual. You can subclass
ViewerOpenGL and modify HandleKey to get special shortcut navigation
results.

Configuring the ViewerOpenGL methods for particular hardcoded situations
would require re-coding every time your viewing requirements change (say
from corner of part).

<helpful list subscriber>
=20
          [Original message to which the helpful list subscriber
replied:]
=20
Hi,



My goal is to use openvrml to generate six rendered images (one view
from the center of each of the six =1Csides of the bounding box) of a =
part
that I am going to manufacture.

The part is described in a VRML file.



To do this I need to calculate the bounding box, and then configure the
ViewerOpenGL to have the proper viewpoints. If I have two objects, a
VrmlScene which Ive loaded and a ViewerOpenGL-derived class that Im
rendering with, whats the =1Cright way to accomplish this?



Your help is much appreciated!





-- Gregory Hassett, Paris, France

[openvrml-develop] Boudning boxes, setting viewpoints & perspectives

[Gregory H. <gha...@ya...>]

Hi,
 
My goal is to use openvrml to generate six rendered images (one view
from the center of each of the six "sides" of the bounding box) of a
part that I am going to manufacture.
The part is described in a VRML file.
 
To do this I need to calculate the bounding box, and then configure the
ViewerOpenGL to have the proper viewpoints.  If I have two objects, a
VrmlScene which I've loaded and a ViewerOpenGL-derived class that I'm
rendering with, what's the "right" way to accomplish this?
 
Your help is much appreciated!
 
 
-- Gregory Hassett, Paris, France

Re: [openvrml-develop] Re: Vrml97Parser Performance

[Braden M. <br...@en...>]

On Mon, 2002-01-21 at 17:26, Damian McGuckin wrote:
> On 21 Jan 2002, Braden McDaniel wrote:
> 
> > On Fri, 2001-12-28 at 06:50, Damian McGuckin wrote:
> > >
> > > In our quest for speed, we inline'd _GetNextChar() and it turns out that
> > > on a 1000Mhz Pentium, isNewLineChar() is chewing up 10% of the time on
> > > nextToken().
> > >
> > > I was thinking of inlining all the routines in the anonymous namespace.
> >
> > Where inlining is showing a tangible performance increase, I certainly
> > think we should do it.
> 
> We certainly saw the gains under Windows and Linux.
> 
> > I'd like at some point to rewrite our parser to use Spirit instead of
> > ANTLR. I don't know if that'll end up giving us a speed boost, but I'd
> > be surprised if it hurt us. (My primary motivation is that I think
> > Spirit would be a less annoying dependency.) Spirit's still evolving
> > kinda rapidly, and the ANTLR parser works, so I'm not in a rush to get
> > this done.
> 
> I just had a look at Spirit. Looks cool.  Why would it be a less annoying
> dependency?  I must admit I have really only used Lex & Yacc in anger over
> the years.

Spirit grammars are C++ code, so there's no code generation step. You
just compile it--using the Spirit library--like any other C++ code.

> > It looks like you're focusing on the scanner here for performance
> > improvements; and I think any improvements there should help us
> > regardless of whether we're using ANTLR or Spirit.
> 
> The performance of the scanner increases the speed of loading the initial
> scene.  As this is the first thing somebody notices, then it sticks the
> hardest in their mind if it is slow.  I'd like to try and come up with a
> compiled form of the ASCII files.  You'd be amazed at how much time is
> spent reading integers and reals when you are loading a 6MB Ascii VRML
> file.

Ah, yes... The elusive "binary VRML format". :-)

-- 
Braden McDaniel                           e-mail: <br...@en...>
<http://endoframe.com>                    Jabber: <br...@ja...>

Re: [openvrml-develop] Re: Vrml97Parser Performance

[Damian M. <da...@er...>]

On 21 Jan 2002, Braden McDaniel wrote:

> On Fri, 2001-12-28 at 06:50, Damian McGuckin wrote:
> >
> > In our quest for speed, we inline'd _GetNextChar() and it turns out that
> > on a 1000Mhz Pentium, isNewLineChar() is chewing up 10% of the time on
> > nextToken().
> >
> > I was thinking of inlining all the routines in the anonymous namespace.
>
> Where inlining is showing a tangible performance increase, I certainly
> think we should do it.

We certainly saw the gains under Windows and Linux.

> I'd like at some point to rewrite our parser to use Spirit instead of
> ANTLR. I don't know if that'll end up giving us a speed boost, but I'd
> be surprised if it hurt us. (My primary motivation is that I think
> Spirit would be a less annoying dependency.) Spirit's still evolving
> kinda rapidly, and the ANTLR parser works, so I'm not in a rush to get
> this done.

I just had a look at Spirit. Looks cool.  Why would it be a less annoying
dependency?  I must admit I have really only used Lex & Yacc in anger over
the years.

> It looks like you're focusing on the scanner here for performance
> improvements; and I think any improvements there should help us
> regardless of whether we're using ANTLR or Spirit.

The performance of the scanner increases the speed of loading the initial
scene.  As this is the first thing somebody notices, then it sticks the
hardest in their mind if it is slow.  I'd like to try and come up with a
compiled form of the ASCII files.  You'd be amazed at how much time is
spent reading integers and reals when you are loading a 6MB Ascii VRML
file.

Thanks - Damian (McGuckin)

Pacific Engineering Systems International, 22/8 Campbell St, Artarmon NSW 2064
Ph:+61-2-99063377 .. Fx:+61-2-99063468   | unsolicited email not wanted here !
Views and opinions here are mine and not those of any past or present employer

Re: [openvrml-develop] Re: Vrml97Parser Performance

[Braden M. <br...@en...>]

On Fri, 2001-12-28 at 06:50, Damian McGuckin wrote:
> 
> In our quest for speed, we inline'd _GetNextChar() and it turns out that
> on a 1000Mhz Pentium, isNewLineChar() is chewing up 10% of the time on
> nextToken().
> 
> I was thinking of inlining all the routines in the anonymous namespace.

Where inlining is showing a tangible performance increase, I certainly
think we should do it.

> Some questions:
> 
> a)	Could we come up some 'ctype' like scheme whereby we assume
> 	that char is signed and we do a table lookup?

I imagine so.

> b)	Alternatively, is a test such as
> 
> 	(isdigit(c) || ('A' <= c && c <= 'F') || ('a' <= c && c <= 'f'))
> 
> 	always valid for a hex digit.

Looks good to me.

> While I fully appreciate that one should get it right before you make it
> fast, we find that if something takes ages to load, be it a VRML file or
> a web page, then users will complain.

I'd like at some point to rewrite our parser to use Spirit instead of
ANTLR. I don't know if that'll end up giving us a speed boost, but I'd
be surprised if it hurt us. (My primary motivation is that I think
Spirit would be a less annoying dependency.) Spirit's still evolving
kinda rapidly, and the ANTLR parser works, so I'm not in a rush to get
this done.

It looks like you're focusing on the scanner here for performance
improvements; and I think any improvements there should help us
regardless of whether we're using ANTLR or Spirit.

-- 
Braden McDaniel                           e-mail: <br...@en...>
<http://endoframe.com>                    Jabber: <br...@ja...>

[openvrml-develop] Fogging the mirror

[Braden M. <br...@en...>]

For the past few weeks, my responsiveness has been spotty and sporadic.
As some of you know, I've recently relocated from Miami to Washington,
DC, and started a new job. To make things more interesting, I've had a
rash of problems with my machine that can only be attributable to what
must be some *really* evil hardware karma I picked up somewhere.

Over the next week, I'll be trying to catch up on any e-mail that I
didn't get to before. If, by next weekend, you're still waiting on my
response to something raised in the last few weeks, *please* pester me.

-- 
Braden McDaniel                           e-mail: <br...@en...>
<http://endoframe.com>                    Jabber: <br...@ja...>

[openvrml-develop] Transparency

[Michael S. <Mic...@gm...>]

Hi,

in December 2001, this is arrived, but I don't know how to handle the 
Transparancy object correct.
Where and how may a Z-Sort implemented to sort all objects an draw the 
transparancy object
at last

Thanx Micha.

Re: [openvrml-develop] Re: Installation of GTKLOOKAT

[Thomas F. <tf...@ro...>]

Thomas Flynn wrote:

<snip>
> 
> I'm still trying to figure out why the gtkgl macro fails on my 
> system...if I manually edit the makefiles, all seems to be fine...
> 


The problem turned out to be in the gtkgl macro (pointing at OV_PATH_GL 
when should point at OV_CHECK_GL). I'll submit a patch for this plus 
some other gtklookat issues I've found.

Tom

Re: [openvrml-develop] Re: Installation of GTKLOOKAT

[Thomas F. <tf...@ro...>]

Damian McGuckin wrote:

>>>Damian McGuckin had the following problem building GTKLOOKAT from
>>>sources
>>>
>>>aclocal: configure.in: 9: macro `AM_PATH_GTK' not found in library
>>>
>>>It won't be found in /usr/local/share/aclocal because it came with the
>>>release of the O/S.
>>>
>>>What's the problem?
>>>
>>>
<snip>
> 
> Obviously the installation process is relying on some macro set-up by the
> rpm.  Any ideas what it is?  Tom, do you have some m4 macro related to gtk
> in the area /usr/local/share/aclocal, or anywhere else for that matter?
> 


Installing the gtk rpm put gtk.m4 under /usr/share/aclocal on my machine.

I'm still trying to figure out why the gtkgl macro fails on my 
system...if I manually edit the makefiles, all seems to be fine...

Later,
Tom