From bagvian at gmail.com  Wed Oct  1 17:08:40 2008
From: bagvian at gmail.com (bagvian)
Date: Wed, 1 Oct 2008 17:08:40 +0200
Subject: graphics speed from 4.1.2 to 5.0.2 and virtual graphical window
Message-ID: <e578e190810010808jac39e7enf326a854053b32f2@mail.gmail.com>

Hi there,

My problem is the following: I want to generate many plots and export
them as, say gif file to see them later.
I wrote an example that does such a task:

for i=0:9
  // Create a graphival window
  scf(i);
  f=gcf()
  f.figure_size= [1600,1200];

  // Plot a first random function with using a couple of graphical options
  subplot(2,1,1); plot(rand(1,100),'b-x');
  a=gca();
  a.tight_limits="on";
  legends(["rand0"],[2],opt="ur");
  xgrid(1);

  // Plot a second random function
  subplot(2,1,2); plot(rand(1,100),'b-x');
  a=gca();
  a.tight_limits="on";
  legends(["rand1"],[2],opt="ur");
  xgrid(1);

  // Export plot into gif and delete the current graphical window
  xs2gif(i,sprintf('image_%d.gif',i));
  xdel(i);
end

This program runs within 75s with Scilab 5.0.2 on my computer.
Now, it runs within only 5s with Scilab 4.1.2 on the same computer.

Is there an explanation for this lack of performance reached by 5.0.2?
Is there a way to improve the function above to get a fast result with 5.0.2?

The best way would be not to create a visible graphical window but
create a "virtual graphical window" instead that cannot be seen but
that can be used to export graphics.
Any idea about how to do such a thing?

Thanks a lot in advance.
Cheers, B.


From geraldonetto at gmail.com  Wed Oct  1 17:29:32 2008
From: geraldonetto at gmail.com (Geraldo Netto)
Date: Wed, 1 Oct 2008 12:29:32 -0300
Subject: [scilab-Users] graphics speed from 4.1.2 to 5.0.2 and virtual graphical window
In-Reply-To: <e578e190810010808jac39e7enf326a854053b32f2@mail.gmail.com>
References: <e578e190810010808jac39e7enf326a854053b32f2@mail.gmail.com>
Message-ID: <faf3083f0810010829w3b5497c6na90117901c58ce8e@mail.gmail.com>

Hi Guys,

It seems a java bottleneck
i mean this performance issue that comes from java architecture

while i'm pretty newbie to ensure this
i guess it can be improved by changing
the swing to another 'widget toolkit' like swt[1]
which uses a mix of java and native api to draw the interfaces....

[1] http://en.wikipedia.org/wiki/Standard_Widget_Toolkit

See Ya
(and i hope it helps :))

Geraldo
S?o Paulo, Brasil, -3gmt
site: http://exdev.sf.net/
msn: geraldo_boca_at_hotmail.com
skype: geraldo-netto
icq: 145-061-456

2008/10/1 bagvian <bagvian at gmail.com>:
> Hi there,
>
> My problem is the following: I want to generate many plots and export
> them as, say gif file to see them later.
> I wrote an example that does such a task:
>
> for i=0:9
>  // Create a graphival window
>  scf(i);
>  f=gcf()
>  f.figure_size= [1600,1200];
>
>  // Plot a first random function with using a couple of graphical options
>  subplot(2,1,1); plot(rand(1,100),'b-x');
>  a=gca();
>  a.tight_limits="on";
>  legends(["rand0"],[2],opt="ur");
>  xgrid(1);
>
>  // Plot a second random function
>  subplot(2,1,2); plot(rand(1,100),'b-x');
>  a=gca();
>  a.tight_limits="on";
>  legends(["rand1"],[2],opt="ur");
>  xgrid(1);
>
>  // Export plot into gif and delete the current graphical window
>  xs2gif(i,sprintf('image_%d.gif',i));
>  xdel(i);
> end
>
> This program runs within 75s with Scilab 5.0.2 on my computer.
> Now, it runs within only 5s with Scilab 4.1.2 on the same computer.
>
> Is there an explanation for this lack of performance reached by 5.0.2?
> Is there a way to improve the function above to get a fast result with 5.0.2?
>
> The best way would be not to create a visible graphical window but
> create a "virtual graphical window" instead that cannot be seen but
> that can be used to export graphics.
> Any idea about how to do such a thing?
>
> Thanks a lot in advance.
> Cheers, B.
>


From sylvestre.ledru at scilab.org  Wed Oct  1 17:42:29 2008
From: sylvestre.ledru at scilab.org (Sylvestre Ledru)
Date: Wed, 01 Oct 2008 17:42:29 +0200
Subject: [scilab-Users] graphics speed from 4.1.2 to 5.0.2 and virtual
	graphical window
In-Reply-To: <faf3083f0810010829w3b5497c6na90117901c58ce8e@mail.gmail.com>
References: <e578e190810010808jac39e7enf326a854053b32f2@mail.gmail.com>
	 <faf3083f0810010829w3b5497c6na90117901c58ce8e@mail.gmail.com>
Message-ID: <1222875749.31019.236.camel@korcula.inria.fr>


> while i'm pretty newbie to ensure this
> i guess it can be improved by changing
> the swing to another 'widget toolkit' like swt[1]
> which uses a mix of java and native api to draw the interfaces....
> 
> [1] http://en.wikipedia.org/wiki/Standard_Widget_Toolkit
On this URL, you can read:
"More "real-life" benchmarks show no clear winner [between Swing & SWT],
and showed that the results greatly depend on the context and the
environments[11]."

But if someone wants to create a SWT GUI of Scilab, he is most welcome
(and most of the GUI code has been rewritten this the possibility to
change of GUI API quite easily).

Sylvestre




From geraldonetto at gmail.com  Wed Oct  1 17:48:16 2008
From: geraldonetto at gmail.com (Geraldo Netto)
Date: Wed, 1 Oct 2008 12:48:16 -0300
Subject: [scilab-Users] graphics speed from 4.1.2 to 5.0.2 and virtual graphical window
In-Reply-To: <1222875749.31019.236.camel@korcula.inria.fr>
References: <e578e190810010808jac39e7enf326a854053b32f2@mail.gmail.com>
	 <faf3083f0810010829w3b5497c6na90117901c58ce8e@mail.gmail.com>
	 <1222875749.31019.236.camel@korcula.inria.fr>
Message-ID: <faf3083f0810010848ucf024c6r5052b5ad2de8ec7c@mail.gmail.com>

;)

btw, what about javolution: http://javolution.org/
would be interesting using such a component?
again, i'm pretty newbie to say so, but at least
it seems interesting...

See Ya,

Geraldo
S?o Paulo, Brasil, -3gmt
site: http://exdev.sf.net/
msn: geraldo_boca_at_hotmail.com
skype: geraldo-netto
icq: 145-061-456

2008/10/1 Sylvestre Ledru <sylvestre.ledru at scilab.org>:
>
>> while i'm pretty newbie to ensure this
>> i guess it can be improved by changing
>> the swing to another 'widget toolkit' like swt[1]
>> which uses a mix of java and native api to draw the interfaces....
>>
>> [1] http://en.wikipedia.org/wiki/Standard_Widget_Toolkit
> On this URL, you can read:
> "More "real-life" benchmarks show no clear winner [between Swing & SWT],
> and showed that the results greatly depend on the context and the
> environments[11]."
>
> But if someone wants to create a SWT GUI of Scilab, he is most welcome
> (and most of the GUI code has been rewritten this the possibility to
> change of GUI API quite easily).
>
> Sylvestre
>
>
>


From marc.fuentes at inrialpes.fr  Wed Oct  1 17:51:48 2008
From: marc.fuentes at inrialpes.fr (marc.fuentes at inrialpes.fr)
Date: Wed, 1 Oct 2008 17:51:48 +0200
Subject: [scilab-Users] graphics speed from 4.1.2 to 5.0.2 and virtual
	graphical window
In-Reply-To: <faf3083f0810010829w3b5497c6na90117901c58ce8e@mail.gmail.com>
References: <e578e190810010808jac39e7enf326a854053b32f2@mail.gmail.com> <faf3083f0810010829w3b5497c6na90117901c58ce8e@mail.gmail.com>
Message-ID: <20081001155148.GA16033@delacroix.inrialpes.fr>

On 12:29 Wed 01 Oct     , Geraldo Netto wrote:
> Hi Guys,
> 
> It seems a java bottleneck
> i mean this performance issue that comes from java architecture
> 
> while i'm pretty newbie to ensure this
> i guess it can be improved by changing
> the swing to another 'widget toolkit' like swt[1]
> which uses a mix of java and native api to draw the interfaces....
> 
> [1] http://en.wikipedia.org/wiki/Standard_Widget_Toolkit
> 
> See Ya
> (and i hope it helps :))
> 
> Geraldo
> S?o Paulo, Brasil, -3gmt
> site: http://exdev.sf.net/
> msn: geraldo_boca_at_hotmail.com
> skype: geraldo-netto
> icq: 145-061-456
> 
> 2008/10/1 bagvian <bagvian at gmail.com>:
> > Hi there,
> >
> > My problem is the following: I want to generate many plots and export
> > them as, say gif file to see them later.
> > I wrote an example that does such a task:
> >
> > for i=0:9
> >  // Create a graphival window
> >  scf(i);
> >  f=gcf()
> >  f.figure_size= [1600,1200];
> >
> >  // Plot a first random function with using a couple of graphical options
> >  subplot(2,1,1); plot(rand(1,100),'b-x');
> >  a=gca();
> >  a.tight_limits="on";
> >  legends(["rand0"],[2],opt="ur");
> >  xgrid(1);
> >
> >  // Plot a second random function
> >  subplot(2,1,2); plot(rand(1,100),'b-x');
> >  a=gca();
> >  a.tight_limits="on";
> >  legends(["rand1"],[2],opt="ur");
> >  xgrid(1);
> >
> >  // Export plot into gif and delete the current graphical window
> >  xs2gif(i,sprintf('image_%d.gif',i));
> >  xdel(i);
> > end
> >
> > This program runs within 75s with Scilab 5.0.2 on my computer.
> > Now, it runs within only 5s with Scilab 4.1.2 on the same computer.
> >
> > Is there an explanation for this lack of performance reached by 5.0.2?
> > Is there a way to improve the function above to get a fast result with 5.0.2?
> >
> > The best way would be not to create a visible graphical window but
> > create a "virtual graphical window" instead that cannot be seen but
> > that can be used to export graphics.
> > Any idea about how to do such a thing?
> >
> > Thanks a lot in advance.
> > Cheers, B.
> >
> 
Indeed, scilab wants to mimick matlab using java, and as a consequence
it becomes very slow. In matlab using -nojvm enables matlab to go faster
but does not prevent plotting graphics.
But with scilab 5.0.2 , disabling java support at compile time prevents you to plot
graphics(you must compile with option "--without-gui")
The solution : go on with scilab-4.1.2!



From cyril.leroux at stage.cnes.fr  Thu Oct  2 13:14:53 2008
From: cyril.leroux at stage.cnes.fr (LeRoux Cyril)
Date: Thu, 2 Oct 2008 13:14:53 +0200
Subject: Ginput - How to retrieve the mouse cursor positioning
Message-ID: <092785B790DCD043BA45401EDA43D9B50304C43A@cst-xch-003.cnesnet.ad.cnes.fr>

Hi,

Does anybody know how to get the mouse cursor positioning with Scilab ?
I want to get the coordinates of an area in a graphic window. Is there
any similar function to Matlab ginput function ?

Thanks,

Cyril Le Roux


From ricci2 at unisi.it  Thu Oct  2 18:21:25 2008
From: ricci2 at unisi.it (Maso Ricci)
Date: Thu, 02 Oct 2008 18:21:25 +0200
Subject: xset  for subplot
Message-ID: <48E4F505.2000905@unisi.it>


-- 
Hi,

I plotted some subplots using the xsetech command.
now I need to change the color map of only ONE plot but if I use xset it 
changes the color of all the subplots.

Does anybody know whether there is a kind of xset command working only 
with subplot

thanks in advance

Maso





Maso Ricci Ph.D
Dept. of Chemical and Biosystem Science
University of Siena
Via A. Moro, 2 ? 53100 Siena
ITALY
Tel: +39 0577 234367
Fax: +39 0577 234177
email: ricci2 at unisi.it


From michelidea at hotmail.com  Fri Oct  3 15:12:05 2008
From: michelidea at hotmail.com (D M)
Date: Fri, 3 Oct 2008 13:12:05 +0000
Subject: translating from matlab
Message-ID: <COL108-W737B941DF35E876FD9D124C03C0@phx.gbl>


Hi all,

i'd like to 'port' some matlab scribts from school to scilab.

-----------file.m----------------
v=[exp(j*pi*10/50) exp(-j*pi*10/50)];
b=poly(v);
a=poly(0.9*v);

// frequency response of filter
...
...
----------file.m-----------------

The first line is no problem, but i'm getting in trouble with poly. A poly command exists in scilab, but as far as i know has a different meaning.
Is there an equivalent to the matlab-poly in scilab? The stuff with the frequency response I will do with frmag, but first I have to solve the poly-problem.

Thank you!

fb
_________________________________________________________________
Explore the seven wonders of the world
http://search.msn.com/results.aspx?q=7+wonders+world&mkt=en-US&form=QBRE

From vincent.couvert at scilab.org  Fri Oct  3 15:30:33 2008
From: vincent.couvert at scilab.org (Vincent COUVERT)
Date: Fri, 03 Oct 2008 15:30:33 +0200
Subject: [scilab-Users] translating from matlab
In-Reply-To: <COL108-W737B941DF35E876FD9D124C03C0@phx.gbl>
References: <COL108-W737B941DF35E876FD9D124C03C0@phx.gbl>
Message-ID: <48E61E79.6090603@scilab.org>

Hi,

By default Scilab poly takes input values as polynomial coefficients but 
Matlab poly takes theses values as roots for polynomial.

Scilab function has an option to precize your input values are coefficients.

Vincent

D M a ?crit :
> Hi all,
>
> i'd like to 'port' some matlab scribts from school to scilab.
>
> -----------file.m----------------
> v=[exp(j*pi*10/50) exp(-j*pi*10/50)];
> b=poly(v);
> a=poly(0.9*v);
>
> // frequency response of filter
> ...
> ...
> ----------file.m-----------------
>
> The first line is no problem, but i'm getting in trouble with poly. A poly command exists in scilab, but as far as i know has a different meaning.
> Is there an equivalent to the matlab-poly in scilab? The stuff with the frequency response I will do with frmag, but first I have to solve the poly-problem.
>
> Thank you!
>
> fb
> _________________________________________________________________
> Explore the seven wonders of the world
> http://search.msn.com/results.aspx?q=7+wonders+world&mkt=en-US&form=QBRE



From WeilandDo at gmx.de  Tue Oct  7 23:59:36 2008
From: WeilandDo at gmx.de (Dominik Weiland)
Date: Tue, 7 Oct 2008 22:59:36 +0100
Subject: Creating a scicos model using a script
Message-ID: <934F9990AA164D2BA408BF3F6DB152FD@newccacc98cbf8>

Hello,

 

There may be an obvious solution to this question but I couldn't find it.

Is there a way to create a scicos model using a scilab command rather than
graphically building the model "by-hand" ?

 

Thanks,

Dominik

 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20081007/63a76570/attachment.htm>

From grocer.toolbox at gmail.com  Wed Oct  8 08:21:48 2008
From: grocer.toolbox at gmail.com (Eric Dubois)
Date: Wed, 8 Oct 2008 08:21:48 +0200
Subject: adding help chapters under a common folder
Message-ID: <b04eeae20810072321w346d6047x80736c83ddfeef6f@mail.gmail.com>

Hi Everybody.

In Scilab 5.0 I want to add help chapters under a common folder as for
instance is done for Scicos.

for instance if I have 2 help directorys say c:/help/chapter1/jar and
c:/help/chapter2/jar, then
--> add_help_chapter('chapter 1','c:/help/chapter1/jar');
add_help_chapter('chapter 2','c:/help/chapter2/jar');
--> help

shows 2 folders: chapter1/chapter1 and chapter2/chapter2

But what I would like to see is a generic folder say Newchapters with
inside the 2 folders chapter1 and chapter2 (without their subfolders
with the same name).

How can I realize this?

Thanks.
Eric.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20081008/3e695306/attachment.htm>

From nicolas.vergnes at steria.cnes.fr  Wed Oct  8 09:03:18 2008
From: nicolas.vergnes at steria.cnes.fr (Vergnes Nicolas)
Date: Wed, 8 Oct 2008 09:03:18 +0200
Subject: [scilab-Users] Ginput - How to retrieve the mouse cursor positioning
Message-ID: <EC2C1F912EFD8F498B04551CD2B0408D0480C78D@cst-xch-002.cnesnet.ad.cnes.fr>

Hi, 
You can use xgetmouse and xclick 

-----Message d'origine-----
De : LeRoux Cyril [mailto:cyril.leroux at stage.cnes.fr] 
Envoy? : jeudi 2 octobre 2008 13:15
? : users at lists.scilab.org
Objet : [scilab-Users] Ginput - How to retrieve the mouse cursor positioning

Hi,

Does anybody know how to get the mouse cursor positioning with Scilab ?
I want to get the coordinates of an area in a graphic window. Is there any similar function to Matlab ginput function ?

Thanks,

Cyril Le Roux


From frank at photonita.com.br  Wed Oct  8 18:05:16 2008
From: frank at photonita.com.br (Frank Hrebabetzky)
Date: Wed, 08 Oct 2008 13:05:16 -0300
Subject: Delete default menu in GUI
Message-ID: <48ECDA3C.50603@photonita.com.br>

When I create a GUI with 'figure', it comes with a default menu which I 
don't want. I found the 'delmenu' function, but it requires the name of 
the menu items. So:
How do I systematically search for all of theses names?
   - or -
Is there a simpler way to delete the whole menu, or to suppress it 
firsthand?
-- 
Frank Hrebabetzky	Tel. (48) 3239 2258
Photonita Ltda.		http://www.photonita.com.br
Brazil


From robertgordonkey at yahoo.com  Wed Oct  8 18:08:41 2008
From: robertgordonkey at yahoo.com (Robert Key)
Date: Wed, 8 Oct 2008 09:08:41 -0700 (PDT)
Subject: plot produces a blank window
Message-ID: <307998.65627.qm@web45601.mail.sp1.yahoo.com>

Hi,
 I am running scilab 5.0.2 on linux 86_64. The console window works find however when I plot a simple graph a grey window is launched but there is no plot in it. The error message in the terminal is 

Exception in thread "AWT-EventQueue-0" java.lang.UnsatisfiedLinkError: no gluegen-rt in java.library.path
        at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1682)
        at java.lang.Runtime.loadLibrary0(Runtime.java:823)
        at java.lang.System.loadLibrary(System.java:1030)
        at com.sun.gluegen.runtime.NativeLibLoader.loadLibraryInternal(NativeLibLoader.java:102)
        at com.sun.gluegen.runtime.NativeLibLoader.access$000(NativeLibLoader.java:51)


 I have my classpath setup correctly and included the thirdparty stuff in the classpath.
How do I fix this?
Thanks.

Rob Key



      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20081008/18b87f17/attachment.htm>

From sylvestre.ledru at inria.fr  Wed Oct  8 20:40:03 2008
From: sylvestre.ledru at inria.fr (Sylvestre Ledru)
Date: Wed, 08 Oct 2008 20:40:03 +0200
Subject: [scilab-Users] plot produces a blank window
In-Reply-To: <307998.65627.qm@web45601.mail.sp1.yahoo.com>
References: <307998.65627.qm@web45601.mail.sp1.yahoo.com>
Message-ID: <1223491203.3236.46.camel@zlarin>

Hi,

Le mercredi 08 octobre 2008 ? 09:08 -0700, Robert Key a ?crit :
> Hi,
>  I am running scilab 5.0.2 on linux 86_64. The console window works
> find however when I plot a simple graph a grey window is launched but
> there is no plot in it. The error message in the terminal is 
> 
> Exception in thread "AWT-EventQueue-0" java.lang.UnsatisfiedLinkError:
> no gluegen-rt in java.library.path
>         at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1682)
>         at java.lang.Runtime.loadLibrary0(Runtime.java:823)
>         at java.lang.System.loadLibrary(System.java:1030)
>         at
> com.sun.gluegen.runtime.NativeLibLoader.loadLibraryInternal(NativeLibLoader.java:102)
>         at com.sun.gluegen.runtime.NativeLibLoader.access
> $000(NativeLibLoader.java:51)
> 
>  I have my classpath setup correctly and included the thirdparty stuff
> in the classpath.
> How do I fix this?

libgluegen-rt.so needs to be available in the LD_LIBRARY_PATH.
(The Java <=> C JOGL connector)

Sylvestre




From grocer.toolbox at gmail.com  Wed Oct  8 21:53:32 2008
From: grocer.toolbox at gmail.com (Eric Dubois)
Date: Wed, 8 Oct 2008 21:53:32 +0200
Subject: adding help chapters under a common folder
Message-ID: <b04eeae20810081253m33f9e9ddh34b8a9bb732bf472@mail.gmail.com>

Hi Everybody.

In Scilab 5.0 I want to add help chapters under a common folder as for
instance is done for Scicos.

for instance if I have 2 help directorys say c:/help/chapter1/jar and
c:/help/chapter2/jar, then
--> add_help_chapter('chapter 1','c:/help/chapter1/jar');
add_help_chapter('chapter 2','c:/help/chapter2/jar');
--> help

shows 2 folders: chapter1/chapter1 and chapter2/chapter2

But what I would like to see is a generic folder say Newchapters with
inside the 2 folders chapter1 and chapter2 (without their subfolders
with the same name).

How can I realize this?

Thanks.
Eric.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20081008/919f2a16/attachment.htm>

From robertgordonkey at yahoo.com  Thu Oct  9 08:49:32 2008
From: robertgordonkey at yahoo.com (Robert Key)
Date: Wed, 8 Oct 2008 23:49:32 -0700 (PDT)
Subject: [scilab-Users] plot produces a blank window
Message-ID: <470624.20000.qm@web45603.mail.sp1.yahoo.com>

Hi,
Thanks so much for your reply. I have placed a libgluegen-rt.so 64bit in my /usr/local/lib directory and did a ldconfig. The plot is still blank but I will work on the error mesage.
I will also try a binary version for i386 linux.
Thanks for your time.
Scilab is wonderful - even without the plots!
Robert Key

 Rob Key



----- Original Message ----
From: Sylvestre Ledru <sylvestre.ledru at inria.fr>
To: users at lists.scilab.org
Sent: Wednesday, October 8, 2008 8:40:03 PM
Subject: Re: [scilab-Users] plot produces a blank window

Hi,

Le mercredi 08 octobre 2008 ? 09:08 -0700, Robert Key a ?crit :
> Hi,
>  I am running scilab 5.0.2 on linux 86_64. The console window works
> find however when I plot a simple graph a grey window is launched but
> there is no plot in it. The error message in the terminal is 
> 
> Exception in thread "AWT-EventQueue-0" java.lang.UnsatisfiedLinkError:
> no gluegen-rt in java.library.path
>         at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1682)
>         at java.lang.Runtime.loadLibrary0(Runtime.java:823)
>         at java.lang.System.loadLibrary(System.java:1030)
>         at
> com.sun.gluegen.runtime.NativeLibLoader.loadLibraryInternal(NativeLibLoader..java:102)
>         at com.sun.gluegen.runtime.NativeLibLoader.access
> $000(NativeLibLoader.java:51)
> 
>  I have my classpath setup correctly and included the thirdparty stuff
> in the classpath.
> How do I fix this?

libgluegen-rt.so needs to be available in the LD_LIBRARY_PATH.
(The Java <=> C JOGL connector)

Sylvestre


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20081008/7cef80f2/attachment.htm>

From robertgordonkey at yahoo.com  Thu Oct  9 14:05:41 2008
From: robertgordonkey at yahoo.com (Robert Key)
Date: Thu, 9 Oct 2008 05:05:41 -0700 (PDT)
Subject: plots are still blank
Message-ID: <265452.78698.qm@web45602.mail.sp1.yahoo.com>

Hi all,
 I use scilab 5.0.2 on linux x86_64. jogl 64 bit is installed and LD_LIBRARY_PATH and the CLASSPATH are set up correctly but I still get the error:

java.lang.reflect.InvocationTargetException
        at java.awt.EventQueue.invokeAndWait(EventQueue.java:997)
        at javax.swing.SwingUtilities.invokeAndWait(SwingUtilities.java:1323)
        at org.scilab.modules.gui.bridge.tab.SwingScilabTab.draw(Unknown Source)
        at org.scilab.modules.gui.tab.ScilabTabBridge.draw(Unknown Source)
        at org.scilab.modules.gui.bridge.ScilabBridge.draw(Unknown Source)
        at org.scilab.modules.gui.tab.ScilabTab.draw(Unknown Source)
        at org.scilab.modules.gui.graphicWindow.ScilabRendererProperties.forceDi
splay(Unknown Source)
        at org.scilab.modules.renderer.figureDrawing.DrawableFigureGL.drawCanvas
(Unknown Source)
Caused by: java.lang.NoClassDefFoundError: Could not initialize class com.sun.op
engl.impl.x11.X11GLDrawableFactory

This class com.sun.opengl.impl.x11.X11GLDrawableFactory exists in the gluegen-rt.jar and is in the classpath.
Also all the libgluegen-rt.so and related libraries are in the LD_LIBRARY_PATH and ldconfig has been run many times.

How can I correct this.
Many thanks,

 Rob Key



      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20081009/ec0c6fda/attachment.htm>

From robertgordonkey at yahoo.com  Fri Oct 10 10:38:11 2008
From: robertgordonkey at yahoo.com (Robert Key)
Date: Fri, 10 Oct 2008 01:38:11 -0700 (PDT)
Subject: [scilab-Users] SOLVED: plots are still blank
Message-ID: <218066.98489.qm@web45607.mail.sp1.yahoo.com>

Hi I have managed to solve my own plotting problem. Here is the solution in case anyone has the same problem:

I placed all run time libraries required by scilab in /opt/jdk1.6.0_07/jre/lib/amd64 (this would be i386 on 32 bit linux)
such as libgluegen-rt.so and all the jogl*.so. Somehow they were not found when placed in  /usr/local/lib and ldconfig then run.

I placed all the gluegen-rt.jar and jogl*.jar class libraries in /opt/jdk1.6.0_07/jre/lib/ext. If you do this then it is not necessary to include them in your CLASSPATH variable. I tried putting then in a directory and including that in the CLASSPATH but this did not work for some reason.

Now my plots and sciocos all work fantastically on linux x86_64!!!
Thanks for a great scientific software application.
Cheers and happy computing,
  Robert Key

 Rob Key



----- Original Message ----
From: Robert Key <robertgordonkey at yahoo.com>
To: users at lists.scilab.org
Sent: Thursday, October 9, 2008 2:05:41 PM
Subject: [scilab-Users] plots are still blank


Hi all,
 I use scilab 5.0.2 on linux x86_64. jogl 64 bit is installed and LD_LIBRARY_PATH and the CLASSPATH are set up correctly but I still get the error:

java.lang.reflect.InvocationTargetException
        at java.awt.EventQueue.invokeAndWait(EventQueue.java:997)
        at javax.swing.SwingUtilities.invokeAndWait(SwingUtilities.java:1323)
        at org.scilab.modules.gui.bridge.tab.SwingScilabTab.draw(Unknown Source)
        at org.scilab.modules.gui.tab.ScilabTabBridge.draw(Unknown Source)
        at org.scilab.modules.gui.bridge.ScilabBridge.draw(Unknown Source)
        at org.scilab.modules.gui.tab.ScilabTab.draw(Unknown Source)
        at org.scilab.modules.gui.graphicWindow.ScilabRendererProperties.forceDi
splay(Unknown Source)
        at org.scilab.modules.renderer.figureDrawing.DrawableFigureGL.drawCanvas
(Unknown Source)
Caused by: java.lang.NoClassDefFoundError: Could not initialize class com.sun.op
engl.impl.x11.X11GLDrawableFactory

This class com.sun.opengl.impl.x11.X11GLDrawableFactory exists in the gluegen-rt.jar and is in the classpath.
Also all the libgluegen-rt.so and related libraries are in the LD_LIBRARY_PATH and ldconfig has been run many times.

How can I correct this.
Many thanks,

 Rob Key


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20081010/e6079249/attachment.htm>

From sylvestre.ledru at scilab.org  Fri Oct 10 10:41:30 2008
From: sylvestre.ledru at scilab.org (Sylvestre Ledru)
Date: Fri, 10 Oct 2008 10:41:30 +0200
Subject: [scilab-Users] SOLVED: plots are still blank
In-Reply-To: <218066.98489.qm@web45607.mail.sp1.yahoo.com>
References: <218066.98489.qm@web45607.mail.sp1.yahoo.com>
Message-ID: <1223628090.28352.81.camel@korcula.inria.fr>

Thanks for the feedback. It is much appreciated.

Would you mind updating this wiki page:
http://wiki.scilab.org/Compiling_Scilab_5.x_under_GNU-Linux_Unix_-_FAQ

with these information ?

Cheers,
Sylvestre


 Le vendredi 10 octobre 2008 ? 01:38 -0700, Robert Key a ?crit :
> Hi I have managed to solve my own plotting problem. Here is the
> solution in case anyone has the same problem:
> 
> I placed all run time libraries required by scilab
> in /opt/jdk1.6.0_07/jre/lib/amd64 (this would be i386 on 32 bit linux)
> such as libgluegen-rt.so and all the jogl*.so. Somehow they were not
> found when placed in  /usr/local/lib and ldconfig then run.
> 
> I placed all the gluegen-rt.jar and jogl*.jar class libraries
> in /opt/jdk1.6.0_07/jre/lib/ext. If you do this then it is not
> necessary to include them in your CLASSPATH variable. I tried putting
> then in a directory and including that in the CLASSPATH but this did
> not work for some reason.
> 
> Now my plots and sciocos all work fantastically on linux x86_64!!!
> Thanks for a great scientific software application.
> Cheers and happy computing,
>   Robert Key
>  
> Rob Key
> 
> 
> 
> ----- Original Message ----
> From: Robert Key <robertgordonkey at yahoo.com>
> To: users at lists.scilab.org
> Sent: Thursday, October 9, 2008 2:05:41 PM
> Subject: [scilab-Users] plots are still blank
> 
> Hi all,
>  I use scilab 5.0.2 on linux x86_64. jogl 64 bit is installed and
> LD_LIBRARY_PATH and the CLASSPATH are set up correctly but I still get
> the error:
> 
> java.lang.reflect.InvocationTargetException
>         at java.awt.EventQueue.invokeAndWait(EventQueue.java:997)
>         at
> javax.swing.SwingUtilities.invokeAndWait(SwingUtilities.java:1323)
>         at
> org.scilab.modules.gui.bridge.tab.SwingScilabTab.draw(Unknown Source)
>         at org.scilab.modules.gui.tab.ScilabTabBridge.draw(Unknown
> Source)
>         at org.scilab.modules.gui.bridge.ScilabBridge.draw(Unknown
> Source)
>         at org.scilab.modules.gui.tab.ScilabTab.draw(Unknown Source)
>         at
> org.scilab.modules.gui.graphicWindow.ScilabRendererProperties.forceDi
> splay(Unknown Source)
>         at
> org.scilab.modules.renderer.figureDrawing.DrawableFigureGL.drawCanvas
> (Unknown Source)
> Caused by: java.lang.NoClassDefFoundError: Could not initialize class
> com.sun.op
> engl.impl.x11.X11GLDrawableFactory
> 
> This class com.sun.opengl.impl.x11.X11GLDrawableFactory exists in the
> gluegen-rt.jar and is in the classpath.
> Also all the libgluegen-rt.so and related libraries are in the
> LD_LIBRARY_PATH and ldconfig has been run many times.
> 
> How can I correct this.
> Many thanks,
>  
> Rob Key
> 
> 
> 
> 
> 



From morgan.fitzgibbon at m4s.be  Fri Oct 10 13:42:15 2008
From: morgan.fitzgibbon at m4s.be (Morgan Fitzgibbon)
Date: Fri, 10 Oct 2008 12:42:15 +0100
Subject: Problems with sublpot
Message-ID: <48EF3F97.9020001@m4s.be>

Hi,

When I am trying to use subplot with SciLab 5.0.2 it does not appear to 
split the drawing screen as expected? I am trying the following;

scf(1);
subplot(2,1,1),
plot2d(myfreq,20*log10(abs(Gol)),logflag='ln')

I would expect the drawing screen to be split in half and the plot to be 
plotted on the upper half?Instead the whole drawing window is being 
used? Any suggestions?

Thanks
Morgan




From UVV-mail at yandex.ru  Sat Oct 11 05:32:32 2008
From: UVV-mail at yandex.ru (Vyacheslav V. Yurkov)
Date: Sat, 11 Oct 2008 11:32:32 +0800
Subject: Scilab 5.0.1 Failure
Message-ID: <48F01E50.90203@yandex.ru>

Hi.
I use Slackware Linux 12.1 and Scilab 5.0.1.
When I'm trying to call set up window for /lookup table/ element from 
non_linear palette scilab has failed and I get following messages in 
console:
terminate called after throwing an instance of 
'GiwsException::JniCallMethodException'
  what():  Exception when calling Java method : index less than zero.
 at javax.swing.JMenu.getItem(Unknown Source)
 at 
org.scilab.modules.gui.bridge.menubar.SwingScilabMenuBar.setSubMenuEnabled(Unknown 
Source)
 at 
org.scilab.modules.gui.bridge.CallScilabBridge.setFigureSubMenuEnabled(Unknown 
Source)
index less than zero.
 at javax.swing.JMenu.getItem(Unknown Source)
 at 
org.scilab.modules.gui.bridge.menubar.SwingScilabMenuBar.setSubMenuEnabled(Unknown 
Source)
 at 
org.scilab.modules.gui.bridge.CallScilabBridge.setFigureSubMenuEnabled(Unknown 
Source)

./scilab: line 194:  4211 Aborted                 "$SCILABBIN" $*
Are there any suggetions?

-- 
Vyacheslav V. Yurkov <uvv.mail at gmail.com>	Process Control Engineer
Norilsk						Tel: +79069001870

Department of process control systems
Open Joint Stock Company MMC Norilsk Nickel, brass works

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20081011/66bfb946/attachment.htm>

From simone.mannori at scilab.org  Sat Oct 11 08:29:15 2008
From: simone.mannori at scilab.org (Simone Mannori)
Date: Sat, 11 Oct 2008 08:29:15 +0200
Subject: [scilab-Users] Scilab 5.0.1 Failure
In-Reply-To: <48F01E50.90203@yandex.ru>
References: <48F01E50.90203@yandex.ru>
Message-ID: <1223706555.3835.15.camel@pinarellu.inria.fr>

Dear Sir,

the 5.0.2 version does not crash (under Windows) but produce an alarming
"Scilab critical error .... with "unsetmenu" function".

We are investing the problem on Linux too.

I think that you have found a bug. 

Do you know how to use Bugzilla?
Have systematically organized bug reports can optimize our bug fixing
effort.

Do not hesitate to ask help or signal other bug(s) on Scilab/Scicos :).

Thanks for using Scilab/Scicos.

Simone Mannori - The Scilab Consortium (DIGITEO)
Scilab/Scicos Embedded Applications

//============================================================================


On Sat, 2008-10-11 at 11:32 +0800, Vyacheslav V." Yurkov wrote:
> Hi.
> I use Slackware Linux 12.1 and Scilab 5.0.1.
> When I'm trying to call set up window for lookup table element from
> non_linear palette scilab has failed and I get following messages in
> console:
> terminate called after throwing an instance of
> 'GiwsException::JniCallMethodException'
>   what():  Exception when calling Java method : index less than zero.
>  at javax.swing.JMenu.getItem(Unknown Source)
>  at
> org.scilab.modules.gui.bridge.menubar.SwingScilabMenuBar.setSubMenuEnabled(Unknown Source)
>  at
> org.scilab.modules.gui.bridge.CallScilabBridge.setFigureSubMenuEnabled(Unknown Source)
> index less than zero.
>  at javax.swing.JMenu.getItem(Unknown Source)
>  at
> org.scilab.modules.gui.bridge.menubar.SwingScilabMenuBar.setSubMenuEnabled(Unknown Source)
>  at
> org.scilab.modules.gui.bridge.CallScilabBridge.setFigureSubMenuEnabled(Unknown Source)
> 
> ./scilab: line 194:  4211 Aborted                 "$SCILABBIN" $*
> Are there any suggetions?
> -- 
> Vyacheslav V. Yurkov <uvv.mail at gmail.com>	Process Control Engineer
> Norilsk						Tel: +79069001870
> 
> Department of process control systems
> Open Joint Stock Company MMC Norilsk Nickel, brass works



From cugcop at brabon.com  Fri Oct 10 18:53:28 2008
From: cugcop at brabon.com (Ilene Hancock)
Date: Sat, 11 Oct 2008 01:53:28 +0900
Subject: I need this Document Tomorrow
Message-ID: <01c92b44$27aa1c00$e64699da@cugcop>

Hallo

 I need this document translated, and the translation is to be of high quality. 
The attorney asked me to be especially careful with page 2. As I need to 
submit the document tomorrow, please have it checked and translated for me today.
 
 I have deposited $150 to your credit card account that you gave me the last time. 
Let me know if any questions occur.

 P.S. The document is in the ZIP-compressed MS Word file attached to this message. 
I look forward to getting the result ASAP.

See you then
Ilene Hancock
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Document.zip
Type: application/zip
Size: 32730 bytes
Desc: not available
URL: <https://lists.scilab.org/pipermail/users/attachments/20081011/31953454/attachment.zip>

From robertgordonkey at yahoo.com  Sat Oct 11 19:35:18 2008
From: robertgordonkey at yahoo.com (Robert Key)
Date: Sat, 11 Oct 2008 10:35:18 -0700 (PDT)
Subject: [scilab-Users] SOLVED: plots are still blank
Message-ID: <731441.73651.qm@web45612.mail.sp1.yahoo.com>

Hi Sylvestre I have looked at the url for the faq. I don't know how to update the faq. Do I need to register to that mailing list?
Please let me know how to update the faq and I will gladly do it.
Thanks,
  Robert Key

 Rob Key



----- Original Message ----
From: Sylvestre Ledru <sylvestre.ledru at scilab.org>
To: users at lists.scilab.org
Sent: Friday, October 10, 2008 10:41:30 AM
Subject: Re: [scilab-Users] SOLVED: plots are still blank

Thanks for the feedback. It is much appreciated.

Would you mind updating this wiki page:
http://wiki.scilab.org/Compiling_Scilab_5.x_under_GNU-Linux_Unix_-_FAQ

with these information ?

Cheers,
Sylvestre


Le vendredi 10 octobre 2008 ? 01:38 -0700, Robert Key a ?crit :
> Hi I have managed to solve my own plotting problem. Here is the
> solution in case anyone has the same problem:
> 
> I placed all run time libraries required by scilab
> in /opt/jdk1.6.0_07/jre/lib/amd64 (this would be i386 on 32 bit linux)
> such as libgluegen-rt.so and all the jogl*.so. Somehow they were not
> found when placed in  /usr/local/lib and ldconfig then run.
> 
> I placed all the gluegen-rt.jar and jogl*.jar class libraries
> in /opt/jdk1.6.0_07/jre/lib/ext. If you do this then it is not
> necessary to include them in your CLASSPATH variable. I tried putting
> then in a directory and including that in the CLASSPATH but this did
> not work for some reason.
> 
> Now my plots and sciocos all work fantastically on linux x86_64!!!
> Thanks for a great scientific software application.
> Cheers and happy computing,
>   Robert Key
>  
> Rob Key
> 
> 
> 
> ----- Original Message ----
> From: Robert Key <robertgordonkey at yahoo.com>
> To: users at lists.scilab.org
> Sent: Thursday, October 9, 2008 2:05:41 PM
> Subject: [scilab-Users] plots are still blank
> 
> Hi all,
>  I use scilab 5.0.2 on linux x86_64. jogl 64 bit is installed and
> LD_LIBRARY_PATH and the CLASSPATH are set up correctly but I still get
> the error:
> 
> java.lang.reflect.InvocationTargetException
>         at java.awt.EventQueue.invokeAndWait(EventQueue.java:997)
>         at
> javax.swing.SwingUtilities.invokeAndWait(SwingUtilities.java:1323)
>         at
> org.scilab.modules.gui..bridge.tab.SwingScilabTab.draw(Unknown Source)
>         at org.scilab.modules.gui.tab.ScilabTabBridge.draw(Unknown
> Source)
>         at org.scilab.modules.gui.bridge.ScilabBridge.draw(Unknown
> Source)
>         at org.scilab.modules.gui.tab.ScilabTab.draw(Unknown Source)
>         at
> org.scilab.modules.gui.graphicWindow.ScilabRendererProperties.forceDi
> splay(Unknown Source)
>         at
> org.scilab.modules.renderer.figureDrawing.DrawableFigureGL.drawCanvas
> (Unknown Source)
> Caused by: java.lang.NoClassDefFoundError: Could not initialize class
> com.sun.op
> engl.impl.x11.X11GLDrawableFactory
> 
> This class com.sun.opengl.impl.x11.X11GLDrawableFactory exists in the
> gluegen-rt.jar and is in the classpath.
> Also all the libgluegen-rt.so and related libraries are in the
> LD_LIBRARY_PATH and ldconfig has been run many times.
> 
> How can I correct this.
> Many thanks,
>  
> Rob Key
> 
> 
> 
> 
> 


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20081011/f18ca3c4/attachment.htm>

From sylvestre.ledru at scilab.org  Sun Oct 12 01:13:25 2008
From: sylvestre.ledru at scilab.org (Sylvestre Ledru)
Date: Sun, 12 Oct 2008 01:13:25 +0200
Subject: [scilab-Users] SOLVED: plots are still blank
In-Reply-To: <731441.73651.qm@web45612.mail.sp1.yahoo.com>
References: <731441.73651.qm@web45612.mail.sp1.yahoo.com>
Message-ID: <1223766805.3326.3.camel@zlarin>

Hello Rob,

Actually, it is pretty easy.
1) create an account the wiki
http://wiki.scilab.org/UserPreferences
2) go the to the page
http://wiki.scilab.org/Compiling_Scilab_5.x_under_GNU-Linux_Unix_-_FAQ
3) edit the page
4) update the content (basically, copy/paste an example and modify it)

Don't worry to break things. Each page is versionned and it is
straighforward to revert a bad modification/mistakes.

Sylvestre

Le samedi 11 octobre 2008 ? 10:35 -0700, Robert Key a ?crit :
> Hi Sylvestre I have looked at the url for the faq. I don't know how to
> update the faq. Do I need to register to that mailing list?
> Please let me know how to update the faq and I will gladly do it.
> Thanks,
>   Robert Key
>  
> Rob Key
> 
> 
> 
> ----- Original Message ----
> From: Sylvestre Ledru <sylvestre.ledru at scilab.org>
> To: users at lists.scilab.org
> Sent: Friday, October 10, 2008 10:41:30 AM
> Subject: Re: [scilab-Users] SOLVED: plots are still blank
> 
> Thanks for the feedback. It is much appreciated.
> 
> Would you mind updating this wiki page:
> http://wiki.scilab.org/Compiling_Scilab_5.x_under_GNU-Linux_Unix_-_FAQ
> 
> with these information ?
> 
> Cheers,
> Sylvestre
> 
> 
> Le vendredi 10 octobre 2008 ? 01:38 -0700, Robert Key a ?crit :
> > Hi I have managed to solve my own plotting problem. Here is the
> > solution in case anyone has the same problem:
> > 
> > I placed all run time libraries required by scilab
> > in /opt/jdk1.6.0_07/jre/lib/amd64 (this would be i386 on 32 bit
> linux)
> > such as libgluegen-rt.so and all the jogl*.so. Somehow they were not
> > found when placed in  /usr/local/lib and ldconfig then run.
> > 
> > I placed all the gluegen-rt.jar and jogl*.jar class libraries
> > in /opt/jdk1.6.0_07/jre/lib/ext. If you do this then it is not
> > necessary to include them in your CLASSPATH variable. I tried
> putting
> > then in a directory and including that in the CLASSPATH but this did
> > not work for some reason.
> > 
> > Now my plots and sciocos all work fantastically on linux x86_64!!!
> > Thanks for a great scientific software application.
> > Cheers and happy computing,
> >  Robert Key
> >  
> > Rob Key
> > 
> > 
> > 
> > ----- Original Message ----
> > From: Robert Key <robertgordonkey at yahoo.com>
> > To: users at lists.scilab.org
> > Sent: Thursday, October 9, 2008 2:05:41 PM
> > Subject: [scilab-Users] plots are still blank
> > 
> > Hi all,
> >  I use scilab 5.0.2 on linux x86_64. jogl 64 bit is installed and
> > LD_LIBRARY_PATH and the CLASSPATH are set up correctly but I still
> get
> > the error:
> > 
> > java.lang.reflect.InvocationTargetException
> >        at java.awt.EventQueue.invokeAndWait(EventQueue.java:997)
> >        at
> > javax.swing.SwingUtilities.invokeAndWait(SwingUtilities.java:1323)
> >        at
> > org.scilab.modules.gui.bridge.tab.SwingScilabTab.draw(Unknown
> Source)
> >        at org.scilab.modules.gui.tab.ScilabTabBridge.draw(Unknown
> > Source)
> >        at org.scilab.modules.gui.bridge.ScilabBridge.draw(Unknown
> > Source)
> >        at org.scilab.modules.gui.tab.ScilabTab.draw(Unknown Source)
> >        at
> >
> org.scilab.modules.gui.graphicWindow.ScilabRendererProperties.forceDi
> > splay(Unknown Source)
> >        at
> >
> org.scilab.modules.renderer.figureDrawing.DrawableFigureGL.drawCanvas
> > (Unknown Source)
> > Caused by: java.lang.NoClassDefFoundError: Could not initialize
> class
> > com.sun.op
> > engl.impl.x11.X11GLDrawableFactory
> > 
> > This class com.sun.opengl.impl.x11.X11GLDrawableFactory exists in
> the
> > gluegen-rt.jar and is in the classpath.
> > Also all the libgluegen-rt.so and related libraries are in the
> > LD_LIBRARY_PATH and ldconfig has been run many times.
> > 
> > How can I correct this.
> > Many thanks,
> >  
> > Rob Key
> > 
> > 
> > 
> > 
> > 
> 
> 
> 




From windlifephysics at gmail.com  Sun Oct 12 07:54:58 2008
From: windlifephysics at gmail.com (Wind Zero)
Date: Sun, 12 Oct 2008 13:54:58 +0800
Subject: Problem in graph
Message-ID: <544b74f40810112254s69959d0bqb176aec68b45713b@mail.gmail.com>

Hi, everybody.
I am running scilab 5.0.2 on fedora 9 86_64.  My  laptop  is  Gateway.  The
console window  and  the  editor  window  works  well.  However  when  I
plot  a  simple  graph, a black or grey window is  launched but there is no
plot in it. And there is no error message.
I think it may be the problem of the graphic controller. My graphic
controller is the product of Intel. So I installed the newest driver. But
the problem is still existing. Could somebody  give me some help? Thank you
very much!

Zero

-- 
Xu Zhang
Wind_Zero
Department of physics
Wuhan University, P.R.China
windlifephysics at gmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20081012/a1bd4b80/attachment.htm>

From sdr at durietz.se  Sun Oct 12 13:36:12 2008
From: sdr at durietz.se (Stefan Du Rietz)
Date: Sun, 12 Oct 2008 13:36:12 +0200
Subject: findobj()
Message-ID: <48F1E12C.4020803@durietz.se>

I am using Scilab 4.1.2 under Win XP.

1. findobj('Tag','my figure tag') does not find a graphic 
figure, though it can find GUI figures and uicontrols. Is 
that correct?

2. I have also tried findobj() in Scilab 5.0.2, but because 
that has only the graphic figure, findobj() cannot find it.

3. I have problems all the time with the graphics in Scilab 
5.0.2 and often it hangs and must be stopped with Task 
Manager, which also takes long time... ;-(

Would be grateful for your comments.

Stefan


From robertgordonkey at yahoo.com  Sun Oct 12 16:54:14 2008
From: robertgordonkey at yahoo.com (Robert Key)
Date: Sun, 12 Oct 2008 07:54:14 -0700 (PDT)
Subject: [scilab-Users] SOLVED: plots are still blank
Message-ID: <524161.64031.qm@web45615.mail.sp1.yahoo.com>

Hi Sylvestre,
 I have updated the FAQ. I hope the wording is ok. I have included a little more infomation.
Cheers,

 Rob Key



----- Original Message ----
From: Sylvestre Ledru <sylvestre.ledru at scilab.org>
To: users at lists.scilab.org
Sent: Sunday, October 12, 2008 1:13:25 AM
Subject: Re: [scilab-Users] SOLVED: plots are still blank

Hello Rob,

Actually, it is pretty easy.
1) create an account the wiki
http://wiki.scilab.org/UserPreferences
2) go the to the page
http://wiki.scilab.org/Compiling_Scilab_5.x_under_GNU-Linux_Unix_-_FAQ
3) edit the page
4) update the content (basically, copy/paste an example and modify it)

Don't worry to break things. Each page is versionned and it is
straighforward to revert a bad modification/mistakes.

Sylvestre

Le samedi 11 octobre 2008 ? 10:35 -0700, Robert Key a ?crit :
> Hi Sylvestre I have looked at the url for the faq. I don't know how to
> update the faq. Do I need to register to that mailing list?
> Please let me know how to update the faq and I will gladly do it.
> Thanks,
>   Robert Key
>  
> Rob Key
> 
> 
> 
> ----- Original Message ----
> From: Sylvestre Ledru <sylvestre.ledru at scilab.org>
> To: users at lists.scilab.org
> Sent: Friday, October 10, 2008 10:41:30 AM
> Subject: Re: [scilab-Users] SOLVED: plots are still blank
> 
> Thanks for the feedback. It is much appreciated.
> 
> Would you mind updating this wiki page:
> http://wiki.scilab.org/Compiling_Scilab_5.x_under_GNU-Linux_Unix_-_FAQ
> 
> with these information ?
> 
> Cheers,
> Sylvestre
> 
> 
> Le vendredi 10 octobre 2008 ? 01:38 -0700, Robert Key a ?crit :
> > Hi I have managed to solve my own plotting problem. Here is the
> > solution in case anyone has the same problem:
> > 
> > I placed all run time libraries required by scilab
> > in /opt/jdk1.6.0_07/jre/lib/amd64 (this would be i386 on 32 bit
> linux)
> > such as libgluegen-rt.so and all the jogl*.so. Somehow they were not
> > found when placed in  /usr/local/lib and ldconfig then run.
> > 
> > I placed all the gluegen-rt.jar and jogl*.jar class libraries
> > in /opt/jdk1.6.0_07/jre/lib/ext. If you do this then it is not
> > necessary to include them in your CLASSPATH variable.. I tried
> putting
> > then in a directory and including that in the CLASSPATH but this did
> > not work for some reason.
> > 
> > Now my plots and sciocos all work fantastically on linux x86_64!!!
> > Thanks for a great scientific software application.
> > Cheers and happy computing,
> >  Robert Key
> >  
> > Rob Key
> > 
> > 
> > 
> > ----- Original Message ----
> > From: Robert Key <robertgordonkey at yahoo.com>
> > To: users at lists.scilab.org
> > Sent: Thursday, October 9, 2008 2:05:41 PM
> > Subject: [scilab-Users] plots are still blank
> > 
> > Hi all,
> >  I use scilab 5.0.2 on linux x86_64. jogl 64 bit is installed and
> > LD_LIBRARY_PATH and the CLASSPATH are set up correctly but I still
> get
> > the error:
> > 
> > java.lang.reflect.InvocationTargetException
> >        at java.awt.EventQueue.invokeAndWait(EventQueue.java:997)
> >        at
> > javax.swing.SwingUtilities.invokeAndWait(SwingUtilities.java:1323)
> >        at
> > org.scilab.modules.gui.bridge.tab.SwingScilabTab.draw(Unknown
> Source)
> >        at org.scilab.modules.gui.tab.ScilabTabBridge.draw(Unknown
> > Source)
> >        at org.scilab.modules.gui.bridge.ScilabBridge.draw(Unknown
> > Source)
> >        at org.scilab.modules.gui.tab.ScilabTab.draw(Unknown Source)
> >        at
> >
> org.scilab.modules.gui.graphicWindow.ScilabRendererProperties.forceDi
> > splay(Unknown Source)
> >        at
> >
> org.scilab.modules.renderer.figureDrawing.DrawableFigureGL.drawCanvas
> > (Unknown Source)
> > Caused by: java.lang.NoClassDefFoundError: Could not initialize
> class
> > com.sun.op
> > engl.impl.x11.X11GLDrawableFactory
> > 
> > This class com.sun.opengl.impl.x11.X11GLDrawableFactory exists in
> the
> > gluegen-rt.jar and is in the classpath.
> > Also all the libgluegen-rt.so and related libraries are in the
> > LD_LIBRARY_PATH and ldconfig has been run many times.
> > 
> > How can I correct this.
> > Many thanks,
> >  
> > Rob Key
> > 
> > 
> > 
> > 
> > 
> 
> 
> 


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20081012/4a7d4b20/attachment.htm>

From antonelli at unicas.it  Mon Oct 13 10:31:34 2008
From: antonelli at unicas.it (ga)
Date: Mon, 13 Oct 2008 01:31:34 -0700 (PDT)
Subject: Scicos problem in Scilab 5.0.x
Message-ID: <66ca2523-3f18-4f6b-8075-50ba612d1d04@d70g2000hsc.googlegroups.com>

Hi,

I'm experiencing some problems with Scicos under Scilab 5.0.x for WXP
SP3.

- Pal Tree command is not working. The new window appears with the
blocks name but without icons. When I try to drag the block's name
nothing happens;

- One of the new features of Scilab 5 should have been the possibility
to use the Scilab's shell when Scicos is running. This is not possible
in my case. I have the prompt and can write a command (such as <a=2>)
but it simply frozen when entering [RETURN] still keeping Scicos
working;

Not specific to Scicos:

- The fonts' rendering of any graphic windows is poor;

In addition, I'm still not able to use any graphic window under Linux
with an Intel graphic card and I did not get useful replies other than
to disable the hardware accelerator.

Best,
g.




From simone.mannori at scilab.org  Mon Oct 13 10:45:31 2008
From: simone.mannori at scilab.org (Simone Mannori)
Date: Mon, 13 Oct 2008 10:45:31 +0200
Subject: [scilab-Users] Scicos problem in Scilab 5.0.x
In-Reply-To: <66ca2523-3f18-4f6b-8075-50ba612d1d04@d70g2000hsc.googlegroups.com>
References: 	 <66ca2523-3f18-4f6b-8075-50ba612d1d04@d70g2000hsc.googlegroups.com>
Message-ID: <1223887532.3080.13.camel@pinarellu.inria.fr>

Bonjour, 

On Mon, 2008-10-13 at 01:31 -0700, ga wrote:
> Hi,
> 
> I'm experiencing some problems with Scicos under Scilab 5.0.x for WXP
> SP3.
> 
> - Pal Tree command is not working. The new window appears with the
> blocks name but without icons. When I try to drag the block's name
> nothing happens;
> 
because there is no TCL/TK support for xclick/xgetmouse Scilab
primitives. The TCL/TK windows are shown but there are not active. 

> - One of the new features of Scilab 5 should have been the possibility
> to use the Scilab's shell when Scicos is running. This is not possible
> in my case. I have the prompt and can write a command (such as <a=2>)
> but it simply frozen when entering [RETURN] still keeping Scicos
> working;
> 

As above: seteventhandler is bugged/not fully functional inside Scilab
5 


> Not specific to Scicos:
> 
> - The fonts' rendering of any graphic windows is poor;
> 

probably some fonts are not installed on your machine ?


> In addition, I'm still not able to use any graphic window under Linux
> with an Intel graphic card and I did not get useful replies other than
> to disable the hardware accelerator.

Unfortunately, Scilab 5.0.x does not offer support for Intel integrated
cards under Linux.

We are actively working on a solution for these problems for the next
Scilab 5.1.x

May I suggest to post some bug reports using Bugzilla ?
http://bugzilla.scilab.org/

Thanks for using Scilab :)


Simone Mannori - The Scilab Consortium 
Scilab/Scicos Embedded Applications

//**-------------------------------------------------------------------------





From sylvestre.ledru at scilab.org  Mon Oct 13 11:18:18 2008
From: sylvestre.ledru at scilab.org (Sylvestre Ledru)
Date: Mon, 13 Oct 2008 11:18:18 +0200
Subject: [scilab-Users] Scicos problem in Scilab 5.0.x
In-Reply-To: <1223887532.3080.13.camel@pinarellu.inria.fr>
References: 	 <66ca2523-3f18-4f6b-8075-50ba612d1d04@d70g2000hsc.googlegroups.com>
	 <1223887532.3080.13.camel@pinarellu.inria.fr>
Message-ID: <1223889498.18230.48.camel@korcula.inria.fr>


> 
> > Not specific to Scicos:
> > 
> > - The fonts' rendering of any graphic windows is poor;
Could you send a screenshot of this ?

> 

> Unfortunately, Scilab 5.0.x does not offer support for Intel integrated
> cards under Linux.
> 
> We are actively working on a solution for these problems for the next
> Scilab 5.1.x
> 
> May I suggest to post some bug reports using Bugzilla ?
> http://bugzilla.scilab.org/
Well, if it is to report that the graphic doesn't work, don't waste your
time on this. We are most aware of this! (cf bug #3525) ;)

Sylvestre




From simone.mannori at scilab.org  Mon Oct 13 11:41:41 2008
From: simone.mannori at scilab.org (Simone Mannori)
Date: Mon, 13 Oct 2008 11:41:41 +0200
Subject: [scilab-Users] Scicos problem in Scilab 5.0.x
In-Reply-To: <66ca2523-3f18-4f6b-8075-50ba612d1d04@d70g2000hsc.googlegroups.com>
References: 	 <66ca2523-3f18-4f6b-8075-50ba612d1d04@d70g2000hsc.googlegroups.com>
Message-ID: <1223890901.3080.19.camel@pinarellu.inria.fr>

Hi,

> - One of the new features of Scilab 5 should have been the possibility
> to use the Scilab's shell when Scicos is running. This is not possible
> in my case. I have the prompt and can write a command (such as <a=2>)
> but it simply frozen when entering [RETURN] still keeping Scicos
> working;
> 

about this point my previous repose was _WRONG_ :

Scilab 5 console is "active" but is not able to compute while Scicos is
running: basically there are no changes between Scilab 4.x and 5.x

Please uses "Tools"-"Calc" to switch between the two environment.

"Faro' piu' attenzione la prossima volta".

Simone Mannori






From sylvestre.ledru at scilab.org  Mon Oct 13 15:19:24 2008
From: sylvestre.ledru at scilab.org (Sylvestre Ledru)
Date: Mon, 13 Oct 2008 15:19:24 +0200
Subject: Update of the Wiki [Was: SOLVED: plots are still blank]
In-Reply-To: <524161.64031.qm@web45615.mail.sp1.yahoo.com>
References: <524161.64031.qm@web45615.mail.sp1.yahoo.com>
Message-ID: <1223903964.18230.91.camel@korcula.inria.fr>

Hello Rob,

Thanks for this! It was mainly OK. I added some extra information about
the SCI/etc/classpath.xml which can be used to set the path to the jar
libs and also that *.so can be set thanks to the LD_LIBRARY_PATH
variable.
I also did some cosmetic changes! 
Don't hesitate to update other pages on the wiki.

Thanks again.
Cheers,
Sylvestre

Le dimanche 12 octobre 2008 ? 07:54 -0700, Robert Key a ?crit :
> Hi Sylvestre,
>  I have updated the FAQ. I hope the wording is ok. I have included a
> little more infomation.
> Cheers,
>  
> Rob Key
> 
> 
> 
> ----- Original Message ----
> From: Sylvestre Ledru <sylvestre.ledru at scilab.org>
> To: users at lists..scilab.org
> Sent: Sunday, October 12, 2008 1:13:25 AM
> Subject: Re: [scilab-Users] SOLVED: plots are still blank
> 
> Hello Rob,
> 
> Actually, it is pretty easy.
> 1) create an account the wiki
> http://wiki.scilab.org/UserPreferences
> 2) go the to the page
> http://wiki.scilab.org/Compiling_Scilab_5.x_under_GNU-Linux_Unix_-_FAQ
> 3) edit the page
> 4) update the content (basically, copy/paste an example and modify it)
> 
> Don't worry to break things. Each page is versionned and it is
> straighforward to revert a bad modification/mistakes.
> 
> Sylvestre
> 
> Le samedi 11 octobre 2008 ? 10:35 -0700, Robert Key a ?crit :
> > Hi Sylvestre I have looked at the url for the faq. I don't know how
> to
> > update the faq. Do I need to register to that mailing list?
> > Please let me know how to update the faq and I will gladly do it.
> > Thanks,
> >  Robert Key
> >  
> > Rob Key
> > 
> > 
> > 
> > ----- Original Message ----
> > From: Sylvestre Ledru <sylvestre.ledru at scilab.org>
> > To: users at lists.scilab.org
> > Sent: Friday, October 10, 2008 10:41:30 AM
> > Subject: Re: [scilab-Users] SOLVED: plots are still blank
> > 
> > Thanks for the feedback. It is much appreciated.
> > 
> > Would you mind updating this wiki page:
> >
> http://wiki.scilab..org/Compiling_Scilab_5.x_under_GNU-Linux_Unix_-_FAQ
> > 
> > with these information ?
> > 
> > Cheers,
> > Sylvestre
> > 
> > 
> > Le vendredi 10 octobre 2008 ? 01:38 -0700, Robert Key a ?crit :
> > > Hi I have managed to solve my own plotting problem. Here is the
> > > solution in case anyone has the same problem:
> > > 
> > > I placed all run time libraries required by scilab
> > > in /opt/jdk1.6.0_07/jre/lib/amd64 (this would be i386 on 32 bit
> > linux)
> > > such as libgluegen-rt.so and all the jogl*.so. Somehow they were
> not
> > > found when placed in  /usr/local/lib and ldconfig then run.
> > > 
> > > I placed all the gluegen-rt.jar and jogl*.jar class libraries
> > > in /opt/jdk1.6.0_07/jre/lib/ext. If you do this then it is not
> > > necessary to include them in your CLASSPATH variable. I tried
> > putting
> > > then in a directory and including that in the CLASSPATH but this
> did
> > > not work for some reason.
> > > 
> > > Now my plots and sciocos all work fantastically on linux x86_64!!!
> > > Thanks for a great scientific software application.
> > > Cheers and happy computing,
> > >  Robert Key
> > >  
> > > Rob Key
> > > 
> > > 
> > > 
> > > ----- Original Message ----
> > > From: Robert Key <robertgordonkey at yahoo.com>
> > > To: users at lists.scilab.org
> > > Sent: Thursday, October 9, 2008 2:05:41 PM
> > > Subject: [scilab-Users] plots are still blank
> > > 
> > > Hi all,
> > >  I use scilab 5.0.2 on linux x86_64. jogl 64 bit is installed and
> > > LD_LIBRARY_PATH and the CLASSPATH are set up correctly but I still
> > get
> > > the error:
> > > 
> > > java.lang.reflect.InvocationTargetException
> > >        at java.awt.EventQueue.invokeAndWait(EventQueue.java:997)
> > >        at
> > >
> javax.swing.SwingUtilities..invokeAndWait(SwingUtilities.java:1323)
> > >        at
> > > org.scilab.modules.gui.bridge.tab.SwingScilabTab.draw(Unknown
> > Source)
> > >        at org.scilab.modules.gui.tab.ScilabTabBridge.draw(Unknown
> > > Source)
> > >        at org.scilab.modules.gui.bridge.ScilabBridge.draw(Unknown
> > > Source)
> > >        at org.scilab.modules.gui.tab.ScilabTab.draw(Unknown
> Source)
> > >        at
> > >
> >
> org.scilab.modules.gui.graphicWindow.ScilabRendererProperties.forceDi
> > > splay(Unknown Source)
> > >        at
> > >
> >
> org.scilab.modules.renderer.figureDrawing.DrawableFigureGL.drawCanvas
> > > (Unknown Source)
> > > Caused by: java.lang.NoClassDefFoundError: Could not initialize
> > class
> > > com.sun..op
> > > engl.impl.x11.X11GLDrawableFactory
> > > 
> > > This class com.sun.opengl.impl.x11.X11GLDrawableFactory exists in
> > the
> > > gluegen-rt.jar and is in the classpath.
> > > Also all the libgluegen-rt.so and related libraries are in the
> > > LD_LIBRARY_PATH and ldconfig has been run many times.
> > > 
> > > How can I correct this.
> > > Many thanks,
> > >  
> > > Rob Key
> > > 
> > > 
> > > 
> > > 
> > > 
> > 
> > 
> > 
> 
> 
> 
> 



From sylvestre.ledru at scilab.org  Mon Oct 13 16:28:54 2008
From: sylvestre.ledru at scilab.org (Sylvestre Ledru)
Date: Mon, 13 Oct 2008 16:28:54 +0200
Subject: Problem with the graphic in Scilab 5.0
Message-ID: <1223908134.18230.115.camel@korcula.inria.fr>

Hello,

I am writing a page on the Scilab Wiki trying to explain where the
problem comes from, where it happens and how to fix it
(thanks to a crappy workaround).
The URL is the following:
http://wiki.scilab.org/Graphical_issues_with_Scilab_5.0

Don't hesitate to report other workarounds or ideas.

Sylvestre




From antonelli at unicas.it  Mon Oct 13 16:43:59 2008
From: antonelli at unicas.it (Gianluca Antonelli)
Date: Mon, 13 Oct 2008 16:43:59 +0200
Subject: [scilab-Users] Scicos problem in Scilab 5.0.x
In-Reply-To: <1223889498.18230.48.camel@korcula.inria.fr>
References: <66ca2523-3f18-4f6b-8075-50ba612d1d04@d70g2000hsc.googlegroups.com>	 <1223887532.3080.13.camel@pinarellu.inria.fr> <1223889498.18230.48.camel@korcula.inria.fr>
Message-ID: <48F35EAF.2080107@unicas.it>



Sylvestre Ledru wrote:
>>> Not specific to Scicos:
>>>
>>> - The fonts' rendering of any graphic windows is poor;
> Could you send a screenshot of this ?

here we are... (hope the mailing list support attachments).

the figure has been obtained with:

plot2d;xtitle('test of poor rendering');

what is strange is that the original figure looks fine and the poor 
rendering appears by re-sizing it.

just as a remind: wxp sp 3, scilab 5.0.x

best,
g.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: test.jpg
Type: image/jpeg
Size: 92096 bytes
Desc: not available
URL: <https://lists.scilab.org/pipermail/users/attachments/20081013/6d77da85/attachment.jpg>

From allan.cornet at scilab.org  Mon Oct 13 17:00:45 2008
From: allan.cornet at scilab.org (Allan CORNET)
Date: Mon, 13 Oct 2008 17:00:45 +0200
Subject: [scilab-Users] Scicos problem in Scilab 5.0.x
In-Reply-To: <48F35EAF.2080107@unicas.it>
References: <66ca2523-3f18-4f6b-8075-50ba612d1d04@d70g2000hsc.googlegroups.com>	 <1223887532.3080.13.camel@pinarellu.inria.fr> <1223889498.18230.48.camel@korcula.inria.fr> <48F35EAF.2080107@unicas.it>
Message-ID: <000301c92d44$78826f90$69874eb0$@cornet@scilab.org>

Hi,

plot2d;xtitle('test of poor rendering '+getversion());

It seems to work on my Vista.
I resize and capture this screenshot.

Best Regards

Allan CORNET


-----Message d'origine-----
De?: Gianluca Antonelli [mailto:antonelli at unicas.it] 
Envoy??: lundi 13 octobre 2008 16:44
??: users at lists.scilab.org
Objet?: Re: [scilab-Users] Scicos problem in Scilab 5.0.x



Sylvestre Ledru wrote:
>>> Not specific to Scicos:
>>>
>>> - The fonts' rendering of any graphic windows is poor;
> Could you send a screenshot of this ?

here we are... (hope the mailing list support attachments).

the figure has been obtained with:

plot2d;xtitle('test of poor rendering');

what is strange is that the original figure looks fine and the poor
rendering appears by re-sizing it.

just as a remind: wxp sp 3, scilab 5.0.x

best,
g.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: plot2d.JPG
Type: image/jpeg
Size: 156872 bytes
Desc: not available
URL: <https://lists.scilab.org/pipermail/users/attachments/20081013/c98972b8/attachment.jpe>

From cd at dittmann-arheilgen.de  Mon Oct 13 23:32:56 2008
From: cd at dittmann-arheilgen.de (Christoph Dittmann)
Date: Mon, 13 Oct 2008 23:32:56 +0200
Subject: Inexplicable result with scilab 5.0.2 and matrix multiplication
Message-ID: <48F3BE88.7030603@dittmann-arheilgen.de>

Hi,

I've got a problem with scilab 5.0.2. It seems to print certain values 
wrong. The following example shows the problem:
-->M = [1,2;3,2]; C = M^(-1)*M
  C  =

     2.           1.110D-16
     1.110D-16    2.

This is what I get with scilab 5.0.2 on Ubuntu 8.04 (x86) using the 
precompiled version from http://www.scilab.org/download/ . My problem 
are the 2's on the diagonal since I've no idea where they come from.

With scilab 4.1.2, the one from the Ubuntu 8.04 repository, I get what I 
expect:
-->M = [1,2;3,2]; C = M^(-1)*M
  C  =

     1.           1.110D-16
     5.551D-17    1.


Some further calculations with scilab 5.0.2 show that the problem seems 
to lie in the printing of the result and not its actual numerical value:
-->x = C(1,1)
  x  =

     2.
-->x * 2
  ans =

    2.
-->x + 0.01
  ans  =

     1.01


Could that be a bug or did I just miss something crucial?


Thanks,
Christoph


From sylvestre.ledru at inria.fr  Mon Oct 13 23:40:38 2008
From: sylvestre.ledru at inria.fr (Sylvestre Ledru)
Date: Mon, 13 Oct 2008 23:40:38 +0200
Subject: [scilab-Users] Inexplicable result with scilab 5.0.2 and
	matrix multiplication
In-Reply-To: <48F3BE88.7030603@dittmann-arheilgen.de>
References: <48F3BE88.7030603@dittmann-arheilgen.de>
Message-ID: <1223934038.3427.54.camel@zlarin>

Hello,

It is most probably a bug in the printing functions... I can reproduce
it with the binary (ie not compiled on my computer) but I can't on my
dev version (compiled with gfortran 4.3.1)
Once, we had a simular bug caused by a bug in gfortran. Maybe it is a
regression.

Thanks for pointing this out.
Sylvestre

Le lundi 13 octobre 2008 ? 23:32 +0200, Christoph Dittmann a ?crit :
> Hi,
> 
> I've got a problem with scilab 5.0.2. It seems to print certain values 
> wrong. The following example shows the problem:
> -->M = [1,2;3,2]; C = M^(-1)*M
>   C  =
> 
>      2.           1.110D-16
>      1.110D-16    2.
> 
> This is what I get with scilab 5.0.2 on Ubuntu 8.04 (x86) using the 
> precompiled version from http://www.scilab.org/download/ . My problem 
> are the 2's on the diagonal since I've no idea where they come from.
> 
> With scilab 4.1.2, the one from the Ubuntu 8.04 repository, I get what I 
> expect:
> -->M = [1,2;3,2]; C = M^(-1)*M
>   C  =
> 
>      1.           1.110D-16
>      5.551D-17    1.
> 
> 
> Some further calculations with scilab 5.0.2 show that the problem seems 
> to lie in the printing of the result and not its actual numerical value:
> -->x = C(1,1)
>   x  =
> 
>      2.
> -->x * 2
>   ans =
> 
>     2.
> -->x + 0.01
>   ans  =
> 
>      1.01
> 
> 
> Could that be a bug or did I just miss something crucial?
> 
> 
> Thanks,
> Christoph



From fvogelnew1 at free.fr  Mon Oct 13 23:48:28 2008
From: fvogelnew1 at free.fr (=?UTF-8?B?RnJhbsOnb2lzIFZvZ2Vs?=)
Date: Mon, 13 Oct 2008 23:48:28 +0200
Subject: [scilab-Users] Inexplicable result with scilab 5.0.2 and	matrix
 multiplication
In-Reply-To: <1223934038.3427.54.camel@zlarin>
References: <48F3BE88.7030603@dittmann-arheilgen.de> <1223934038.3427.54.camel@zlarin>
Message-ID: <48F3C22C.20606@free.fr>

Looks like this one:

http://bugzilla.scilab.org/show_bug.cgi?id=2647

Francois


Sylvestre Ledru said on 13/10/2008 23:40:
> Hello,
> 
> It is most probably a bug in the printing functions... I can reproduce
> it with the binary (ie not compiled on my computer) but I can't on my
> dev version (compiled with gfortran 4.3.1)
> Once, we had a simular bug caused by a bug in gfortran. Maybe it is a
> regression.
> 
> Thanks for pointing this out.
> Sylvestre
> 
> Le lundi 13 octobre 2008 ? 23:32 +0200, Christoph Dittmann a ?crit :
>> Hi,
>>
>> I've got a problem with scilab 5.0.2. It seems to print certain values 
>> wrong. The following example shows the problem:
>> -->M = [1,2;3,2]; C = M^(-1)*M
>>   C  =
>>
>>      2.           1.110D-16
>>      1.110D-16    2.
>>
>> This is what I get with scilab 5.0.2 on Ubuntu 8.04 (x86) using the 
>> precompiled version from http://www.scilab.org/download/ . My problem 
>> are the 2's on the diagonal since I've no idea where they come from.
>>
>> With scilab 4.1.2, the one from the Ubuntu 8.04 repository, I get what I 
>> expect:
>> -->M = [1,2;3,2]; C = M^(-1)*M
>>   C  =
>>
>>      1.           1.110D-16
>>      5.551D-17    1.
>>
>>
>> Some further calculations with scilab 5.0.2 show that the problem seems 
>> to lie in the printing of the result and not its actual numerical value:
>> -->x = C(1,1)
>>   x  =
>>
>>      2.
>> -->x * 2
>>   ans =
>>
>>     2.
>> -->x + 0.01
>>   ans  =
>>
>>      1.01
>>
>>
>> Could that be a bug or did I just miss something crucial?
>>
>>
>> Thanks,
>> Christoph
> 
> 
> 


From sylvestre.ledru at inria.fr  Mon Oct 13 23:52:23 2008
From: sylvestre.ledru at inria.fr (Sylvestre Ledru)
Date: Mon, 13 Oct 2008 23:52:23 +0200
Subject: [scilab-Users] Inexplicable result with scilab 5.0.2
	and	matrix multiplication
In-Reply-To: <48F3C22C.20606@free.fr>
References: <48F3BE88.7030603@dittmann-arheilgen.de>
	 <1223934038.3427.54.camel@zlarin>  <48F3C22C.20606@free.fr>
Message-ID: <1223934743.3427.67.camel@zlarin>

Grr, I forgot that bug. Thanks for digging it up.

Sylvestre

Le lundi 13 octobre 2008 ? 23:48 +0200, Fran?ois Vogel a ?crit :
> Looks like this one:
> 
> http://bugzilla.scilab.org/show_bug.cgi?id=2647
> 
> Francois
> 
> 
> Sylvestre Ledru said on 13/10/2008 23:40:
> > Hello,
> > 
> > It is most probably a bug in the printing functions... I can reproduce
> > it with the binary (ie not compiled on my computer) but I can't on my
> > dev version (compiled with gfortran 4.3.1)
> > Once, we had a simular bug caused by a bug in gfortran. Maybe it is a
> > regression.
> > 
> > Thanks for pointing this out.
> > Sylvestre
> > 
> > Le lundi 13 octobre 2008 ? 23:32 +0200, Christoph Dittmann a ?crit :
> >> Hi,
> >>
> >> I've got a problem with scilab 5.0.2. It seems to print certain values 
> >> wrong. The following example shows the problem:
> >> -->M = [1,2;3,2]; C = M^(-1)*M
> >>   C  =
> >>
> >>      2.           1.110D-16
> >>      1.110D-16    2.
> >>
> >> This is what I get with scilab 5.0.2 on Ubuntu 8.04 (x86) using the 
> >> precompiled version from http://www.scilab.org/download/ . My problem 
> >> are the 2's on the diagonal since I've no idea where they come from.
> >>
> >> With scilab 4.1.2, the one from the Ubuntu 8.04 repository, I get what I 
> >> expect:
> >> -->M = [1,2;3,2]; C = M^(-1)*M
> >>   C  =
> >>
> >>      1.           1.110D-16
> >>      5.551D-17    1.
> >>
> >>
> >> Some further calculations with scilab 5.0.2 show that the problem seems 
> >> to lie in the printing of the result and not its actual numerical value:
> >> -->x = C(1,1)
> >>   x  =
> >>
> >>      2.
> >> -->x * 2
> >>   ans =
> >>
> >>     2.
> >> -->x + 0.01
> >>   ans  =
> >>
> >>      1.01
> >>
> >>
> >> Could that be a bug or did I just miss something crucial?
> >>
> >>
> >> Thanks,
> >> Christoph
> > 
> > 
> > 



From sylvestre.ledru at scilab.org  Wed Oct 15 13:46:05 2008
From: sylvestre.ledru at scilab.org (Sylvestre Ledru)
Date: Wed, 15 Oct 2008 13:46:05 +0200
Subject: [scilab-Users] Inexplicable result with scilab 5.0.2
	and	matrix multiplication
In-Reply-To: <48F3C22C.20606@free.fr>
References: <48F3BE88.7030603@dittmann-arheilgen.de>
	 <1223934038.3427.54.camel@zlarin>  <48F3C22C.20606@free.fr>
Message-ID: <1224071165.18336.83.camel@korcula.inria.fr>

For those who are interested in this bug, the discussion continues in
the bugtracker.
(You can add your email address to the bug report to receive updates).

Sylvestre

Le lundi 13 octobre 2008 ? 23:48 +0200, Fran?ois Vogel a ?crit : 
> Looks like this one:
> 
> http://bugzilla.scilab.org/show_bug.cgi?id=2647
> 
> Francois
> 
> 
> Sylvestre Ledru said on 13/10/2008 23:40:
> > Hello,
> > 
> > It is most probably a bug in the printing functions... I can reproduce
> > it with the binary (ie not compiled on my computer) but I can't on my
> > dev version (compiled with gfortran 4.3.1)
> > Once, we had a simular bug caused by a bug in gfortran. Maybe it is a
> > regression.
> > 
> > Thanks for pointing this out.
> > Sylvestre
> > 
> > Le lundi 13 octobre 2008 ? 23:32 +0200, Christoph Dittmann a ?crit :
> >> Hi,
> >>
> >> I've got a problem with scilab 5.0.2. It seems to print certain values 
> >> wrong. The following example shows the problem:
> >> -->M = [1,2;3,2]; C = M^(-1)*M
> >>   C  =
> >>
> >>      2.           1.110D-16
> >>      1.110D-16    2.
> >>
> >> This is what I get with scilab 5.0.2 on Ubuntu 8.04 (x86) using the 
> >> precompiled version from http://www.scilab.org/download/ . My problem 
> >> are the 2's on the diagonal since I've no idea where they come from.
> >>
> >> With scilab 4.1.2, the one from the Ubuntu 8.04 repository, I get what I 
> >> expect:
> >> -->M = [1,2;3,2]; C = M^(-1)*M
> >>   C  =
> >>
> >>      1.           1.110D-16
> >>      5.551D-17    1.
> >>
> >>
> >> Some further calculations with scilab 5.0.2 show that the problem seems 
> >> to lie in the printing of the result and not its actual numerical value:
> >> -->x = C(1,1)
> >>   x  =
> >>
> >>      2.
> >> -->x * 2
> >>   ans =
> >>
> >>     2.
> >> -->x + 0.01
> >>   ans  =
> >>
> >>      1.01
> >>
> >>
> >> Could that be a bug or did I just miss something crucial?
> >>
> >>
> >> Thanks,
> >> Christoph
> > 
> > 
> > cd 



From UVV-mail at yandex.ru  Thu Oct 16 14:44:12 2008
From: UVV-mail at yandex.ru (Vyacheslav V. Yurkov)
Date: Thu, 16 Oct 2008 20:44:12 +0800
Subject: [scilab-Users] graphics speed from 4.1.2 to 5.0.2 and virtual
        graphical window
In-Reply-To: <20081001155148.GA16033@delacroix.inrialpes.fr>
References: <e578e190810010808jac39e7enf326a854053b32f2@mail.gmail.com> <faf3083f0810010829w3b5497c6na90117901c58ce8e@mail.gmail.com> <20081001155148.GA16033@delacroix.inrialpes.fr>
Message-ID: <48F7371C.6010201@yandex.ru>

marc.fuentes at inrialpes.fr wrote:
> Indeed, scilab wants to mimick matlab using java, and as a consequence
> it becomes very slow. In matlab using -nojvm enables matlab to go faster
> but does not prevent plotting graphics.
> But with scilab 5.0.2 , disabling java support at compile time prevents you to plot
> graphics(you must compile with option "--without-gui")
> The solution : go on with scilab-4.1.2!
>   
So will it be faster?
Or people who plots graphics should stay in scilab-4.1.2 and don't wait 
better solution?

-- 
Vyacheslav V. Yurkov <uvv.mail at gmail.com>	Process Control Engineer
Norilsk						Tel: +79069001870

Department of process control systems
Open Joint Stock Company MMC Norilsk Nickel, brass works

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20081016/15004771/attachment.htm>

From sylvestre.ledru at scilab.org  Thu Oct 16 14:55:36 2008
From: sylvestre.ledru at scilab.org (Sylvestre Ledru)
Date: Thu, 16 Oct 2008 14:55:36 +0200
Subject: [scilab-Users] graphics speed from 4.1.2 to 5.0.2 and virtual
	graphical window
In-Reply-To: <48F7371C.6010201@yandex.ru>
References: <e578e190810010808jac39e7enf326a854053b32f2@mail.gmail.com>
	 <faf3083f0810010829w3b5497c6na90117901c58ce8e@mail.gmail.com>
	 <20081001155148.GA16033@delacroix.inrialpes.fr>
	 <48F7371C.6010201@yandex.ru>
Message-ID: <1224161736.18336.185.camel@korcula.inria.fr>

Le jeudi 16 octobre 2008 ? 20:44 +0800, Vyacheslav V. Yurkov a ?crit :
> marc.fuentes at inrialpes.fr wrote: 
> > Indeed, scilab wants to mimick matlab using java, and as a consequence
> > it becomes very slow. In matlab using -nojvm enables matlab to go faster
> > but does not prevent plotting graphics.
> > But with scilab 5.0.2 , disabling java support at compile time prevents you to plot
> > graphics(you must compile with option "--without-gui")
> > The solution : go on with scilab-4.1.2!
> >   
> So will it be faster?
Depends your needs, your OS and your hardware.

Under Windows and Linux with Nvidia (and recent ATI drivers), the
performance are good enough in most situations.
(don't forget that the first access to a graphic feature is slow due to
the load of the libs).

And in the computing engine, Scilab 5 family is much faster than the 4.

Sylvestre




From UVV-mail at yandex.ru  Thu Oct 16 15:12:34 2008
From: UVV-mail at yandex.ru (Vyacheslav V. Yurkov)
Date: Thu, 16 Oct 2008 21:12:34 +0800
Subject: [scilab-Users] graphics speed from 4.1.2 to 5.0.2 and virtual
 graphical window
In-Reply-To: <1224161736.18336.185.camel@korcula.inria.fr>
References: <e578e190810010808jac39e7enf326a854053b32f2@mail.gmail.com>  <faf3083f0810010829w3b5497c6na90117901c58ce8e@mail.gmail.com>   <20081001155148.GA16033@delacroix.inrialpes.fr>         <48F7371C.6010201@yandex.ru> <1224161736.18336.185.camel@korcula.inria.fr>
Message-ID: <48F73DC2.0@yandex.ru>

Sylvestre Ledru wrote:
> Le jeudi 16 octobre 2008 ? 20:44 +0800, Vyacheslav V. Yurkov a ?crit :
>   
> Depends your needs, your OS and your hardware.
>
> Under Windows and Linux with Nvidia (and recent ATI drivers), the
> performance are good enough in most situations.
> (don't forget that the first access to a graphic feature is slow due to
> the load of the libs).
>
> And in the computing engine, Scilab 5 family is much faster than the 4.
>
> Sylvestr
I'm using Slackware Linux 12.1 with nvidia drivers version x86-96.43.07.
Hardware: Pentium-4 2.8 GHz, 512 Mb memory, video card - MX440.
Even button "stop" don't respond to pushing on it during simulation for 
one-two minutes.

-- 
Vyacheslav V. Yurkov <uvv.mail at gmail.com>	Process Control Engineer
Norilsk						Tel: +79069001870

Department of process control systems
Open Joint Stock Company MMC Norilsk Nickel, brass works

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20081016/b95ba32d/attachment.htm>

From sylvestre.ledru at scilab.org  Thu Oct 16 15:14:34 2008
From: sylvestre.ledru at scilab.org (Sylvestre Ledru)
Date: Thu, 16 Oct 2008 15:14:34 +0200
Subject: [scilab-Users] graphics speed from 4.1.2 to 5.0.2 and virtual
	graphical window
In-Reply-To: <48F73DC2.0@yandex.ru>
References: <e578e190810010808jac39e7enf326a854053b32f2@mail.gmail.com>
	 <faf3083f0810010829w3b5497c6na90117901c58ce8e@mail.gmail.com>
	 <20081001155148.GA16033@delacroix.inrialpes.fr>
	 <48F7371C.6010201@yandex.ru> <1224161736.18336.185.camel@korcula.inria.fr>
	  <48F73DC2.0@yandex.ru>
Message-ID: <1224162874.18336.191.camel@korcula.inria.fr>

Le jeudi 16 octobre 2008 ? 21:12 +0800, Vyacheslav V. Yurkov a ?crit :
> Sylvestre Ledru wrote: 
> > Le jeudi 16 octobre 2008 ? 20:44 +0800, Vyacheslav V. Yurkov a ?crit :
> >   
> > Depends your needs, your OS and your hardware.
> > 
> > Under Windows and Linux with Nvidia (and recent ATI drivers), the
> > performance are good enough in most situations.
> > (don't forget that the first access to a graphic feature is slow due to
> > the load of the libs).
> > 
> > And in the computing engine, Scilab 5 family is much faster than the 4.
> > 
> > Sylvestr
> I'm using Slackware Linux 12.1 with nvidia drivers version
> x86-96.43.07.
> Hardware: Pentium-4 2.8 GHz, 512 Mb memory, video card - MX440.
Do you know if the acceleration of your X server is enable ?


> Even button "stop" don't respond to pushing on it during simulation
> for one-two minutes.
That is indeed an anormal issue.

Sylvestre




From Francis.Drossaert at pgs.com  Thu Oct 16 15:17:09 2008
From: Francis.Drossaert at pgs.com (Francis Drossaert)
Date: Thu, 16 Oct 2008 14:17:09 +0100
Subject: [scilab-Users] graphics speed from 4.1.2 to 5.0.2 and virtual graphical window
In-Reply-To: <48F73DC2.0@yandex.ru>
References: <e578e190810010808jac39e7enf326a854053b32f2@mail.gmail.com>  <faf3083f0810010829w3b5497c6na90117901c58ce8e@mail.gmail.com>   <20081001155148.GA16033@delacroix.inrialpes.fr>         <48F7371C.6010201@yandex.ru> <1224161736.18336.185.camel@korcula.inria.fr> <48F73DC2.0@yandex.ru>
Message-ID: <6D1269A4C2D21240807B91421B187255068DF546@lonms21.onshore.pgs.com>

I have not noticed any slowing down in Scilab other than the plotting of the results in a figure. I am doing fairly intense calculations e.g. fft of hundreds of columns in a matrix. All in all I am satisfied. Copy and past works now (very very useful for me) and the auto fill in is great to. The console looks better and more pleasing to the eye.
 

________________________________

From: Vyacheslav V. Yurkov [mailto:UVV-mail at yandex.ru] 
Sent: 16 October 2008 14:13
To: users at lists.scilab.org
Subject: Re: [scilab-Users] graphics speed from 4.1.2 to 5.0.2 and virtual graphical window


Sylvestre Ledru wrote: 

	Le jeudi 16 octobre 2008 ? 20:44 +0800, Vyacheslav V. Yurkov a ?crit :
	  
	Depends your needs, your OS and your hardware.
	
	Under Windows and Linux with Nvidia (and recent ATI drivers), the
	performance are good enough in most situations.
	(don't forget that the first access to a graphic feature is slow due to
	the load of the libs).
	
	And in the computing engine, Scilab 5 family is much faster than the 4.
	
	Sylvestr

I'm using Slackware Linux 12.1 with nvidia drivers version x86-96.43.07.
Hardware: Pentium-4 2.8 GHz, 512 Mb memory, video card - MX440.
Even button "stop" don't respond to pushing on it during simulation for one-two minutes.

-- 
Vyacheslav V. Yurkov <uvv.mail at gmail.com> <mailto:uvv.mail at gmail.com> 	Process Control Engineer
Norilsk						Tel: +79069001870

Department of process control systems
Open Joint Stock Company MMC Norilsk Nickel, brass works

This email and any files contained therein is confidential and may contain privileged information.  If you are not the named addressee(s) or you have otherwise received this in error, you should not distribute or copy this e-mail or use any of its content for any purpose. Please notify the sender immediately by e-mail if you have received this e-mail in error and delete it from your system
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20081016/f29ae968/attachment.htm>

From UVV-mail at yandex.ru  Thu Oct 16 15:39:03 2008
From: UVV-mail at yandex.ru (Vyacheslav V. Yurkov)
Date: Thu, 16 Oct 2008 21:39:03 +0800
Subject: [scilab-Users] graphics speed from 4.1.2 to 5.0.2 and virtual
 graphical window
In-Reply-To: <1224162874.18336.191.camel@korcula.inria.fr>
References: <e578e190810010808jac39e7enf326a854053b32f2@mail.gmail.com>  <faf3083f0810010829w3b5497c6na90117901c58ce8e@mail.gmail.com>   <20081001155148.GA16033@delacroix.inrialpes.fr>         <48F7371C.6010201@yandex.ru> <1224161736.18336.185.camel@korcula.inria.fr>       <48F73DC2.0@yandex.ru> <1224162874.18336.191.camel@korcula.inria.fr>
Message-ID: <48F743F7.3010204@yandex.ru>

Sylvestre Ledru wrote:
> Le jeudi 16 octobre 2008 ? 21:12 +0800, Vyacheslav V. Yurkov a ?crit :
>   
> Do you know if the acceleration of your X server is enable ?
>
>   
Of course. Sometimes I play games using 3D acceleration.
>> Even button "stop" don't respond to pushing on it during simulation
>> for one-two minutes.
>>     
> That is indeed an anormal issue.
>   
Maybe it would be useful: I use jre-6u6.
Sorry for disinformation but it happens not with all models. Only with 
my model I've enclosed.

2 Francis Drossaert: I'm talking about graphics in scicos emulation.

-- 
Vyacheslav V. Yurkov <uvv.mail at gmail.com>	Process Control Engineer
Norilsk						Tel: +79069001870

Department of process control systems
Open Joint Stock Company MMC Norilsk Nickel, brass works

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20081016/b5e8f2a9/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: model.cos
Type: application/octet-stream
Size: 318880 bytes
Desc: not available
URL: <https://lists.scilab.org/pipermail/users/attachments/20081016/b5e8f2a9/attachment.obj>

From simone.mannori at scilab.org  Thu Oct 16 17:21:24 2008
From: simone.mannori at scilab.org (Simone Mannori)
Date: Thu, 16 Oct 2008 17:21:24 +0200
Subject: [scilab-Users] graphics speed from 4.1.2 to 5.0.2 and virtual
	graphical window
In-Reply-To: <48F743F7.3010204@yandex.ru>
References: <e578e190810010808jac39e7enf326a854053b32f2@mail.gmail.com>
	 <faf3083f0810010829w3b5497c6na90117901c58ce8e@mail.gmail.com>
	 <20081001155148.GA16033@delacroix.inrialpes.fr>
	 <48F7371C.6010201@yandex.ru> <1224161736.18336.185.camel@korcula.inria.fr>
	       <48F73DC2.0@yandex.ru> <1224162874.18336.191.camel@korcula.inria.fr>
	  <48F743F7.3010204@yandex.ru>
Message-ID: <1224170484.3081.31.camel@pinarellu.inria.fr>

Bonjour Vyacheslav,

the model that you have send with the last email has some convergence
problem in both Scilab 4.1.2 and Scilab 5.0.

These are pure SUNDIALS issues: probably your systems ha some critical
convergence parameters to fix.
May you  be so kind to send me a version with the string in English ? 

The "stop" button issue and others Scicos related problems could be
related to the relatively old model of NVIDIA card (MX440, driver
86-96.43.07). Are you sure that is not possible to find more recent
driver like this one ?

http://www.nvidia.com/object/linux_display_x86_96.43.07.html

Thanks for using Scilab and Scicos

Simone Mannori - The Scilab Consortium (DIGITEO)
Scilab/Scicos Embedded Applications

//**----------------------------------------------------------------------

On Thu, 2008-10-16 at 21:39 +0800, Vyacheslav V. Yurkov wrote:
> Sylvestre Ledru wrote: 
> > Le jeudi 16 octobre 2008 ? 21:12 +0800, Vyacheslav V. Yurkov a ?crit :
> >   
> > Do you know if the acceleration of your X server is enable ?
> > 
> >   
> Of course. Sometimes I play games using 3D acceleration.
> > > Even button "stop" don't respond to pushing on it during simulation
> > > for one-two minutes.
> > >     
> > That is indeed an anormal issue.
> >   
> Maybe it would be useful: I use jre-6u6.
> Sorry for disinformation but it happens not with all models. Only with
> my model I've enclosed.
> 
> 2 Francis Drossaert: I'm talking about graphics in scicos emulation.
> -- 
> Vyacheslav V. Yurkov <uvv.mail at gmail.com>	Process Control Engineer
> Norilsk						Tel: +79069001870
> 
> Department of process control systems
> Open Joint Stock Company MMC Norilsk Nickel, brass works



From sylvestre.ledru at scilab.org  Fri Oct 17 17:47:07 2008
From: sylvestre.ledru at scilab.org (Sylvestre Ledru)
Date: Fri, 17 Oct 2008 17:47:07 +0200
Subject: [scilab-Users] Inexplicable result with scilab 5.0.2
	and	matrix multiplication
In-Reply-To: <1224071165.18336.83.camel@korcula.inria.fr>
References: <48F3BE88.7030603@dittmann-arheilgen.de>
	 <1223934038.3427.54.camel@zlarin>  <48F3C22C.20606@free.fr>
	 <1224071165.18336.83.camel@korcula.inria.fr>
Message-ID: <1224258427.18336.380.camel@korcula.inria.fr>

It is a bug of gfortran 4.3 which didn't exist in the version 4.2.
I submitted a bug report:
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37863
which has been very quickly confirmed as being a bug.

Sylvestre

Le mercredi 15 octobre 2008 ? 13:46 +0200, Sylvestre Ledru a ?crit :
> For those who are interested in this bug, the discussion continues in
> the bugtracker.
> (You can add your email address to the bug report to receive updates).
> 
> Sylvestre
> 
> Le lundi 13 octobre 2008 ? 23:48 +0200, Fran?ois Vogel a ?crit : 
> > Looks like this one:
> > 
> > http://bugzilla.scilab.org/show_bug.cgi?id=2647
> > 
> > Francois
> > 
> > 
> > Sylvestre Ledru said on 13/10/2008 23:40:
> > > Hello,
> > > 
> > > It is most probably a bug in the printing functions... I can reproduce
> > > it with the binary (ie not compiled on my computer) but I can't on my
> > > dev version (compiled with gfortran 4.3.1)
> > > Once, we had a simular bug caused by a bug in gfortran. Maybe it is a
> > > regression.
> > > 
> > > Thanks for pointing this out.
> > > Sylvestre
> > > 
> > > Le lundi 13 octobre 2008 ? 23:32 +0200, Christoph Dittmann a ?crit :
> > >> Hi,
> > >>
> > >> I've got a problem with scilab 5.0.2. It seems to print certain values 
> > >> wrong. The following example shows the problem:
> > >> -->M = [1,2;3,2]; C = M^(-1)*M
> > >>   C  =
> > >>
> > >>      2.           1.110D-16
> > >>      1.110D-16    2.
> > >>
> > >> This is what I get with scilab 5.0.2 on Ubuntu 8.04 (x86) using the 
> > >> precompiled version from http://www.scilab.org/download/ . My problem 
> > >> are the 2's on the diagonal since I've no idea where they come from.
> > >>
> > >> With scilab 4.1.2, the one from the Ubuntu 8.04 repository, I get what I 
> > >> expect:
> > >> -->M = [1,2;3,2]; C = M^(-1)*M
> > >>   C  =
> > >>
> > >>      1.           1.110D-16
> > >>      5.551D-17    1.
> > >>
> > >>
> > >> Some further calculations with scilab 5.0.2 show that the problem seems 
> > >> to lie in the printing of the result and not its actual numerical value:
> > >> -->x = C(1,1)
> > >>   x  =
> > >>
> > >>      2.
> > >> -->x * 2
> > >>   ans =
> > >>
> > >>     2.
> > >> -->x + 0.01
> > >>   ans  =
> > >>
> > >>      1.01
> > >>
> > >>
> > >> Could that be a bug or did I just miss something crucial?
> > >>
> > >>
> > >> Thanks,
> > >> Christoph
> > > 
> > > 
> > > cd 
> 



From UVV-mail at yandex.ru  Sat Oct 18 10:17:05 2008
From: UVV-mail at yandex.ru (Vyacheslav V. Yurkov)
Date: Sat, 18 Oct 2008 16:17:05 +0800
Subject: Transfer function
Message-ID: <48F99B81.2020901@yandex.ru>

Is it possible to get transfer function from data array in scilab?

-- 
Vyacheslav V. Yurkov <uvv.mail at gmail.com>	Process Control Engineer
Norilsk						Tel: +79069001870

Department of process control systems
Open Joint Stock Company MMC Norilsk Nickel, brass works



From c_korn at gmx.de  Sat Oct 18 22:25:04 2008
From: c_korn at gmx.de (Christoph Korn)
Date: Sat, 18 Oct 2008 22:25:04 +0200
Subject: No exact results
Message-ID: <48FA4620.3060400@gmx.de>

Why doesn't scilab-5.0.2 output exact results?

Examples:
-->sin(%pi)
 ans  =
 
    1.225E-16

-->M=[1,2;3,2]; C=M^-1 * M
 C  =
 
    1.           1.110E-16 
    1.110E-16    1. 

I mean, even maxima can do so ;-)

http://img300.imageshack.us/img300/3764/bildschirmfotowxmaxima0vc4.png


From vaborg at gmail.com  Tue Oct 21 17:16:47 2008
From: vaborg at gmail.com (Ivan Maximov)
Date: Tue, 21 Oct 2008 17:16:47 +0200
Subject: Matrix exponent calculation
Message-ID: <20081021171647.28af2a90.vaborg@gmail.com>

Hi!

It's just interesting to know under what kind of conditions I get
a convergence problem when computing a matrix exponent dimension 16x16?
because sometimes it appears and I don't know a reason for that and sometimes
Scilab computes the matrix exponent (I mean expm() function).

thanks in advance.

---------
best regards,
Ivan



From sylvestre.ledru at scilab.org  Wed Oct 22 11:27:56 2008
From: sylvestre.ledru at scilab.org (Sylvestre Ledru)
Date: Wed, 22 Oct 2008 11:27:56 +0200
Subject: [scilab-Users] No exact results
In-Reply-To: <48FA4620.3060400@gmx.de>
References: <48FA4620.3060400@gmx.de>
Message-ID: <1224667676.14725.20.camel@korcula.inria.fr>

Le samedi 18 octobre 2008 ? 22:25 +0200, Christoph Korn a ?crit :
> Why doesn't scilab-5.0.2 output exact results?
Because Scilab is a numerical software. %pi is a predefined variable (3.14xxx)
and it is this given value which is used by the sin function. Not the
fact that it is working with pi itself.

> Examples:
> -->sin(%pi)
>  ans  =
>  
>     1.225E-16

> I mean, even maxima can do so ;-)
Because it is a symbolic computing software.

S




From patufet99 at gmail.com  Wed Oct 22 11:35:27 2008
From: patufet99 at gmail.com (Patufet P)
Date: Wed, 22 Oct 2008 11:35:27 +0200
Subject: Question with multiple 3d plot colors
Message-ID: <ae7092b40810220235r55754f37pa3265ae21998d68f@mail.gmail.com>

Hello,

I would like to know how several surfaces on the same graph could be
shown with the color corresponding to the height Z.
With a single surface this works great:

f=scf();
f.color_map=jetcolormap(256);
t=[0:0.3:2*%pi]';
z=sin(t)*cos(t');
plot3d(t,t,z)
e=gce();
e.color_flag=1;

but if I add different surfaces afterwards

plot3d(t,t,z+5)
e=gce();
e.color_flag=1;

the colors are mapped for each new surface.

What I would like is that one color corresponds to one height for all
the graphs.

Thank you for any hints.

Regards.


From c_korn at gmx.de  Wed Oct 22 12:58:29 2008
From: c_korn at gmx.de (Christoph Korn)
Date: Wed, 22 Oct 2008 12:58:29 +0200
Subject: [scilab-Users] No exact results
In-Reply-To: <1224667676.14725.20.camel@korcula.inria.fr>
References: <48FA4620.3060400@gmx.de> <1224667676.14725.20.camel@korcula.inria.fr>
Message-ID: <48FF0755.7050907@gmx.de>

An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20081022/5399a5e1/attachment.htm>

From enrico.segre at weizmann.ac.il  Wed Oct 22 12:56:23 2008
From: enrico.segre at weizmann.ac.il (Enrico Segre)
Date: Wed, 22 Oct 2008 12:56:23 +0200
Subject: Question with multiple 3d plot colors
Message-ID: <1224672983.10756.6.camel@segre-pc2.weizmann.ac.il>

AFAIU that should be possible only if you first transform your surfaces
to  facelet representation (see help genfac3d) and then either you call
plot3d(xf,yf,zf) lumping all the facelets together, or
plot3d(xf,yf,list(zf,colors)) for each one of them, specifying at hand
the colors of each piece as function of z 
(i.e. colors=1+(zf-zmin)*(ncolors-1)/(zmax-zmin) ).

hth, Enrico




From c_korn at gmx.de  Wed Oct 22 21:29:08 2008
From: c_korn at gmx.de (Christoph Korn)
Date: Wed, 22 Oct 2008 21:29:08 +0200
Subject: Error in scilab compiled for amd64
Message-ID: <48FF7F04.4020406@gmx.de>

Hello,
I have compiled scilab 5.0.2 for hardy amd64.

The sources and binaries can be found in this ppa:
https://launchpad.net/~getdeb.packages/+archive

You can checkout the sources by downloading the files:
https://launchpad.net/~getdeb.packages/+archive/+files/scilab_5.0.2.orig.tar.gz
https://launchpad.net/~getdeb.packages/+archive/+files/scilab_5.0.2-1~ppa1.diff.gz

tar xzf scilab_5.0.2.orig.tar.gz
cd scilab-5.0.2
zcat ../scilab_5.0.2-1~ppa1.diff.gz | patch -p1

The i386 packages run fine.

In amd64 scilab gives this error:
http://nopaste.com/p/arvctbquq/txt

This is the /usr/share/scilab/etc/scilab.start file:
http://nopaste.com/p/astHMiTdk/txt

The output of some calculations is also odd:
http://img1.abload.de/img/bildschirmfoto-consoleby3.png


From sylvestre.ledru at inria.fr  Wed Oct 22 21:58:26 2008
From: sylvestre.ledru at inria.fr (Sylvestre Ledru)
Date: Wed, 22 Oct 2008 21:58:26 +0200
Subject: [scilab-Users] Error in scilab compiled for amd64
In-Reply-To: <48FF7F04.4020406@gmx.de>
References: <48FF7F04.4020406@gmx.de>
Message-ID: <1224705506.3319.174.camel@zlarin>

Hello Christoph,

Could you send your config.log ?

I am mainly interested by the flags:
IS_64_BITS_CPU & USE_DYNAMIC_STACK 

Thx,
Sylvestre
PS: This discussion would be a bit more appropriated on the dev ML.

Le mercredi 22 octobre 2008 ? 21:29 +0200, Christoph Korn a ?crit :
> Hello,
> I have compiled scilab 5.0.2 for hardy amd64.
> 
> The sources and binaries can be found in this ppa:
> https://launchpad.net/~getdeb.packages/+archive
> 
> You can checkout the sources by downloading the files:
> https://launchpad.net/~getdeb.packages/+archive/+files/scilab_5.0.2.orig.tar.gz
> https://launchpad.net/~getdeb.packages/+archive/+files/scilab_5.0.2-1~ppa1.diff.gz
> 
> tar xzf scilab_5.0.2.orig.tar.gz
> cd scilab-5.0.2
> zcat ../scilab_5.0.2-1~ppa1.diff.gz | patch -p1
> 
> The i386 packages run fine.
> 
> In amd64 scilab gives this error:
> http://nopaste.com/p/arvctbquq/txt
> 
> This is the /usr/share/scilab/etc/scilab.start file:
> http://nopaste.com/p/astHMiTdk/txt
> 
> The output of some calculations is also odd:
> http://img1.abload.de/img/bildschirmfoto-consoleby3.png



From Dan_Gill at mksinst.com  Wed Oct 22 21:58:40 2008
From: Dan_Gill at mksinst.com (Dan_Gill at mksinst.com)
Date: Wed, 22 Oct 2008 15:58:40 -0400
Subject: [scilab-Users] Error in scilab compiled for amd64
Message-ID: <OF852574EA.006DE5F1-ON852574EA.006DE5F1-852574EA.006DE5F1@mksinst.com>

Return Receipt
                                                                           
   Your       [scilab-Users] Error in scilab compiled for amd64            
   document:                                                               
                                                                           
   was        Dan_Gill at mksinst.com                                         
   received                                                                
   by:                                                                     
                                                                           
   at:        10/22/2008 04:00:22 PM                                       
                                                                           






From patufet99 at gmail.com  Thu Oct 23 09:07:46 2008
From: patufet99 at gmail.com (Patufet P)
Date: Thu, 23 Oct 2008 09:07:46 +0200
Subject: [scilab-Users] Re: Question with multiple 3d plot colors
In-Reply-To: <1224672983.10756.6.camel@segre-pc2.weizmann.ac.il>
References: <1224672983.10756.6.camel@segre-pc2.weizmann.ac.il>
Message-ID: <ae7092b40810230007u4511d681w7d49dcd3256a1042@mail.gmail.com>

Your suggestion works well. In the "colorbar" help there is an example
where you can map the z space to the color scale: dsearch(zz,
linspace(zmin, zmax, nb_col+1));
My example works with:

f=scf();
nb_col = 64;
f.color_map=jetcolormap(nb_col)
zmax=6;
zmin=-1;
colorbar(zmin,zmax,fmt="%.1f")
t=[0:0.3:2*%pi]';
z=sin(t)*cos(t');
[xx,yy,zz]=genfac3d(t,t,z);
zcol=dsearch(zz, linspace(zmin, zmax, nb_col+1));
plot3d(xx,yy,list(zz,zcol))

[xx,yy,zz]=genfac3d(t,t,z+5);
zcol=dsearch(zz, linspace(zmin, zmax, nb_col+1));
plot3d(xx,yy,list(zz,zcol))

Thank you for your help.

Regards

On Wed, Oct 22, 2008 at 12:56 PM, Enrico Segre
<enrico.segre at weizmann.ac.il> wrote:
> AFAIU that should be possible only if you first transform your surfaces
> to  facelet representation (see help genfac3d) and then either you call
> plot3d(xf,yf,zf) lumping all the facelets together, or
> plot3d(xf,yf,list(zf,colors)) for each one of them, specifying at hand
> the colors of each piece as function of z
> (i.e. colors=1+(zf-zmin)*(ncolors-1)/(zmax-zmin) ).
>
> hth, Enrico
>
>
>


From michael.baudin at scilab.org  Thu Oct 23 09:23:46 2008
From: michael.baudin at scilab.org (Michael Baudin)
Date: Thu, 23 Oct 2008 09:23:46 +0200
Subject: [scilab-Users] Matrix exponent calculation
In-Reply-To: <20081021171647.28af2a90.vaborg@gmail.com>
References: <20081021171647.28af2a90.vaborg@gmail.com>
Message-ID: <49002682.5000907@scilab.org>

Hi,
Can you send the particular matrix for which the
problem appear ?
Regards,
Micha?l Baudin


Ivan Maximov a ?crit :
> Hi!
>
> It's just interesting to know under what kind of conditions I get
> a convergence problem when computing a matrix exponent dimension 16x16?
> because sometimes it appears and I don't know a reason for that and sometimes
> Scilab computes the matrix exponent (I mean expm() function).
>
> thanks in advance.
>
> ---------
> best regards,
> Ivan
>
>   



From osvaldo at dcc.ufmg.br  Thu Oct 23 16:12:29 2008
From: osvaldo at dcc.ufmg.br (Osvaldo Sergio Farhat de Carvalho)
Date: Thu, 23 Oct 2008 12:12:29 -0200
Subject: [scilab-Users] Error in scilab compiled for amd64
Message-ID: <OFC258E024.16169F95-ON832574EB.004E0D6A-832574EB.004E0D6A@grude.ufmg.br>

Return Receipt
                                                                           
   Your       [scilab-Users] Error in scilab compiled for amd64            
   document:                                                               
                                                                           
   was        osvaldo at dcc.ufmg.br                                          
   received                                                                
   by:                                                                     
                                                                           
   at:        23/10/2008 12:12:32                                          
                                                                           


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20081023/5711804a/attachment.htm>

From geodesicos at gmail.com  Thu Oct 23 17:04:55 2008
From: geodesicos at gmail.com (Alfredo Jaramillo)
Date: Thu, 23 Oct 2008 12:04:55 -0300
Subject: Problems with help search
Message-ID: <42abbe500810230804q305105dbhf879e0bd9c602d4e@mail.gmail.com>

Hi, We work with programming on numeric problems, and on the 4.x versions
the search engine works really good, but starting 5.X versions there are
problems with it. For instance: if we write "poly" on the "search help"
engine we found nothing apropiate about that function. Is this problem
linked with java interface?


Alfredo Jaramillo Palma
Estudiante de Ingenier?a Matem?tica
Universidad De La Frontera
IX Regi?n, Chile


http://antipoemasypoemas.blogspot.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20081023/9711f21b/attachment.htm>

From joshua.auerbach at uvm.edu  Fri Oct 24 18:03:37 2008
From: joshua.auerbach at uvm.edu (Joshua Auerbach)
Date: Fri, 24 Oct 2008 12:03:37 -0400
Subject: [scilab-Users] plotting two y-axes, lines and bars in one plot
In-Reply-To: <b04eeae20808280634v6bc70aedj9f93029eaaac8900@mail.gmail.com>
References: <48B67E77.6010806@web.de> <b04eeae20808280634v6bc70aedj9f93029eaaac8900@mail.gmail.com>
Message-ID: <4901F1D9.9060603@uvm.edu>

I am trying to use your toolbox in scilab 5.02 but cannot seem to get it 
installed.  Are there different installation instructions when using the 
version for 5.02?

Eric Dubois wrote:
> In my toolbox grocer (with E. Michaux), available art 
> http://dubois.ensae.net./grocer.html you will find a function 
> pltseries0 that has many options, including the possibility to draw 2 
> axes and to mix bars and curves.
>  
> Hope it works for you.
>  
> Eric. 
>
>  
> 2008/8/28, Simon <simon.elfert at web.de <mailto:simon.elfert at web.de>>:
>
>     Hi everybody,
>
>     I would like to do some plotting with scilab. The aim is as follows:
>
>     - one plot with two different y-axes
>     - the left y-axis shall be used for some lines (timeseries)
>     - the right y-axis shall be used for a timeseries plotted as bars
>
>     In matlab this should be possible with plotyy(), but there seems
>     to be a possibility in scilab as well using drawaxis.
>     Unfortunately I was not able yet to implement it.
>
>     Do you know, how I can plot both - lines and bars - in one plot
>     and differentiate the y-axes? (by the way... the right y-axis has
>     to scaled reversely topdown, but one step after the oter...)
>
>
>     Thank you, kind regards
>     Simon Elfert
>
>
>
>
>     ps - do you know a good and broad plotting tutorial for scilab?
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20081024/007b80c0/attachment.htm>

From grocer.toolbox at gmail.com  Fri Oct 24 21:12:08 2008
From: grocer.toolbox at gmail.com (Eric Dubois)
Date: Fri, 24 Oct 2008 21:12:08 +0200
Subject: [scilab-Users] plotting two y-axes, lines and bars in one plot
In-Reply-To: <4901F1D9.9060603@uvm.edu>
References: <48B67E77.6010806@web.de>
	 <b04eeae20808280634v6bc70aedj9f93029eaaac8900@mail.gmail.com>
	 <4901F1D9.9060603@uvm.edu>
Message-ID: <b04eeae20810241212h2ac89d1mcf5d9697dd8bce03@mail.gmail.com>

You will find on my website a version that works with Scilab 5.0.2.

Eric.

2008/10/24 Joshua Auerbach <joshua.auerbach at uvm.edu>

>  I am trying to use your toolbox in scilab 5.02 but cannot seem to get it
> installed.  Are there different installation instructions when using the
> version for 5.02?
>
> Eric Dubois wrote:
>
> In my toolbox grocer (with E. Michaux), available art
> http://dubois.ensae.net./grocer.html you will find a function pltseries0
> that has many options, including the possibility to draw 2 axes and to mix
> bars and curves.
>
> Hope it works for you.
>
> Eric.
>
>
> 2008/8/28, Simon <simon.elfert at web.de>:
>>
>> Hi everybody,
>>
>> I would like to do some plotting with scilab. The aim is as follows:
>>
>> - one plot with two different y-axes
>> - the left y-axis shall be used for some lines (timeseries)
>> - the right y-axis shall be used for a timeseries plotted as bars
>>
>> In matlab this should be possible with plotyy(), but there seems to be a
>> possibility in scilab as well using drawaxis. Unfortunately I was not able
>> yet to implement it.
>>
>> Do you know, how I can plot both - lines and bars - in one plot and
>> differentiate the y-axes? (by the way... the right y-axis has to scaled
>> reversely topdown, but one step after the oter...)
>>
>>
>> Thank you, kind regards
>> Simon Elfert
>>
>>
>>
>>
>> ps - do you know a good and broad plotting tutorial for scilab?
>>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20081024/c4c6dc43/attachment.htm>

From debian1sy8 at gmail.com  Sat Oct 25 16:30:43 2008
From: debian1sy8 at gmail.com (Maki 1sy8)
Date: Sat, 25 Oct 2008 16:30:43 +0200
Subject: Using english language in Scilab
Message-ID: <58061b670810250730k22d9a166kdf6c9d04b71bed69@mail.gmail.com>

Hi all,

I'd like to use the english version of Scilab 5.0.2 on a win* XP box.
I looked in the etc/scilab.start directory  but found nothing to
change. Then I added the -l en option to the shortcut but it would
only change the help system not the menus language. The environment
variable is said to be OBSOLETE in the help. I also tried to issue a
setlanguage('en') command but it changes nothing even after I restar
scilab.

How does one force the menus and the help system to be in english, please ?

Regards.


From sylvestre.ledru at inria.fr  Sat Oct 25 16:35:21 2008
From: sylvestre.ledru at inria.fr (Sylvestre Ledru)
Date: Sat, 25 Oct 2008 16:35:21 +0200
Subject: [scilab-Users] Using english language in Scilab
In-Reply-To: <58061b670810250730k22d9a166kdf6c9d04b71bed69@mail.gmail.com>
References: <58061b670810250730k22d9a166kdf6c9d04b71bed69@mail.gmail.com>
Message-ID: <1224945321.21746.36.camel@zlarin>

Le samedi 25 octobre 2008 ? 16:30 +0200, Maki 1sy8 a ?crit :
> Hi all,
> 
> I'd like to use the english version of Scilab 5.0.2 on a win* XP box.
> I looked in the etc/scilab.start directory  but found nothing to
> change. Then I added the -l en option to the shortcut but it would
> only change the help system not the menus language. The environment
> variable is said to be OBSOLETE in the help. I also tried to issue a
> setlanguage('en') command but it changes nothing even after I restar
> scilab.
Try: setlanguage("en_US")
Otherwise, to sent the system variable, LANG=en_US before launching
Scilab.

Sylvestre





From debian1sy8 at gmail.com  Sun Oct 26 11:08:22 2008
From: debian1sy8 at gmail.com (Maki 1sy8)
Date: Sun, 26 Oct 2008 11:08:22 +0100
Subject: [scilab-Users] Using english language in Scilab
In-Reply-To: <1224945321.21746.36.camel@zlarin>
References: <58061b670810250730k22d9a166kdf6c9d04b71bed69@mail.gmail.com>
	 <1224945321.21746.36.camel@zlarin>
Message-ID: <58061b670810260308k5f572fc1m5dc9352c8311d9de@mail.gmail.com>

On Sat, Oct 25, 2008 at 3:35 PM, Sylvestre Ledru
<sylvestre.ledru at inria.fr> wrote:
> Le samedi 25 octobre 2008 ? 16:30 +0200, Maki 1sy8 a ?crit :
>> Hi all,
>>
>> I'd like to use the english version of Scilab 5.0.2 on a win* XP box.
>> I looked in the etc/scilab.start directory  but found nothing to
>> change. Then I added the -l en option to the shortcut but it would
>> only change the help system not the menus language. The environment
>> variable is said to be OBSOLETE in the help. I also tried to issue a
>> setlanguage('en') command but it changes nothing even after I restar
>> scilab.
> Try: setlanguage("en_US")
> Otherwise, to sent the system variable, LANG=en_US before launching
> Scilab.
>
> Sylvestre
>
>

Thanks for your help but still no change!!
I added a  new environment variable as you said but to no avail. When
I tried the obsolete LANGUAGE variable, the banner started up in
english but all menus are greyed out (and still in french) except the
Applications and help menus. Both menu labels (and submenus) are also
in french and when I also checked the language used in the help
system, it was in french!!

There's something wrong with the way the language is set up. Maybe the
language setting should be offered for change during setup! Why is it
so difficult just to set the language ?


From debian1sy8 at gmail.com  Sun Oct 26 12:55:18 2008
From: debian1sy8 at gmail.com (Maki 1sy8)
Date: Sun, 26 Oct 2008 12:55:18 +0100
Subject: Matlab which equivalent
Message-ID: <58061b670810260455l4fbe3748sa491b680e2772b40@mail.gmail.com>

Hi all,

Do you know if scilab provides an equivalent to the Matlab command
'which' ? It allows to know the file containing a given function.

Regards


From fvogelnew1 at free.fr  Sun Oct 26 13:17:17 2008
From: fvogelnew1 at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Vogel?=)
Date: Sun, 26 Oct 2008 13:17:17 +0100
Subject: [scilab-Users] Matlab which equivalent
In-Reply-To: <58061b670810260455l4fbe3748sa491b680e2772b40@mail.gmail.com>
References: <58061b670810260455l4fbe3748sa491b680e2772b40@mail.gmail.com>
Message-ID: <49045FCD.1070409@free.fr>

Maki 1sy8 said on 26/10/2008 12:55:
> Do you know if scilab provides an equivalent to the Matlab command
> 'which' ? It allows to know the file containing a given function.

help whereis
This will give you the name of the library containing a function.

Or use Scipad to open the function itself: File/Open source of...

Francois


From debian1sy8 at gmail.com  Sun Oct 26 13:40:20 2008
From: debian1sy8 at gmail.com (Maki 1sy8)
Date: Sun, 26 Oct 2008 13:40:20 +0100
Subject: [scilab-Users] Matlab which equivalent
In-Reply-To: <49045FCD.1070409@free.fr>
References: <58061b670810260455l4fbe3748sa491b680e2772b40@mail.gmail.com>
	 <49045FCD.1070409@free.fr>
Message-ID: <58061b670810260540w477f5917oe48b5e27161b1607@mail.gmail.com>

On Sun, Oct 26, 2008 at 1:17 PM, Fran?ois Vogel <fvogelnew1 at free.fr> wrote:
> Maki 1sy8 said on 26/10/2008 12:55:
>>
>> Do you know if scilab provides an equivalent to the Matlab command
>> 'which' ? It allows to know the file containing a given function.
>
> help whereis
> This will give you the name of the library containing a function.
>
> Or use Scipad to open the function itself: File/Open source of...
>
> Francois
>

In fact, I already tried whereis. But for example 'whereis poly'
returned me an empty matrix and 'whereis mean' returned me
statisticslib. Which doesn't tell much about where the definitions of
these functions are actually located (even if they are not written in
the scilab language themselves).

BTW, I couldn't see in the FAQ or in the doc the meaning of all the
extension/type of files used and recognized by Scilab. What is the
difference between .sce and .sci, what are their intended uses ? What
exactly are in .sav files, etc ..?


From fvogelnew1 at free.fr  Sun Oct 26 14:07:14 2008
From: fvogelnew1 at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Vogel?=)
Date: Sun, 26 Oct 2008 14:07:14 +0100
Subject: [scilab-Users] Matlab which equivalent
In-Reply-To: <58061b670810260540w477f5917oe48b5e27161b1607@mail.gmail.com>
References: <58061b670810260455l4fbe3748sa491b680e2772b40@mail.gmail.com>	 <49045FCD.1070409@free.fr> <58061b670810260540w477f5917oe48b5e27161b1607@mail.gmail.com>
Message-ID: <49046B82.9040302@free.fr>

Maki 1sy8 said on 26/10/2008 13:40:
> On Sun, Oct 26, 2008 at 1:17 PM, Fran?ois Vogel <fvogelnew1 at free.fr> wrote:
>> Maki 1sy8 said on 26/10/2008 12:55:
>>> Do you know if scilab provides an equivalent to the Matlab command
>>> 'which' ? It allows to know the file containing a given function.
>> help whereis
>> This will give you the name of the library containing a function.
>>
>> Or use Scipad to open the function itself: File/Open source of...
> 
> In fact, I already tried whereis. But for example 'whereis poly'
> returned me an empty matrix and 'whereis mean' returned me
> statisticslib.

whereis only provides the name of the *library* where a Scilab 
keyword/function is stored, and only if it is written in Scilab language.

> Which doesn't tell much about where the definitions of
> these functions are actually located

Scipad does it (see my previous post). It does it basically through 
get_function_path

> (even if they are not written in
> the scilab language themselves).

AFAIK, there is no way of knowing from Scilab where a function coded 
in c or fortran is stored in the Scilab source tree.

> BTW, I couldn't see in the FAQ or in the doc the meaning of all the
> extension/type of files used and recognized by Scilab. What is the
> difference between .sce and .sci,

Just a convention.

> what are their intended uses ?

.sci: functions
.sce: executable (level 0) code

All this only by convention. Scilab doesn't care. You can 
getf("mywonderfulfoobar.thebestscript") if you want to.

> What
> exactly are in .sav files, etc ..?

See help save_format.

Francois


From debian1sy8 at gmail.com  Sun Oct 26 14:47:13 2008
From: debian1sy8 at gmail.com (Maki 1sy8)
Date: Sun, 26 Oct 2008 14:47:13 +0100
Subject: [scilab-Users] Matlab which equivalent
In-Reply-To: <49046B82.9040302@free.fr>
References: <58061b670810260455l4fbe3748sa491b680e2772b40@mail.gmail.com>
	 <49045FCD.1070409@free.fr>
	 <58061b670810260540w477f5917oe48b5e27161b1607@mail.gmail.com>
	 <49046B82.9040302@free.fr>
Message-ID: <58061b670810260647k5fd53475hc1ca21c7c7c80c98@mail.gmail.com>

On Sun, Oct 26, 2008 at 2:07 PM, Fran?ois Vogel <fvogelnew1 at free.fr> wrote:
>> Which doesn't tell much about where the definitions of
>> these functions are actually located
>
> Scipad does it (see my previous post). It does it basically through
> get_function_path
>

All right. I could see the mean function for example.

>> (even if they are not written in
>> the scilab language themselves).
>
> AFAIK, there is no way of knowing from Scilab where a function coded in c or
> fortran is stored in the Scilab source tree.
>

Ok, but it would be nice to know the whereabouts of the file even if
it's in compiled form so that interested users can check the source
file easily if needed.

>> BTW, I couldn't see in the FAQ or in the doc the meaning of all the
>> extension/type of files used and recognized by Scilab. What is the
>> difference between .sce and .sci,
>
> Just a convention.
>
>> what are their intended uses ?
>
> .sci: functions
> .sce: executable (level 0) code
>

Thanks a lot. I intend to stick to this convention. But it's better
being explained so that people can also follow it.


From debian1sy8 at gmail.com  Sun Oct 26 15:24:11 2008
From: debian1sy8 at gmail.com (Maki 1sy8)
Date: Sun, 26 Oct 2008 15:24:11 +0100
Subject: Path and libs
Message-ID: <58061b670810260724q668d3fe2kf82991ecc91b218c@mail.gmail.com>

On Sun, Oct 26, 2008 at 2:07 PM, Fran?ois Vogel <fvogelnew1 at free.fr> wrote:
> You can getf("mywonderfulfoobar.thebestscript") if you want to.
>

I went into the getf help and I saw this:
'getf is an obsolete way for loading functions into scilab from a
file. It is replaced by the function.'

Kind of short: Replaced by which function exactly?  Is it the exec
function the help is talking about a few words later in the same
paragraph ?

I have a few questions related to path and user libs management. In
Matlab, under the File menu, one could set path, or add path with all
subfolders for example so any .m function would be directly known
under Matlab apart from the current directory which is in the 'path'
automatically.

1-  How do I change the current directory in Scilab? Going thru the
french menu   Fichier->Changer le repertoire courant is very
cumbersome: I can't browse  to the directory I want to use! I'm stuck
to the Desktop!

2- What is SCIHOME? It has nothing to do with the current working
directory displayed by the Fichier->Afficher le repertoire courant
command. Why can't I change it thru the Applications->Editeur de
variable menu?

3- I would like to load at startup, say, a directory with a hierarchy
of subdirectories containing user-defined functions (which I call a
user library but then I don't know if Scilab libraries/toolboxes are
managed this way) so that I can use these functions right away. How
can I do this?


From fvogelnew1 at free.fr  Sun Oct 26 15:51:01 2008
From: fvogelnew1 at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Vogel?=)
Date: Sun, 26 Oct 2008 15:51:01 +0100
Subject: [scilab-Users] Path and libs
In-Reply-To: <58061b670810260724q668d3fe2kf82991ecc91b218c@mail.gmail.com>
References: <58061b670810260724q668d3fe2kf82991ecc91b218c@mail.gmail.com>
Message-ID: <490483D5.8000509@free.fr>

Maki 1sy8 said on 26/10/2008 15:24:
> I went into the getf help and I saw this:
> 'getf is an obsolete way for loading functions into scilab from a
> file. It is replaced by the function.'
> 
> Kind of short: Replaced by which function exactly?  Is it the exec
> function the help is talking about a few words later in the same
> paragraph ?

Yes.
I have fixed this in the master branch.

> I have a few questions related to path and user libs management. In
> Matlab, under the File menu, one could set path, or add path with all
> subfolders for example so any .m function would be directly known
> under Matlab apart from the current directory which is in the 'path'
> automatically.
> 
> 1-  How do I change the current directory in Scilab?

See help chdir  or  help cd

> 2- What is SCIHOME?

See the different help files proposed by apropos SCIHOME


> 3- I would like to load at startup, say, a directory with a hierarchy
> of subdirectories containing user-defined functions (which I call a
> user library but then I don't know if Scilab libraries/toolboxes are
> managed this way) so that I can use these functions right away. How
> can I do this?

See help startup

Francois



From debian1sy8 at gmail.com  Sun Oct 26 18:19:12 2008
From: debian1sy8 at gmail.com (Maki 1sy8)
Date: Sun, 26 Oct 2008 18:19:12 +0100
Subject: [scilab-Users] Path and libs
In-Reply-To: <490483D5.8000509@free.fr>
References: <58061b670810260724q668d3fe2kf82991ecc91b218c@mail.gmail.com>
	 <490483D5.8000509@free.fr>
Message-ID: <58061b670810261019x26b3e3bdjf52c0dd20e865922@mail.gmail.com>

Thanks for all the help. It's much appreciated.

The only problem that remains for now is the user interface language.
The help says the default is english but I still couldn't manage to
set it properly.


From fvogelnew1 at free.fr  Sun Oct 26 19:13:57 2008
From: fvogelnew1 at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Vogel?=)
Date: Sun, 26 Oct 2008 19:13:57 +0100
Subject: [scilab-Users] Path and libs
In-Reply-To: <58061b670810261019x26b3e3bdjf52c0dd20e865922@mail.gmail.com>
References: <58061b670810260724q668d3fe2kf82991ecc91b218c@mail.gmail.com>	 <490483D5.8000509@free.fr> <58061b670810261019x26b3e3bdjf52c0dd20e865922@mail.gmail.com>
Message-ID: <4904B365.9040508@free.fr>

Maki 1sy8 said on 26/10/2008 18:19:
> The only problem that remains for now is the user interface language.
> The help says the default is english but I still couldn't manage to
> set it properly.

setdefaultlanguage(en_US)

and restart Scilab.

Right, there is no help page for setdefaultlanguage, and you couldn't 
find it. This is now bug 3697.

http://bugzilla.scilab.org/show_bug.cgi?id=3697

F.


From debian1sy8 at gmail.com  Sun Oct 26 19:35:39 2008
From: debian1sy8 at gmail.com (Maki 1sy8)
Date: Sun, 26 Oct 2008 19:35:39 +0100
Subject: [scilab-Users] Path and libs
In-Reply-To: <4904B365.9040508@free.fr>
References: <58061b670810260724q668d3fe2kf82991ecc91b218c@mail.gmail.com>
	 <490483D5.8000509@free.fr>
	 <58061b670810261019x26b3e3bdjf52c0dd20e865922@mail.gmail.com>
	 <4904B365.9040508@free.fr>
Message-ID: <58061b670810261135q7ddb69aey4a8c2751fddb7db7@mail.gmail.com>

On Sun, Oct 26, 2008 at 7:13 PM, Fran?ois Vogel <fvogelnew1 at free.fr> wrote:
> Maki 1sy8 said on 26/10/2008 18:19:
>>
>> The only problem that remains for now is the user interface language.
>> The help says the default is english but I still couldn't manage to
>> set it properly.
>
> setdefaultlanguage(en_US)
>
> and restart Scilab.
>
> Right, there is no help page for setdefaultlanguage, and you couldn't find
> it. This is now bug 3697.
>
> http://bugzilla.scilab.org/show_bug.cgi?id=3697
>
> F.

This command didn't work after the first restart. And strangely, I got
the following:
-->getdefaultlanguage
 ans  =

 en_US

-->getlanguage
 ans  =

 fr_FR

I issued another -->setdefaultlanguage('en_US') to get the following
comforting message
The language for menus cannot be changed on a running console.
Restart Scilab to apply to menus.

So all is well now, and I thank you very much for all this help you
provided even on Sundays!

>> 3- I would like to load at startup, say, a directory with a hierarchy
>>  of subdirectories containing user-defined functions (which I call a
>>    user library but then I don't know if Scilab libraries/toolboxes are
>>   managed this way) so that I can use these functions right away. How
>>   can I do this?
> See help startup

Ok. So should I issue getd() commands for each needed subdirectory in
the .scilab file ?


From fvogelnew1 at free.fr  Sun Oct 26 19:40:07 2008
From: fvogelnew1 at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Vogel?=)
Date: Sun, 26 Oct 2008 19:40:07 +0100
Subject: [scilab-Users] Path and libs
In-Reply-To: <58061b670810261135q7ddb69aey4a8c2751fddb7db7@mail.gmail.com>
References: <58061b670810260724q668d3fe2kf82991ecc91b218c@mail.gmail.com>	 <490483D5.8000509@free.fr>	 <58061b670810261019x26b3e3bdjf52c0dd20e865922@mail.gmail.com>	 <4904B365.9040508@free.fr> <58061b670810261135q7ddb69aey4a8c2751fddb7db7@mail.gmail.com>
Message-ID: <4904B987.9020704@free.fr>

Maki 1sy8 said on 26/10/2008 19:35:
> Ok. So should I issue getd() commands for each needed subdirectory in
> the .scilab file ?

Yes. That's how I do it.

F.


From debian1sy8 at gmail.com  Sun Oct 26 20:05:37 2008
From: debian1sy8 at gmail.com (Maki 1sy8)
Date: Sun, 26 Oct 2008 20:05:37 +0100
Subject: Help search
Message-ID: <58061b670810261205q621cefd3v1593d42524035537@mail.gmail.com>

Hi again.

What is the meaning of the number on the left when searching for some help ?
Say, I looked for format (as I wanted to reproduce Matlab's format
long g;) and the format command is only proposed at line 8 with number
6 on the left. Notice, when I searched for the function keyword, I
nearly missed it as it's very far down the list. Is there a way to
make the search more to the point ?

A bookmark system would also be an interesting feature to add.

PS : little bug report, even with the user interface in english, the
label still says 'Trouver' instead of 'Find'


From remjg at free.fr  Mon Oct 27 04:02:09 2008
From: remjg at free.fr (Remi G.)
Date: Sun, 26 Oct 2008 21:02:09 -0600
Subject: Strange problem with "integrate" : "Convergence problem..." error
 24 (Scilab 4.1.2)
Message-ID: <49052F31.5070607@free.fr>

Hello!

I'm facing a strange problem with the function "integrate" (and also 
with "intg") with Scilab 4.1.2 :

If I want to integrate x^5 between -%pi and %pi, it works (hopefully) :
  integrate('t^5','t',-%pi,%pi)

But if replace 5 by a variable k (that I've defined before of course) it 
will not work for k>=5!
  i=5
  integrate('t^i','t',-%pi,%pi)

Here is the error message I get :
    !--error 24
  convergence problem...
  At line      26 of function integrate called by :
  integrate('t^i','t',-%pi,%pi)

That's really annoying because I can't code my least square 
approximation procedure which is... for tomorrow!
If someone could give me a way to go through this issue!
Thank you very much!

R?mi.


From galea at cyber-laser.com  Mon Oct 27 07:31:32 2008
From: galea at cyber-laser.com (Antony Galea)
Date: Mon, 27 Oct 2008 15:31:32 +0900
Subject: Scilab 5.0.2
In-Reply-To: <58061b670810261205q621cefd3v1593d42524035537@mail.gmail.com>
References: <58061b670810261205q621cefd3v1593d42524035537@mail.gmail.com>
Message-ID: <D0C937FDA721C3galea@cyber-laser.com>

I am trying to run Scilab 5.0.2 under XP SP3 and get the error message below
from the command prompt.

Error occured during initialization of VM
Could not reserve enough space for object heap

Can anyone help?



From sylvestre.ledru at inria.fr  Mon Oct 27 07:53:29 2008
From: sylvestre.ledru at inria.fr (Sylvestre Ledru)
Date: Mon, 27 Oct 2008 07:53:29 +0100
Subject: [scilab-Users] Scilab 5.0.2
In-Reply-To: <D0C937FDA721C3galea@cyber-laser.com>
References: <58061b670810261205q621cefd3v1593d42524035537@mail.gmail.com>
	 <D0C937FDA721C3galea@cyber-laser.com>
Message-ID: <1225090409.21746.113.camel@zlarin>

Hello Antony,

It is a bug of the 5.0.2 which is going to be fixed in the upcoming
5.0.3.

edit SCI/modules/jvm/etc/jvm_options.xml
and 
replaces <option value="-Xmx1024m"/>
by
<option value="-Xmx256m"/>

Sylvestre

Le lundi 27 octobre 2008 ? 15:31 +0900, Antony Galea a ?crit :
> I am trying to run Scilab 5.0.2 under XP SP3 and get the error message below
> from the command prompt.
> 
> Error occured during initialization of VM
> Could not reserve enough space for object heap
> 
> Can anyone help?
> 



From galea at cyber-laser.com  Mon Oct 27 08:01:04 2008
From: galea at cyber-laser.com (Antony Galea)
Date: Mon, 27 Oct 2008 16:01:04 +0900
Subject: [scilab-Users] Scilab 5.0.2
In-Reply-To: <1225090409.21746.113.camel@zlarin>
References: <58061b670810261205q621cefd3v1593d42524035537@mail.gmail.com> <D0C937FDA721C3galea@cyber-laser.com> <1225090409.21746.113.camel@zlarin>
Message-ID: <D1C93801C71584galea@cyber-laser.com>

Thank you - that was indeed the problem.

Regards,
Antony

>Hello Antony,
>
>It is a bug of the 5.0.2 which is going to be fixed in the upcoming
>5.0.3.
>
>edit SCI/modules/jvm/etc/jvm_options.xml
>and 
>replaces <option value="-Xmx1024m"/>
>by
><option value="-Xmx256m"/>
>
>Sylvestre
>
>Le lundi 27 octobre 2008 ? 15:31 +0900, Antony Galea a ?crit :
>> I am trying to run Scilab 5.0.2 under XP SP3 and get the error message 
>> below
>> from the command prompt.
>> 
>> Error occured during initialization of VM
>> Could not reserve enough space for object heap
>> 
>> Can anyone help?
>> 



From michael.baudin at scilab.org  Mon Oct 27 15:11:24 2008
From: michael.baudin at scilab.org (=?ISO-8859-1?Q?Micha=EBl_Baudin?=)
Date: Mon, 27 Oct 2008 15:11:24 +0100
Subject: [scilab-Users] Strange problem with "integrate" : "Convergence
 problem..." error 24 (Scilab 4.1.2)
In-Reply-To: <49052F31.5070607@free.fr>
References: <49052F31.5070607@free.fr>
Message-ID: <4905CC0C.6080203@scilab.org>

Hi Remi,

I inquired your problem and it seems to me that all is fine,
that is, the problem you face is a limitation of the computationnal
method and not a bug.

The "integrate" command is a Scilab macro which is
based on the "intg" Scilab primitive.

Now try the following Scilab script :

function y=f(x)
  y=sin(x)
endfunction
a = 0;
b=2*%pi;
computed = intg(a,b,f);

The expected result in that case is 0., since -cos(x) is periodic
on the [0,2 PI] range.
The same error message appears.
The following is the call stack, which shows how the quapro library
is used to compute the integral based on a "21-point gauss-kronrod 
approximation".

 differential_equations/sci_gateway/c/sci_intg
 > differential_equations/sci_gateway/fortran/intg.f
 > differential_equations/src/fortran/dqag0.f
 > dqags > quarul

By doing a step-by-step check of the process, it
appears that the numerical integration is correctly computed (near the
machine's epsilon), but a secondary check sets a boolean flag to a
value indicating that the error for that computation is too large.

A simple workaround for your problem may be to decompose the
range in two parts :

i1 = intg(0,%pi,f);
i2 = intg(%pi,2*%pi,f);

then the expected result is the sum of i1 and i2.
Other solutions include traditionnal ode solver (like dassl ) or 
Monte-Carlo methods.

Best regards,

Micha?l Baudin

Remi G. a ?crit :
> Hello!
>
> I'm facing a strange problem with the function "integrate" (and also 
> with "intg") with Scilab 4.1.2 :
>
> If I want to integrate x^5 between -%pi and %pi, it works (hopefully) :
>  integrate('t^5','t',-%pi,%pi)
>
> But if replace 5 by a variable k (that I've defined before of course) 
> it will not work for k>=5!
>  i=5
>  integrate('t^i','t',-%pi,%pi)
>
> Here is the error message I get :
>    !--error 24
>  convergence problem...
>  At line      26 of function integrate called by :
>  integrate('t^i','t',-%pi,%pi)
>
> That's really annoying because I can't code my least square 
> approximation procedure which is... for tomorrow!
> If someone could give me a way to go through this issue!
> Thank you very much!
>
> R?mi.



From michael.baudin at scilab.org  Mon Oct 27 15:20:50 2008
From: michael.baudin at scilab.org (=?ISO-8859-1?Q?Micha=EBl_Baudin?=)
Date: Mon, 27 Oct 2008 15:20:50 +0100
Subject: [scilab-Users] Strange problem with "integrate" : "Convergence
 problem..." error 24 (Scilab 4.1.2)
In-Reply-To: <4905CC0C.6080203@scilab.org>
References: <49052F31.5070607@free.fr> <4905CC0C.6080203@scilab.org>
Message-ID: <4905CE42.5060506@scilab.org>

Oh, and by the way, this is exactly the method used by integrate, so
you can use it with several step instead of only one :

x0=0.;
step = 2*%pi/10
x1=0:step:2*%pi;
integrate("sin(x)",'x',x0,x1)

The value you search for is the last one : 2.d-17 ~ 0

Regards,

Micha?l

Micha?l Baudin a ?crit :
> Hi Remi,
>
> I inquired your problem and it seems to me that all is fine,
> that is, the problem you face is a limitation of the computationnal
> method and not a bug.
>
> The "integrate" command is a Scilab macro which is
> based on the "intg" Scilab primitive.
>
> Now try the following Scilab script :
>
> function y=f(x)
>  y=sin(x)
> endfunction
> a = 0;
> b=2*%pi;
> computed = intg(a,b,f);
>
> The expected result in that case is 0., since -cos(x) is periodic
> on the [0,2 PI] range.
> The same error message appears.
> The following is the call stack, which shows how the quapro library
> is used to compute the integral based on a "21-point gauss-kronrod 
> approximation".
>
> differential_equations/sci_gateway/c/sci_intg
> > differential_equations/sci_gateway/fortran/intg.f
> > differential_equations/src/fortran/dqag0.f
> > dqags > quarul
>
> By doing a step-by-step check of the process, it
> appears that the numerical integration is correctly computed (near the
> machine's epsilon), but a secondary check sets a boolean flag to a
> value indicating that the error for that computation is too large.
>
> A simple workaround for your problem may be to decompose the
> range in two parts :
>
> i1 = intg(0,%pi,f);
> i2 = intg(%pi,2*%pi,f);
>
> then the expected result is the sum of i1 and i2.
> Other solutions include traditionnal ode solver (like dassl ) or 
> Monte-Carlo methods.
>
> Best regards,
>
> Micha?l Baudin
>
> Remi G. a ?crit :
>> Hello!
>>
>> I'm facing a strange problem with the function "integrate" (and also 
>> with "intg") with Scilab 4.1.2 :
>>
>> If I want to integrate x^5 between -%pi and %pi, it works (hopefully) :
>>  integrate('t^5','t',-%pi,%pi)
>>
>> But if replace 5 by a variable k (that I've defined before of course) 
>> it will not work for k>=5!
>>  i=5
>>  integrate('t^i','t',-%pi,%pi)
>>
>> Here is the error message I get :
>>    !--error 24
>>  convergence problem...
>>  At line      26 of function integrate called by :
>>  integrate('t^i','t',-%pi,%pi)
>>
>> That's really annoying because I can't code my least square 
>> approximation procedure which is... for tomorrow!
>> If someone could give me a way to go through this issue!
>> Thank you very much!
>>
>> R?mi.
>



From remjg at free.fr  Mon Oct 27 19:50:10 2008
From: remjg at free.fr (remjg at free.fr)
Date: Mon, 27 Oct 2008 19:50:10 +0100
Subject: [scilab-Users] Strange problem with "integrate" : "Convergence problem..." error 24 (Scilab 4.1.2)
Message-ID: <1225133410.49060d627de33@imp.free.fr>

Thank you very much everything is fine now!
I will post a message later, I've to finish my assignment very quickly...


From remjg at free.fr  Mon Oct 27 19:53:21 2008
From: remjg at free.fr (remjg at free.fr)
Date: Mon, 27 Oct 2008 19:53:21 +0100
Subject: How can I copy the output of scilab (scilab 4.1.2 on linux)
Message-ID: <1225133601.49060e2162eee@imp.free.fr>

Hello!

I need to copy the output of a script (.sce) I've made but I can't copy from the
scilab main window (scilex).
I'm using Scilab 4.1.2 on linux!
Thank you very much for you help!

R?mi.


From centro.calculo at siu.udea.edu.co  Mon Oct 27 20:00:29 2008
From: centro.calculo at siu.udea.edu.co (Centro Calculo)
Date: Mon, 27 Oct 2008 14:00:29 -0500 (COT)
Subject: [scilab-Users] How can I copy the output of scilab (scilab
 4.1.2 on linux)
In-Reply-To: <1225133601.49060e2162eee@imp.free.fr>
Message-ID: <1127448635.669591225134029487.JavaMail.root@siu-mail.udea.edu.co>

You had tried using the diary function 

diary outputs

.... yours functions and instructions

diary(0)

John

----- Mensaje original -----
De: remjg at free.fr
Para: users at lists.scilab.org
Enviados: Lunes, 27 de Octubre de 2008 13:53:21 GMT -05:00 Colombia
Asunto: [scilab-Users] How can I copy the output of scilab (scilab 4.1.2 on linux)

Hello!

I need to copy the output of a script (.sce) I've made but I can't copy from the
scilab main window (scilex).
I'm using Scilab 4.1.2 on linux!
Thank you very much for you help!

R?mi.


-- 
-------------------------------------------------------------------------------------------
Centro Regional de Simulaci?n y C?lculo Avanzado, CRESCA
Sede de Investigaci?n Universitaria (SIU)
Clle. 62 No. 52-59 (Peatonal)
Cra. 53 No. 61-30 (Vehicular)
Tels. (574) 219-6612 (CRESCA), (574) 219-6400 (Conmutador)
Fax. (574) 219-6565
e-mail: centro.calculo at siu.udea.edu.co
website: http://siu.udea.edu.co/cresca
-------------------------------------------------------------------------------------------


From sylvestre.ledru at inria.fr  Mon Oct 27 21:24:26 2008
From: sylvestre.ledru at inria.fr (Sylvestre Ledru)
Date: Mon, 27 Oct 2008 21:24:26 +0100
Subject: [scilab-Users] How can I copy the output of scilab (scilab
	4.1.2 on linux)
In-Reply-To: <1225133601.49060e2162eee@imp.free.fr>
References: <1225133601.49060e2162eee@imp.free.fr>
Message-ID: <1225139066.21746.121.camel@zlarin>

Le lundi 27 octobre 2008 ? 19:53 +0100, remjg at free.fr a ?crit :
> Hello!
> 
> I need to copy the output of a script (.sce) I've made but I can't copy from the
> scilab main window (scilex).
> I'm using Scilab 4.1.2 on linux!
Depends what you want to do but you can do:
scilab -nw -f your-great-script.sce 2>&1 > your-file.log

Sylvestre




From debian1sy8 at gmail.com  Mon Oct 27 22:14:47 2008
From: debian1sy8 at gmail.com (Maki 1sy8)
Date: Mon, 27 Oct 2008 22:14:47 +0100
Subject: [scilab-Users] Strange problem with "integrate" : "Convergence problem..." error 24 (Scilab 4.1.2)
In-Reply-To: <4905CE42.5060506@scilab.org>
References: <49052F31.5070607@free.fr> <4905CC0C.6080203@scilab.org>
	 <4905CE42.5060506@scilab.org>
Message-ID: <58061b670810271414o474ba54bi5354cb34574aa68@mail.gmail.com>

On Mon, Oct 27, 2008 at 3:20 PM, Micha?l Baudin
<michael.baudin at scilab.org> wrote:
> Oh, and by the way, this is exactly the method used by integrate, so
> you can use it with several step instead of only one :
>
> x0=0.;
> step = 2*%pi/10
> x1=0:step:2*%pi;
> integrate("sin(x)",'x',x0,x1)
>

Hi all,

So, it seems that integrate computes the integral on each subinterval
and then sums all 'sub'-integrals automatically ?

By the way, which algorithm is used by Scilab for numerical quadrature ?

Regards


From centro.calculo at siu.udea.edu.co  Tue Oct 28 13:51:52 2008
From: centro.calculo at siu.udea.edu.co (Centro Calculo)
Date: Tue, 28 Oct 2008 07:51:52 -0500 (COT)
Subject: Using the GA toolbox
In-Reply-To: <2050511793.673231225198278198.JavaMail.root@siu-mail.udea.edu.co>
Message-ID: <70445483.673251225198312324.JavaMail.root@siu-mail.udea.edu.co>

Hello everybody. I am trying to use the toolbox ga-1.0 but this don't come with any tutorial. Anybody knows how to use this toolbox or knows other toolbox for genetics algorithms?

Cheers

-- 
-------------------------------------------------------------------------------------------
Centro Regional de Simulaci?n y C?lculo Avanzado, CRESCA
Sede de Investigaci?n Universitaria (SIU)
Clle. 62 No. 52-59 (Peatonal)
Cra. 53 No. 61-30 (Vehicular)
Tels. (574) 219-6612 (CRESCA), (574) 219-6400 (Conmutador)
Fax. (574) 219-6565
e-mail: centro.calculo at siu.udea.edu.co
website: http://siu.udea.edu.co/cresca
-------------------------------------------------------------------------------------------


From michael.baudin at scilab.org  Wed Oct 29 17:17:55 2008
From: michael.baudin at scilab.org (=?ISO-8859-1?Q?Micha=EBl_Baudin?=)
Date: Wed, 29 Oct 2008 17:17:55 +0100
Subject: [scilab-Users] Strange problem with "integrate" : "Convergence
 problem..." error 24 (Scilab 4.1.2)
In-Reply-To: <58061b670810271414o474ba54bi5354cb34574aa68@mail.gmail.com>
References: <49052F31.5070607@free.fr> <4905CC0C.6080203@scilab.org>	 <4905CE42.5060506@scilab.org> <58061b670810271414o474ba54bi5354cb34574aa68@mail.gmail.com>
Message-ID: <49088CB3.5000408@scilab.org>

Hi !

Maki 1sy8 a ?crit :
> Hi all,
>
> So, it seems that integrate computes the integral on each subinterval
> and then sums all 'sub'-integrals automatically ?
>   
Exactly.

> By the way, which algorithm is used by Scilab for numerical quadrature ?
>   
The algorithms used are those of the quadpack library (and not quapro, as I
wrote in my previous message). The algorithm used in the previous
example is a 21-point gauss-kronrod quadrature and is located in the
quarul.f source code :
http://gitweb.scilab.org/?p=scilab;a=blob;f=scilab/modules/differential_equations/src/fortran/quarul.f;h=b35ec0fcad2e2e6e72e1e0065d8ed7bcdd5dd013;hb=HEAD

Regards,

Micha?l


From michael.baudin at scilab.org  Wed Oct 29 17:30:13 2008
From: michael.baudin at scilab.org (=?UTF-8?B?TWljaGHDq2wgQmF1ZGlu?=)
Date: Wed, 29 Oct 2008 17:30:13 +0100
Subject: [scilab-Users] Using the GA toolbox
In-Reply-To: <70445483.673251225198312324.JavaMail.root@siu-mail.udea.edu.co>
References: <70445483.673251225198312324.JavaMail.root@siu-mail.udea.edu.co>
Message-ID: <49088F95.4030108@scilab.org>

Hi,

The genetic algorithms toolbox is now included in Scilab v5.

http://gitweb.scilab.org/?p=scilab;a=tree;f=scilab/modules/genetic_algorithms;h=4d524fcb82b2df6b37a6eec9f0c2530f11dcaef7;hb=HEAD

The documentation for this module is available as a regular Scilab help,
or online :
http://www.scilab.org/product/man/category_genetic_algorithms.html

In the modules/genetic_algorithms/examples subdirectory, you may find
several example based on GAs, for example the GAdemo.sce script :

http://gitweb.scilab.org/?p=scilab;a=blob;f=scilab/modules/genetic_algorithms/examples/GAdemo.sce;h=e5c2928e0b3a0dcadf294460d70e9c1d93577db2;hb=HEAD

This should help you to get started.

If you have questions on this toolbox, you may contact me, or
the author of the toolbox, Yann Collette, who regularily
replies to messages in this forum.

Best regards,

Micha?l Baudin

Centro Calculo a ?crit :
> Hello everybody. I am trying to use the toolbox ga-1.0 but this don't come with any tutorial. Anybody knows how to use this toolbox or knows other toolbox for genetics algorithms?
>
> Cheers
>
>