From sylvestre.ledru at scilab.org  Tue Mar  3 09:57:43 2009
From: sylvestre.ledru at scilab.org (Sylvestre Ledru)
Date: Tue, 03 Mar 2009 09:57:43 +0100
Subject: [scilab-Users] Scilab 5.1 : pb launching in gui mode ....
In-Reply-To: <653EE19D6D5DD440877B54B5A3BDD04903308B9A@sctex002.st-cloud.dassault-avion.fr>
References: 	 <653EE19D6D5DD440877B54B5A3BDD04903308B9A@sctex002.st-cloud.dassault-avion.fr>
Message-ID: <1236070663.9397.87.camel@korcula.inria.fr>

Hello,

This bug might be fixed:
http://bugzilla.scilab.org/show_bug.cgi?id=4185

Could you confirm if it is fixed with the nightly build ?

Thanks,
Sylvestre


Le mercredi 25 f?vrier 2009 ? 11:59 +0100, Tran Chanh a ?crit :
> Hi all,
> 
> Effectively, my pb is related to 'cygwin' & not w/ client linux.
> For now, I can 'survive' w/ the 2nd option.
> 
> Nevertheless, I would need sooner or later this to run via cygwin & as I 'm a 'total beginner' could someone report this bug for me, please ?  
> 
> Thx again for your help
> 
> Salutations+,
> Chanh
> 
> -----Message d'origine-----
> De : Vergnes Nicolas [mailto:nicolas.vergnes at steria.cnes.fr] 
> Envoy? : mercredi 25 f?vrier 2009 10:26
> ? : users at lists.scilab.org
> Objet : RE: [scilab-Users] Scilab 5.1 : pb launching in gui mode ....
> 
> Ih,
> 
> We have the same problem in Windows with cygwin's X server (ssh + X11 tunnneling) with ssh and Exceed's X server the gui lanch without problem at start ( problems come latter )
> 
> From a Linux client with ssh we have'nt problem
> 
> Is jogl or other java graphic library in cause ?
> Is Scilab require hardware acceleration support ?
> 
> 
> 
> 	Nicolas Vergnes
> 	Produits & Logiciels
> 	Cnes
> 
> -----Message d'origine-----
> De : Sylvestre Ledru [mailto:sylvestre.ledru at scilab.org]
> Envoy? : mercredi 25 f?vrier 2009 10:16
> ? : users at lists.scilab.org
> Objet : Re: [scilab-Users] Scilab 5.1 : pb launching in gui mode ....
> 
> Hello,
> 
> A better place would have been the bug tracker. ;)
> 
> Are you running Scilab through a SSH connexion (X11 export) ?
> 
> Sylvestre
> 
> For now, I reported here:
> http://bugzilla.scilab.org/show_bug.cgi?id=4185
> 
> Le vendredi 20 f?vrier 2009 ? 15:03 +0100, Tran Chanh a ?crit :
> > Hi all,
> >  
> > Hope this is the right 'place' for my following pb. 
> > I just installed the binary version of 5.1 for Linux 32 bits & when I 
> > launch './bin/scilab' I've the following msgs :
> >  
> > Could not create a Scilab main class. Error:
> > Exception in thread "main" java.lang.ArrayIndexOutOfBoundsException: 0
> >         at
> > org.scilab.modules.renderer.utils.RenderingCapabilities.getMaxWindowSize(Unknown Source)
> >         at
> > org.scilab.modules.gui.bridge.window.SwingScilabWindow.setDims(Unknown
> > Source)
> >         at
> > org.scilab.modules.gui.bridge.window.SwingScilabWindow.<init>(Unknown
> > Source)
> >         at
> > org.scilab.modules.gui.window.ScilabWindowBridge.createWindow(Unknown
> > Source)
> >         at
> > org.scilab.modules.gui.bridge.ScilabBridge.createWindow(Unknown
> > Source)
> >         at org.scilab.modules.gui.window.ScilabWindow.<init>(Unknown
> > Source)
> >         at
> > org.scilab.modules.gui.window.ScilabWindow.createWindow(Unknown
> > Source)
> >         at org.scilab.modules.jvm.Scilab.<init>(Unknown Source)
> > 
> > Could someone tell me what 's wrong ? 
> > Thx in advance for any help.
> >  
> > Regards,
> >  
> > Chanh TRAN
> > DSI/DPSI/SF
> > St Cloud 2000 - 2F098
> > T?l : 67-5521 (int)
> >         33 1 47 11 55 21 (ext)
> > Fax : 33 1 47 11 59 11
> >  
> > 
> 



From Dan_Gill at mksinst.com  Tue Mar  3 10:00:19 2009
From: Dan_Gill at mksinst.com (Dan_Gill at mksinst.com)
Date: Tue, 3 Mar 2009 04:00:19 -0500
Subject: Dan Gill/US/MKS is out of the office.
Message-ID: <OF5DAFB467.1576F455-ON8525756E.003177C8-8525756E.003177C8@mksinst.com>


I will be out of the office starting Fri 02/27/2009 and will not return
until Mon 03/09/2009.

I will respond to your message when I return.



From srnhnsn at gmail.com  Tue Mar  3 13:37:03 2009
From: srnhnsn at gmail.com (Soren Hansen)
Date: Tue, 3 Mar 2009 13:37:03 +0100
Subject: No text in plots / scicos
Message-ID: <b23aaa410903030437v20265f16pbd9e5ee079ec4958@mail.gmail.com>

Hi

I have installed Scilab-5.1 from source on a PC running Linux. When i try
plotting anything i get a plotting window without any text on the axis etc.
Same thing happens when I open Scicos palettes .. the text is not visible.

Do I need to specify a font-path, or is something else wrong.

Any suggestions are appreciated.

- Soren
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20090303/0247992a/attachment.htm>

From sylvestre.ledru at scilab.org  Tue Mar  3 13:44:30 2009
From: sylvestre.ledru at scilab.org (Sylvestre Ledru)
Date: Tue, 03 Mar 2009 13:44:30 +0100
Subject: [scilab-Users] No text in plots / scicos
In-Reply-To: <b23aaa410903030437v20265f16pbd9e5ee079ec4958@mail.gmail.com>
References: <b23aaa410903030437v20265f16pbd9e5ee079ec4958@mail.gmail.com>
Message-ID: <1236084270.9397.109.camel@korcula.inria.fr>

Le mardi 03 mars 2009 ? 13:37 +0100, Soren Hansen a ?crit :
> Hi
> 
> I have installed Scilab-5.1 from source on a PC running Linux. When i
> try plotting anything i get a plotting window without any text on the
> axis etc. Same thing happens when I open Scicos palettes .. the text
> is not visible.
> 
> Do I need to specify a font-path, or is something else wrong.
Depends on the distribution... but it is common that font packages are
not installed by default with the JVM. 
You can install them (if you are running Mandriva, Yann could help you)
or change the font in the Scilab menu.

Sylvestre




From srnhnsn at gmail.com  Tue Mar  3 15:43:55 2009
From: srnhnsn at gmail.com (Soren Hansen)
Date: Tue, 3 Mar 2009 15:43:55 +0100
Subject: [scilab-Users] No text in plots / scicos
In-Reply-To: <1236084270.9397.109.camel@korcula.inria.fr>
References: <b23aaa410903030437v20265f16pbd9e5ee079ec4958@mail.gmail.com>
	 <1236084270.9397.109.camel@korcula.inria.fr>
Message-ID: <b23aaa410903030643y777fe0aeu601f0202d37a9f40@mail.gmail.com>

On Tue, Mar 3, 2009 at 1:44 PM, Sylvestre Ledru
<sylvestre.ledru at scilab.org>wrote:

> Le mardi 03 mars 2009 ? 13:37 +0100, Soren Hansen a ?crit :
> > Hi
> >
> > I have installed Scilab-5.1 from source on a PC running Linux. When i
> > try plotting anything i get a plotting window without any text on the
> > axis etc. Same thing happens when I open Scicos palettes .. the text
> > is not visible.
> >
> > Do I need to specify a font-path, or is something else wrong.
> Depends on the distribution... but it is common that font packages are
> not installed by default with the JVM.
> You can install them (if you are running Mandriva, Yann could help you)
> or change the font in the Scilab menu.
>

Ok, thanks.

I found a solution. I am using Ubuntu 8.04 with the and I had installed both
Suns java environment and OpenJDK, but compiled Scilab with OpenJDK. When
rebuilding Scilab with Suns java I got the axis texts correct. Apparently I
am missing some fonts in OpenJDK

- Soren
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20090303/6029fa55/attachment.htm>

From xr at geo.hmg.inpg.fr  Thu Mar  5 22:02:28 2009
From: xr at geo.hmg.inpg.fr (xavier roguiez)
Date: Thu, 05 Mar 2009 22:02:28 +0100
Subject: No data plots ?
Message-ID: <49B03DE4.3000104@geo.hmg.inpg.fr>

Hello,

I've developped some macro who was running perfectly under scilab 4.1.2.
I've tried to run them under scilab 5.1. But it doesn't work.

The calculations seems to be done, but nothing appear on figure.
I attempt some sample with the command plot2d or plot, but the screen is 
desesperaly white...

Does anybody has encountered same problems ?

Thanks in advance for any solution
xavier


From xr at geo.hmg.inpg.fr  Thu Mar  5 22:07:31 2009
From: xr at geo.hmg.inpg.fr (xavier roguiez)
Date: Thu, 05 Mar 2009 22:07:31 +0100
Subject: [scilab-Users] No data plots ?
In-Reply-To: <49B03DE4.3000104@geo.hmg.inpg.fr>
References: <49B03DE4.3000104@geo.hmg.inpg.fr>
Message-ID: <49B03F13.9060909@geo.hmg.inpg.fr>

xavier roguiez a ?crit :
> Hello,
>
> I've developped some macro who was running perfectly under scilab 4.1.2.
> I've tried to run them under scilab 5.1. But it doesn't work.
>
> The calculations seems to be done, but nothing appear on figure.
> I attempt some sample with the command plot2d or plot, but the screen 
> is desesperaly white...
>
> Does anybody has encountered same problems ?
>
> Thanks in advance for any solution
> xavier
>
Sorry, I found new information on my problem :

 ATTENTION : A cause des limitations des pilotes de votre carte 
graphique, qui ne sont pas capable de g?rer OpenGL, Scilab ne pourra pas
       tracer de graphiques. Veuillez mettre vos pilotes ? 
jour.                                                                         


xavier


From steeve102000 at yahoo.com  Fri Mar  6 05:20:25 2009
From: steeve102000 at yahoo.com (Stephen Thomas)
Date: Thu, 5 Mar 2009 20:20:25 -0800 (PST)
Subject: mopen command help needed
Message-ID: <987881.65747.qm@web111313.mail.gq1.yahoo.com>

hi im a newuser to scilab..not much prior experience in any other equivalent software too

i want to open a file for reading a set of x y z values from scilab
 i copeied the code frm
?
?http://spiff.rit.edu/classes/phys317/lectures/readwrite.html

Code

 fid = mopen(filename, "r");
  if (fid == -1)
    error("cannot open file for reading");
  end

to scipad editor and tried to execute via  "Load into scilab" i had created an empty file in home directory named "filename "via gedit
before execution .
While execution says "Undefined variable 'filename'
then i changed the code to 
 
fid = mopen("filename", "r");

ie added " " before and after filename then it executes..
what is the problem here i cant understand it..I just copied it as it is from
?http://spiff.rit.edu/classes/phys317/lectures/readwrite.html
Is the syntax used here is wrong??
Where should i get the correct syntax for the function mopen and other functions then?

i tried 'help mopen'
and it displayed

Calling sequence

[fd.err]=mopen( file [,mode ,swap]) 

Is this the proper syntax ?? I dont think so ,where should i get the corret sequence then?
 somebody please help..
Thanks





--------------------------------------------------------------------------------------------------? "Be good, be kind, be humane, and charitable; love your fellows; ? console the afflicted; pardon those who have done you wrong."? Zoroaster.






      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20090305/fac5d7b4/attachment.htm>

From Francis.Drossaert at pgs.com  Fri Mar  6 10:25:22 2009
From: Francis.Drossaert at pgs.com (Francis Drossaert)
Date: Fri, 6 Mar 2009 09:25:22 -0000
Subject: [scilab-Users] mopen command help needed
In-Reply-To: <987881.65747.qm@web111313.mail.gq1.yahoo.com>
References: <987881.65747.qm@web111313.mail.gq1.yahoo.com>
Message-ID: <6D1269A4C2D21240807B91421B187255068DF63F@lonms21.onshore.pgs.com>

Hi Stephen,

 

1. Make sure that the filename is correct. If your data file is called
xyz.dat then the first command should be fid = mopen('xyz.dat','r') or
you do: filename  =  'xyz.dat'; fid =  mopen(filename,'r');

2. Make sure that the data file is in the current working directory or
make sure that the path is correct for example: fid =
mopen('../data/xyz.dat','r').

 

Cheers,

 

Francis

 

 

From: Stephen Thomas [mailto:steeve102000 at yahoo.com] 
Sent: 06 March 2009 04:20
To: users at lists.scilab.org
Subject: [scilab-Users] mopen command help needed

 

hi im a newuser to scilab..not much prior experience in any other
equivalent software too

i want to open a file for reading a set of x y z values from scilab

   i copeied the code frm

   

 http://spiff.rit.edu/classes/phys317/lectures/readwrite.html

  

  Code

  

   fid = mopen(filename, "r");

  if (fid == -1)

    error("cannot open file for reading");

  end

  

to scipad editor and tried to execute via  "Load into scilab" i had
created an empty file in home directory named "filename "via gedit

before execution .

While execution says "Undefined variable 'filename'

then i changed the code to 

   

fid = mopen("filename", "r");

  

ie added " " before and after filename then it executes..

what is the problem here i cant understand it...I just copied it as it
is from

 http://spiff.rit.edu/classes/phys317/lectures/readwrite.html

Is the syntax used here is wrong??

Where should i
 get the correct syntax for the function mopen and other functions then?

  

i tried 'help mopen'

  and it displayed

  

Calling sequence

  

[fd.err]=mopen( file [,mode ,swap]) 

  

  Is this the proper syntax ?? I dont think so ,where should i get the
corret sequence then?

 somebody please help..

Thanks

  

  

  

  

  

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

  

"Be good, be kind, be humane, and charitable; love your fellows; 

  

console the afflicted; pardon those who have done you wrong."

  

Zoroaster. 



 



 <http://ubuntucounter.geekosophical.net> 

 


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/20090306/7c1b1f64/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ~WRD000.jpg
Type: image/jpeg
Size: 823 bytes
Desc: ~WRD000.jpg
URL: <https://lists.scilab.org/pipermail/users/attachments/20090306/7c1b1f64/attachment.jpg>

From ascv at computer.org  Fri Mar  6 11:24:58 2009
From: ascv at computer.org (Prof. Ausberto S. Castro V. - UNASP-SP)
Date: Fri, 06 Mar 2009 07:24:58 -0300
Subject: [scilab-Users] No data plots ?
In-Reply-To: <49B03DE4.3000104@geo.hmg.inpg.fr>
References: <49B03DE4.3000104@geo.hmg.inpg.fr>
Message-ID: <49b0f9fc.2535640a.724c.1921@mx.google.com>

At 18:02 5/3/2009, xr at geo.hmg.inpg.fr  wrote:
>I've developped some macro who was running perfectly under scilab 4.1.2.
>I've tried to run them under scilab 5.1. But it doesn't work.
>
>The calculations seems to be done, but nothing appear on figure.
>I attempt some sample with the command plot2d or 
>plot, but the screen is desesperaly white...
>Does anybody has encountered same problems ?

Yes, I have the same problem...! Any plot ... crash!


==========================================================
Prof. Ausberto S. Castro V.                             ascv at computer.org
Ph.D. Computer Science                          Ausberto.Castro at unasp.edu.br
UNASP - Centro Universit?rio Adventista de S?o Paulo
Sao Paulo - SP
Brazil
        http:/www.unasp.edu.br/
==========================================================

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20090306/16bcee06/attachment.htm>

From steeve102000 at yahoo.com  Fri Mar  6 11:26:17 2009
From: steeve102000 at yahoo.com (Stephen Thomas)
Date: Fri, 6 Mar 2009 02:26:17 -0800 (PST)
Subject: [scilab-Users] mopen command help needed
In-Reply-To: <6D1269A4C2D21240807B91421B187255068DF63F@lonms21.onshore.pgs.com>
Message-ID: <914012.36548.qm@web111308.mail.gq1.yahoo.com>

Thank You Francis I got it

--------------------------------------------------------------------------------------------------? "Be good, be kind, be humane, and charitable; love your fellows; ? console the afflicted; pardon those who have done you wrong."? Zoroaster.





--- On Fri, 3/6/09, Francis Drossaert <Francis.Drossaert at pgs.com> wrote:
From: Francis Drossaert <Francis.Drossaert at pgs.com>
Subject: RE: [scilab-Users] mopen command help needed
To: users at lists.scilab.org
Date: Friday, March 6, 2009, 2:55 PM




 
 







Hi Stephen, 

 ? 

1. Make sure that the filename is correct. If your data file is
called xyz.dat then the first command should be fid = mopen(?xyz.dat?,?r?)
or you do: filename ?= ??xyz.dat?; fid =? mopen(filename,?r?); 

2. Make sure that the data file is in the current working directory
or make sure that the path is correct for example: fid = mopen(?.../data/xyz.dat?,?r?). 

 ? 

Cheers, 

 ? 

Francis 

 ? 

 ? 



From: Stephen Thomas
[mailto:steeve102000 at yahoo.com] 

Sent: 06 March 2009 04:20

To: users at lists.scilab.org

Subject: [scilab-Users] mopen command help needed 



 ? 


 
  
  hi im a newuser to
  scilab..not much prior experience in any other equivalent software too 
  i want to open a file for reading a set of x y z values from scilab

  ?i copeied the code frm

  ?

?http://spiff.rit.edu/classes/phys317/lectures/readwrite.html

  

  Code

  

  ?fid = mopen(filename, "r");

? if (fid == -1)

??? error("cannot open file for reading");

? end

  

to scipad editor and tried to execute via? "Load into scilab" i had created an empty file in home directory named "filename "via gedit

before execution .

While execution says "Undefined variable 'filename'

then i changed the code to 

  ?

fid = mopen("filename", "r");

  

ie added " " before and after filename then it executes..

what is the problem here i cant understand it...I just copied it as it is from

?http://spiff.rit.edu/classes/phys317/lectures/readwrite.html

Is the syntax used here is wrong??

Where should i get the correct syntax for the function mopen and other functions then?

  

i tried 'help mopen'

  and it displayed

  

Calling sequence

  

[fd.err]=mopen( file [,mode ,swap]) 

  

  Is this the proper syntax ?? I dont think so ,where should i get the corret sequence then?

?somebody please help..

Thanks

  

  

  

  

   
  
  -------------------------------------------------------------------------------------------------- 
  
  ?  
  
  "Be
  good, be kind, be humane, and charitable; love your fellows;  
  
  ?  
  
  console
  the afflicted; pardon those who have done you wrong." 
  
  ?  
  
  Zoroaster.

  

   
  
   ? 
  
   
  
  
   
  
  
 


 ? 



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/20090306/4bbc90c7/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ~WRD000.jpg
Type: image/jpeg
Size: 823 bytes
Desc: ~WRD000.jpg
URL: <https://lists.scilab.org/pipermail/users/attachments/20090306/4bbc90c7/attachment.jpg>

From xr at geo.hmg.inpg.fr  Fri Mar  6 11:29:05 2009
From: xr at geo.hmg.inpg.fr (xavier roguiez)
Date: Fri, 06 Mar 2009 11:29:05 +0100
Subject: [scilab-Users] No data plots ?
In-Reply-To: <49b0f9fc.2535640a.724c.1921@mx.google.com>
References: <49B03DE4.3000104@geo.hmg.inpg.fr> <49b0f9fc.2535640a.724c.1921@mx.google.com>
Message-ID: <49B0FAF1.7070908@geo.hmg.inpg.fr>

Prof. Ausberto S. Castro V. - UNASP-SP a ?crit :
> At 18:02 5/3/2009, xr at geo.hmg.inpg.fr  wrote:
>> I've developped some macro who was running perfectly under scilab 4.1.2.
>> I've tried to run them under scilab 5.1. But it doesn't work.
>>
>> The calculations seems to be done, but nothing appear on figure.
>> I attempt some sample with the command plot2d or plot, but the screen 
>> is desesperaly white...
>> Does anybody has encountered same problems ?
>
> Yes, I have the same problem...! Any plot ... crash!
>
I've updated the driver Nvidia who was unable to read OpenGL, and now 
it's works perfectly.

xavier


From xr at geo.hmg.inpg.fr  Fri Mar  6 11:37:21 2009
From: xr at geo.hmg.inpg.fr (xavier roguiez)
Date: Fri, 06 Mar 2009 11:37:21 +0100
Subject: Poor quality
Message-ID: <49B0FCE1.4040807@geo.hmg.inpg.fr>

Hello

Is there something to change in order to get better quality on graphics 
export ?
With the 4.1.2 scilab version, i was plotting  the data and exporting in 
eps, then converting into pdf, you can see the result in graph412.pdf
With the 5.1, the same result is very poor, as you can see it on the 
result in graph51.pdf

I suspect some problems with the fonts ? But don't know where to do it ?

Does anybody can help me ?

thanks in advance
xavier
-------------- next part --------------
A non-text attachment was scrubbed...
Name: graph412.pdf
Type: application/pdf
Size: 7296 bytes
Desc: not available
URL: <https://lists.scilab.org/pipermail/users/attachments/20090306/5e86caff/attachment.pdf>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: graph51.pdf
Type: application/pdf
Size: 18532 bytes
Desc: not available
URL: <https://lists.scilab.org/pipermail/users/attachments/20090306/5e86caff/attachment-0001.pdf>

From steeve102000 at yahoo.com  Fri Mar  6 11:47:21 2009
From: steeve102000 at yahoo.com (Stephen Thomas)
Date: Fri, 6 Mar 2009 02:47:21 -0800 (PST)
Subject: creating a 3d plot please help
Message-ID: <843702.5973.qm@web111316.mail.gq1.yahoo.com>


Hi all ,
????????? Im trying to create a 3d plot..by reading data from a file..
The datafile has 3 set of values i ve randomly put.
say like this..
X??? Y????? Z
1??? 2????? 3.5
3??? 6.0??? 8.4
7??? 9?? ??? 3

i want to plot the surface? covered by these values..

i used plot3d1(X ,Y,Z) command and i got a rectagular prism shaped surface...with the faces being drawn through outermost set of xyz points

Is there any free suface plotting function in scilab..other than this ractangular facets..

Or can any one tell me which is the best method of plotting such type of non related 3 D data??
Thanks
Stephen


--------------------------------------------------------------------------------------------------? "Be good, be kind, be humane, and charitable; love your fellows; ? console the afflicted; pardon those who have done you wrong."? Zoroaster.






      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20090306/922b7689/attachment.htm>

From sylvestre.ledru at scilab.org  Fri Mar  6 19:42:15 2009
From: sylvestre.ledru at scilab.org (Sylvestre Ledru)
Date: Fri, 06 Mar 2009 19:42:15 +0100
Subject: [scilab-Users] Poor quality
In-Reply-To: <49B0FCE1.4040807@geo.hmg.inpg.fr>
References: <49B0FCE1.4040807@geo.hmg.inpg.fr>
Message-ID: <1236364935.9397.381.camel@korcula.inria.fr>

Le vendredi 06 mars 2009 ? 11:37 +0100, xavier roguiez a ?crit :
> Hello
> 
> Is there something to change in order to get better quality on graphics 
> export ?
> With the 4.1.2 scilab version, i was plotting  the data and exporting in 
> eps, then converting into pdf, you can see the result in graph412.pdf
> With the 5.1, the same result is very poor, as you can see it on the 
> result in graph51.pdf
> 
> I suspect some problems with the fonts ? But don't know where to do it ?
At least a part of your issues is probably caused by this bug:
http://bugzilla.scilab.org/show_bug.cgi?id=4173

Sylvestre




From pierre.juillard at gmail.com  Sat Mar  7 00:42:31 2009
From: pierre.juillard at gmail.com (Pierre JUILLARD)
Date: Sat, 7 Mar 2009 00:42:31 +0100
Subject: [scilab-Users] creating a 3d plot please help
In-Reply-To: <843702.5973.qm@web111316.mail.gq1.yahoo.com>
References: <843702.5973.qm@web111316.mail.gq1.yahoo.com>
Message-ID: <fac5e9bb0903061542u69cbb574s79ce09a13ef7f989@mail.gmail.com>

Hi Stephens,


I can't readily tell how to do in Scilab, but I think you miss a problem (or
maybe it's me...).

The thing is, your example shows a point cloud, not a mesh: you can't
directly get a surface from a point cloud...
How would Scilab or any other software know how to link the correct set of
points between themselves?

However, if your point cloud is one that would correspond to a "relatively
flat" shell mesh (in the sense it's not a compact ball of points everywhere
for instance...), you can use Delaunay3D meshing to have a triangular mesh
corresponding to a surface.

I did a similar thing with ParaView from 3D scans. I wrote the coordinates
in a file understandable by ParaView (csv I think to remember), then I used
the Delaunay3D filter available.
It worked rather well.


Hope I help.

Pierre



2009/3/6 Stephen Thomas <steeve102000 at yahoo.com>

>
> Hi all ,
>           Im trying to create a 3d plot..by reading data from a file..
> The datafile has 3 set of values i ve randomly put.
> say like this..
> X    Y      Z
> 1    2      3.5
> 3    6.0    8.4
> 7    9       3
>
> i want to plot the surface  covered by these values..
>
> i used plot3d1(X ,Y,Z) command and i got a rectagular prism shaped
> surface..with the faces being drawn through outermost set of xyz points
>
> Is there any free suface plotting function in scilab..other than this
> ractangular facets..
>
> Or can any one tell me which is the best method of plotting such type of
> non related 3 D data??
> Thanks
> Stephen
>
>
> *
> --------------------------------------------------------------------------------------------------
> *
>   *"Be good, be kind, be humane, and charitable; love your fellows; *
>   *console the afflicted; pardon those who have done you wrong."*
>   *Zoroaster.
>
> *
>
> [image: The Ubuntu Counter Project - user number # 22904]<http://ubuntucounter.geekosophical.net>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20090307/d3fd3534/attachment.htm>

From d.tschische at cisc.at  Tue Mar 10 15:28:50 2009
From: d.tschische at cisc.at (David Tschische)
Date: Tue, 10 Mar 2009 15:28:50 +0100
Subject: How to replace Matlab stairs function in Scilab
Message-ID: <001801c9a18c$87ec6640$97c532c0$@tschische@cisc.at>

Hi Folks,

I just started to use Scilab and I'm struggling a little bit as I have to
convert some of my scripts from Matlab to Scilab.

Is there any equivalent to Matlab "stairs" function or any suitable way to
draw staircase like steps?

In particular, I would like to have something similar to:

	stairs(rec(:,2),rec(:,1),'DisplayName','TAG0
Receiver','Parent',axes1);

Any suggestions?

Regards,

Dave

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20090310/27db3ad3/attachment.htm>

From michel.audette at kitware.com  Tue Mar 10 20:17:24 2009
From: michel.audette at kitware.com (Michel Audette)
Date: Tue, 10 Mar 2009 15:17:24 -0400
Subject: mex examples
Message-ID: <b16f21b00903101217u5416b1adieba5e1a64d33bcb2@mail.gmail.com>

Hello,

I have some matlab code provided to me by a collaborator, which also
includes c/cpp files. The Scilab wiki alludes to mex examples, but I cannot
find these examples with my copy of version 5.0.3. Where can I find examples
that deal with mex, as well as online instructions?

Best wishes,

-- 
Michel Audette, Ph.D.
R & D Engineer,
Kitware Inc.,
Chapel Hill, N.C.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20090310/31ff299f/attachment.htm>

From michel.audette at kitware.com  Tue Mar 10 21:07:46 2009
From: michel.audette at kitware.com (Michel Audette)
Date: Tue, 10 Mar 2009 16:07:46 -0400
Subject: mex examples
In-Reply-To: <b16f21b00903101217u5416b1adieba5e1a64d33bcb2@mail.gmail.com>
References: <b16f21b00903101217u5416b1adieba5e1a64d33bcb2@mail.gmail.com>
Message-ID: <b16f21b00903101307g2b11d59fq5bcc9770b464ff25@mail.gmail.com>

Hello again,

I've found my answer in the instructions on creating a toolbox.

On another note, when I run mfile2sci, I get
(<unknown>:8633): Gtk-CRITICAL **: gtk_paint_box: assertion `style->depth ==
gdk_drawable_get_depth (window)' failed

and the interface does seem to run properly. Also, the directory of the
input filename is not saved properly.

I have scilab 5.0.3 . Are these issues fixed with the latest release? Is
there some other library that is causing the first problem?

Best wishes,

Michel

On Tue, Mar 10, 2009 at 3:17 PM, Michel Audette
<michel.audette at kitware.com>wrote:

> Hello,
>
> I have some matlab code provided to me by a collaborator, which also
> includes c/cpp files. The Scilab wiki alludes to mex examples, but I cannot
> find these examples with my copy of version 5.0.3. Where can I find examples
> that deal with mex, as well as online instructions?
>
> Best wishes,
>
> --
> Michel Audette, Ph.D.
> R & D Engineer,
> Kitware Inc.,
> Chapel Hill, N.C.
>
>


-- 
Michel Audette, Ph.D.
R & D Engineer,
Kitware Inc.,
Chapel Hill, N.C.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20090310/eaea6e20/attachment.htm>

From sylvestre.ledru at scilab.org  Wed Mar 11 15:08:59 2009
From: sylvestre.ledru at scilab.org (Sylvestre Ledru)
Date: Wed, 11 Mar 2009 15:08:59 +0100
Subject: [scilab-Users] Re: mex examples
In-Reply-To: <b16f21b00903101307g2b11d59fq5bcc9770b464ff25@mail.gmail.com>
References: <b16f21b00903101217u5416b1adieba5e1a64d33bcb2@mail.gmail.com>
	 <b16f21b00903101307g2b11d59fq5bcc9770b464ff25@mail.gmail.com>
Message-ID: <1236780539.11164.253.camel@zlarin>

Hello Michel,

Le mardi 10 mars 2009 ? 16:07 -0400, Michel Audette a ?crit :
> Hello again, 
> 
> I've found my answer in the instructions on creating a toolbox. 
> 
> On another note, when I run mfile2sci, I get 
> (<unknown>:8633): Gtk-CRITICAL **: gtk_paint_box: assertion
> `style->depth == gdk_drawable_get_depth (window)' failed
> 
> and the interface does seem to run properly.
It is just a warning. It should not be an issue.

>  Also, the directory of the input filename is not saved properly. 
What do you mean here ?


> I have scilab 5.0.3 . Are these issues fixed with the latest release?
> Is there some other library that is causing the first problem?
Could you be a bit more specific ? Like showing a piece of code showing
your issue.

Regards,
Sylvestre

> Best wishes, 
> 
> Michel
> 
> On Tue, Mar 10, 2009 at 3:17 PM, Michel Audette
> <michel.audette at kitware.com> wrote:
>         Hello, 
>         
>         I have some matlab code provided to me by a collaborator,
>         which also includes c/cpp files. The Scilab wiki alludes to
>         mex examples, but I cannot find these examples with my copy of
>         version 5.0.3. Where can I find examples that deal with mex,
>         as well as online instructions?
>         
>         Best wishes, 
>         
>         -- 
>         Michel Audette, Ph.D. 
>         R & D Engineer, 
>         Kitware Inc.,
>         Chapel Hill, N.C. 
>         
> 
> 
> 
> -- 
> Michel Audette, Ph.D. 
> R & D Engineer, 
> Kitware Inc.,
> Chapel Hill, N.C. 
> 



From mathieu.dubois at limsi.fr  Wed Mar 11 21:33:54 2009
From: mathieu.dubois at limsi.fr (Mathieu Dubois)
Date: Wed, 11 Mar 2009 21:33:54 +0100 (CET)
Subject: [scilab-Users] How to replace Matlab stairs function in Scilab
In-Reply-To: <001801c9a18c$87ec6640$97c532c0$@tschische@cisc.at>
References: <001801c9a18c$87ec6640$97c532c0$@tschische@cisc.at>
Message-ID: <46681.82.123.141.13.1236803634.squirrel@keo.limsi.fr>

Hi David,

> I just started to use Scilab and I'm struggling a little bit as I have to
> convert some of my scripts from Matlab to Scilab.
>
> Is there any equivalent to Matlab "stairs" function or any suitable way to
> draw staircase like steps?
>

If I understand correctly matlab's stairs function
(http://www.math.carleton.ca/old/help/matlab/MathWorks_R13Doc/techdoc/ref/stairs.html)
you should use plot2d2.

Try:
--> x = 0:.25:10;
--> plot2d2(x,sin(x));

BTW I have found this very simply: type 'step' in the 'search' tab of
scilab's help.

Technically speaking plot2d2 is just a plot2d with a special
polyline_style (see help :).

> In particular, I would like to have something similar to:
>
> 	stairs(rec(:,2),rec(:,1),'DisplayName','TAG0
> Receiver','Parent',axes1);
>

I don't know exactly what these options are supposed to do (I don't have
matlab) but you should be able to find out the right options to plot2d2.

Hope that helps (if it's still needed),
Mathieu



From d.tschische at cisc.at  Thu Mar 12 08:10:57 2009
From: d.tschische at cisc.at (David Tschische)
Date: Thu, 12 Mar 2009 08:10:57 +0100
Subject: [scilab-Users] How to replace Matlab stairs function in Scilab
In-Reply-To: <46681.82.123.141.13.1236803634.squirrel@keo.limsi.fr>
References: <001801c9a18c$87ec6640$97c532c0$@tschische@cisc.at> <46681.82.123.141.13.1236803634.squirrel@keo.limsi.fr>
Message-ID: <000301c9a2e1$b13ff5b0$13bfe110$@tschische@cisc.at>

Thank you Mathieu, "plot2d2" seems to be the answer. 
At least it support most parameters of the Matlab pendent in some way.

Another way I just discovered is using "plot2d" in combination with setting
the polyline settings:

handle = get("hdl");
handle.children(1).polyline_style = 2;

Regards,

Dave


> -----Original Message-----
> From: Mathieu Dubois [mailto:mathieu.dubois at limsi.fr]
> Sent: Mittwoch, 11. M?rz 2009 21:34
> To: users at lists.scilab.org
> Subject: Re: [scilab-Users] How to replace Matlab stairs function in
> Scilab
> 
> Hi David,
> 
> > I just started to use Scilab and I'm struggling a little bit as I
> have to
> > convert some of my scripts from Matlab to Scilab.
> >
> > Is there any equivalent to Matlab "stairs" function or any suitable
> way to
> > draw staircase like steps?
> >
> 
> If I understand correctly matlab's stairs function
> (http://www.math.carleton.ca/old/help/matlab/MathWorks_R13Doc/techdoc/r
> ef/stairs.html)
> you should use plot2d2.
> 
> Try:
> --> x = 0:.25:10;
> --> plot2d2(x,sin(x));
> 
> BTW I have found this very simply: type 'step' in the 'search' tab of
> scilab's help.
> 
> Technically speaking plot2d2 is just a plot2d with a special
> polyline_style (see help :).
> 
> > In particular, I would like to have something similar to:
> >
> > 	stairs(rec(:,2),rec(:,1),'DisplayName','TAG0
> > Receiver','Parent',axes1);
> >
> 
> I don't know exactly what these options are supposed to do (I don't
> have
> matlab) but you should be able to find out the right options to
> plot2d2.
> 
> Hope that helps (if it's still needed),
> Mathieu




From mathieu.dubois at limsi.fr  Thu Mar 12 14:49:45 2009
From: mathieu.dubois at limsi.fr (Mathieu Dubois)
Date: Thu, 12 Mar 2009 14:49:45 +0100
Subject: [scilab-Users] How to replace Matlab stairs function in Scilab
In-Reply-To: <000301c9a2e1$b13ff5b0$13bfe110$@tschische@cisc.at>
References: <001801c9a18c$87ec6640$97c532c0$@tschische@cisc.at> <46681.82.123.141.13.1236803634.squirrel@keo.limsi.fr> <000301c9a2e1$b13ff5b0$13bfe110$@tschische@cisc.at>
Message-ID: <49B912F9.9010202@limsi.fr>

Hi David,
> Thank you Mathieu, "plot2d2" seems to be the answer. 
>   
Glad to help you.
> At least it support most parameters of the Matlab pendent in some way
> Another way I just discovered is using "plot2d" in combination with setting
> the polyline settings:
>
> handle = get("hdl");
> handle.children(1).polyline_style = 2;
>   
This is essentially what plot2d2 does. Extract from the doc:

Note that all the modes proposed by plot2dxx (xx = 1 to 4) can be enabled using plot2d
and setting the polyline_style option to the corresponding number.

So the options to plot2d2 are the options to plot2d... And the way you 
discovered is as valid as plot2d2.

For your example you could try:

--> plot2d2(rec(:,2), rec(:,1),leg='TAG0 Receiver')  // leg is for legend

I don't understand what the parent property is.

Regards,
Mathieu
>   
>> -----Original Message-----
>> From: Mathieu Dubois [mailto:mathieu.dubois at limsi.fr]
>> Sent: Mittwoch, 11. M?rz 2009 21:34
>> To: users at lists.scilab.org
>> Subject: Re: [scilab-Users] How to replace Matlab stairs function in
>> Scilab
>>
>> Hi David,
>>
>>     
>>> I just started to use Scilab and I'm struggling a little bit as I
>>>       
>> have to
>>     
>>> convert some of my scripts from Matlab to Scilab.
>>>
>>> Is there any equivalent to Matlab "stairs" function or any suitable
>>>       
>> way to
>>     
>>> draw staircase like steps?
>>>
>>>       
>> If I understand correctly matlab's stairs function
>> (http://www.math.carleton.ca/old/help/matlab/MathWorks_R13Doc/techdoc/r
>> ef/stairs.html)
>> you should use plot2d2.
>>
>> Try:
>> --> x = 0:.25:10;
>> --> plot2d2(x,sin(x));
>>
>> BTW I have found this very simply: type 'step' in the 'search' tab of
>> scilab's help.
>>
>> Technically speaking plot2d2 is just a plot2d with a special
>> polyline_style (see help :).
>>
>>     
>>> In particular, I would like to have something similar to:
>>>
>>> 	stairs(rec(:,2),rec(:,1),'DisplayName','TAG0
>>> Receiver','Parent',axes1);
>>>
>>>       
>> I don't know exactly what these options are supposed to do (I don't
>> have
>> matlab) but you should be able to find out the right options to
>> plot2d2.
>>
>> Hope that helps (if it's still needed),
>> Mathieu
>>     
>
>
>   

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20090312/2ec4f54d/attachment.htm>

From nicolas.vergnes at steria.cnes.fr  Fri Mar 13 08:17:48 2009
From: nicolas.vergnes at steria.cnes.fr (Vergnes Nicolas)
Date: Fri, 13 Mar 2009 08:17:48 +0100
Subject: [scilab-Users] Scilab 5.1 : pb launching in gui mode ....
In-Reply-To: <1236070663.9397.87.camel@korcula.inria.fr>
Message-ID: <EC2C1F912EFD8F498B04551CD2B0408D0480CAF9@cst-xch-002.cnesnet.ad.cnes.fr>

Hi

Yes 32 version is nightly build is working fine now in remote access !
Is a stable version envisaged in next weeks ?

Regards,
Nicolas Vergnes
Produits & Logiciels
Cnes

-----Message d'origine-----
De : Sylvestre Ledru [mailto:sylvestre.ledru at scilab.org] 
Envoy? : mardi 3 mars 2009 09:58
? : users at lists.scilab.org
Objet : RE: [scilab-Users] Scilab 5.1 : pb launching in gui mode ....

Hello,

This bug might be fixed:
http://bugzilla.scilab.org/show_bug.cgi?id=4185

Could you confirm if it is fixed with the nightly build ?

Thanks,
Sylvestre


Le mercredi 25 f?vrier 2009 ? 11:59 +0100, Tran Chanh a ?crit :
> Hi all,
> 
> Effectively, my pb is related to 'cygwin' & not w/ client linux.
> For now, I can 'survive' w/ the 2nd option.
> 
> Nevertheless, I would need sooner or later this to run via cygwin & as I 'm a 'total beginner' could someone report this bug for me, please ?  
> 
> Thx again for your help
> 
> Salutations+,
> Chanh
> 
> -----Message d'origine-----
> De : Vergnes Nicolas [mailto:nicolas.vergnes at steria.cnes.fr]
> Envoy? : mercredi 25 f?vrier 2009 10:26 ? : users at lists.scilab.org 
> Objet : RE: [scilab-Users] Scilab 5.1 : pb launching in gui mode ....
> 
> Ih,
> 
> We have the same problem in Windows with cygwin's X server (ssh + X11 
> tunnneling) with ssh and Exceed's X server the gui lanch without 
> problem at start ( problems come latter )
> 
> From a Linux client with ssh we have'nt problem
> 
> Is jogl or other java graphic library in cause ?
> Is Scilab require hardware acceleration support ?
> 
> 
> 
> 	Nicolas Vergnes
> 	Produits & Logiciels
> 	Cnes
> 
> -----Message d'origine-----
> De : Sylvestre Ledru [mailto:sylvestre.ledru at scilab.org]
> Envoy? : mercredi 25 f?vrier 2009 10:16 ? : users at lists.scilab.org 
> Objet : Re: [scilab-Users] Scilab 5.1 : pb launching in gui mode ....
> 
> Hello,
> 
> A better place would have been the bug tracker. ;)
> 
> Are you running Scilab through a SSH connexion (X11 export) ?
> 
> Sylvestre
> 
> For now, I reported here:
> http://bugzilla.scilab.org/show_bug.cgi?id=4185
> 
> Le vendredi 20 f?vrier 2009 ? 15:03 +0100, Tran Chanh a ?crit :
> > Hi all,
> >  
> > Hope this is the right 'place' for my following pb. 
> > I just installed the binary version of 5.1 for Linux 32 bits & when 
> > I launch './bin/scilab' I've the following msgs :
> >  
> > Could not create a Scilab main class. Error:
> > Exception in thread "main" java.lang.ArrayIndexOutOfBoundsException: 0
> >         at
> > org.scilab.modules.renderer.utils.RenderingCapabilities.getMaxWindowSize(Unknown Source)
> >         at
> > org.scilab.modules.gui.bridge.window.SwingScilabWindow.setDims(Unkno
> > wn
> > Source)
> >         at
> > org.scilab.modules.gui.bridge.window.SwingScilabWindow.<init>(Unknow
> > n
> > Source)
> >         at
> > org.scilab.modules.gui.window.ScilabWindowBridge.createWindow(Unknow
> > n
> > Source)
> >         at
> > org.scilab.modules.gui.bridge.ScilabBridge.createWindow(Unknown
> > Source)
> >         at org.scilab.modules.gui.window.ScilabWindow.<init>(Unknown
> > Source)
> >         at
> > org.scilab.modules.gui.window.ScilabWindow.createWindow(Unknown
> > Source)
> >         at org.scilab.modules.jvm.Scilab.<init>(Unknown Source)
> > 
> > Could someone tell me what 's wrong ? 
> > Thx in advance for any help.
> >  
> > Regards,
> >  
> > Chanh TRAN
> > DSI/DPSI/SF
> > St Cloud 2000 - 2F098
> > T?l : 67-5521 (int)
> >         33 1 47 11 55 21 (ext)
> > Fax : 33 1 47 11 59 11
> >  
> > 
> 



From dfree.81 at gmail.com  Fri Mar 13 12:43:05 2009
From: dfree.81 at gmail.com (ed dd)
Date: Fri, 13 Mar 2009 12:43:05 +0100
Subject: read nifti file
Message-ID: <b3154a080903130443p51767873h44775ae3ae4c9203@mail.gmail.com>

Hi,
	
I hope to be in the right place, I wanted to ask how I can read *.nii
o .nii.gz file (so nifti) in scilab.

Thanks

Eddy


From rudy.magne at shom.fr  Tue Mar 17 09:49:43 2009
From: rudy.magne at shom.fr (Rudy Magne)
Date: Tue, 17 Mar 2009 08:49:43 +0000
Subject: mfprintf et vecteur
Message-ID: <49BF6427.7030503@shom.fr>

Bonjour,
Est il possible d'?crire dans un fichier texte un vecteur de taille nx 
avec par exemple 3 composantes par lignes (comme il est possible avec 
matlab ou fortran  par exemple)?

ex: fprintf(fid,"%10f.4 %10f.4 %10f.4 \n", vecteur(1:nx))

J'ai essay? avec mfprintf sans succ?s, ni d'ailleurs avec mtlb_fprintf. 
Il existe fprintfMat qui ?crit directement une matrice dans un fichier, 
mais il n'est alors pas possible directement d'?crire  avant et apr?s 
cette matrice.

Merci,
Rudy

 

-------------------------
Rudy Magne
SHOM/HOM/REC        
Vagues et domaine littoral  
13 rue du Chatellier, 29609 BREST Cedex
Tel: (33) 2 98 149 936 
Fax: (33) 2 98 221 864 



From rudy.magne at shom.fr  Tue Mar 17 10:47:57 2009
From: rudy.magne at shom.fr (Rudy Magne)
Date: Tue, 17 Mar 2009 09:47:57 +0000
Subject: mfprintf and vector
Message-ID: <49BF71CD.1040006@shom.fr>

Dear All,

Is is possible to write in a text file a (1,nx) vector with 3 components 
per lines (as it is available in matlab or fortran) with a defined 
format (see below)?

ex: fprintf(fid,"%10f.4 %10f.4 %10f.4 \n", vector(1:nx))

I have tried with mfprintf or mtlb_fprintf without succes.  fprintfMat 
would have done the job but it is then not possible to write before and 
after the written matrix.

Regards,
Rudy

-- 
-------------------------
Rudy Magne
SHOM/HOM/REC        
Vagues et domaine littoral  
13 rue du Chatellier, 29609 BREST Cedex
Tel: (33) 2 98 149 936 
Fax: (33) 2 98 221 864 



From evan at dembskey.org  Tue Mar 17 11:10:46 2009
From: evan at dembskey.org (Evan)
Date: Tue, 17 Mar 2009 12:10:46 +0200
Subject: Generating Wav files in Stereo
Message-ID: <001501c9a6e8$a7819280$f684b780$@org>

Hi all,

Is it possible to generate .wav files, with a different sin wav in each
channel? I want to demonstrate out-of-phase sound. I use the following to
generate a basic wave in at 440 Hz: 

fsampling = 44100;                   	// sampling frequency
t = 0 : 1/fsampling : 4;             	// 4 seconds at this frequency
x = 2 * %pi * 440 * t;
sinwav = sin (x);    				// a wave at 440 Hz - A

savewave ('beep_L.wav',sinwav,fsampling);  

I have found this on the net: sinwav_add = [sinwav ; sinwav_oof]; - however,
no sound is generated using this method.

Regards,
Evan



From hsr at ee.iitm.ac.in  Wed Mar 18 07:15:06 2009
From: hsr at ee.iitm.ac.in (harishankar ramachandran)
Date: Wed, 18 Mar 2009 11:45:06 +0530
Subject: [scilab-Users] mfprintf and vector
In-Reply-To: <49BF71CD.1040006@shom.fr>
References: <49BF71CD.1040006@shom.fr>
Message-ID: <200903181145.06954.hsr@ee.iitm.ac.in>

On Tuesday 17 March 2009 15:17, Rudy Magne wrote:
> Dear All,
>
> Is is possible to write in a text file a (1,nx) vector with 3 components
> per lines (as it is available in matlab or fortran) with a defined
> format (see below)?
>
> ex: fprintf(fid,"%10f.4 %10f.4 %10f.4 \n", vector(1:nx))
>
> I have tried with mfprintf or mtlb_fprintf without succes.  fprintfMat
> would have done the job but it is then not possible to write before and
> after the written matrix.
>
> Regards,
> Rudy

Here is a way of doing it with the fortran i/o commands. It is not as elegant 
as fprintfMat and fscanfMat, since it does not handle arbitrary number of 
preceding header lines. But it does what you want. Ofcourse, I don't know 
whether fortran i/o is supported any longer in scilab v5.x (I am using v4.1)

// Using fortran i/o

// Open the file for operations. Delete existing file if required
f=file('open',"junk.dat","unknown")
// write out a header
header=["First line of header";"Second line of header"];
write(f,header,'(a)')
// write out the matrix
A=[1 2 3;4 5 6;7 8 9];
write(f,A,'(3(1x,1p,g15.8))');
// close the file
file('close',f);

// open the file for reading
f=file('open',"junk.dat","old")

// read in the header (need to know number of lines)
header1=read(f,2,1,"(a)");

// read in the matrix 
B=read(f,-1,3,"(3g16.0)");

// close the file
file('close',f);

// print out header and matrix
disp(header1)
disp(B);

Regards

hari ramachandran



-- 
Dr. Hari Ramachandran, Professor, 332B ESB, EE Dept, IIT-Madras
Interests: Nonlinear Optics, Nonlinear Waves, Plasma Physics, Particle 
           Simulations, Computational Algorithms, Linux.
Off: 91-44-2257-4421                    Fax: 91-44-2257-0120
Res: 91-44-2663-1863             Home Email: omkarbharathi at gmail.com


From rudy.magne at shom.fr  Wed Mar 18 11:39:22 2009
From: rudy.magne at shom.fr (Rudy Magne)
Date: Wed, 18 Mar 2009 10:39:22 +0000
Subject: [scilab-Users] mfprintf and vector
In-Reply-To: <200903181145.06954.hsr@ee.iitm.ac.in>
References: <49BF71CD.1040006@shom.fr> <200903181145.06954.hsr@ee.iitm.ac.in>
Message-ID: <49C0CF5A.4050906@shom.fr>

harishankar ramachandran a ?crit :
> On Tuesday 17 March 2009 15:17, Rudy Magne wrote:
>   
>> Dear All,
>>
>> Is is possible to write in a text file a (1,nx) vector with 3 components
>> per lines (as it is available in matlab or fortran) with a defined
>> format (see below)?
>>
>> ex: fprintf(fid,"%10f.4 %10f.4 %10f.4 \n", vector(1:nx))
>>
>> I have tried with mfprintf or mtlb_fprintf without succes.  fprintfMat
>> would have done the job but it is then not possible to write before and
>> after the written matrix.
>>
>> Regards,
>> Rudy
>>     
>
> Here is a way of doing it with the fortran i/o commands. It is not as elegant 
> as fprintfMat and fscanfMat, since it does not handle arbitrary number of 
> preceding header lines. But it does what you want. Ofcourse, I don't know 
> whether fortran i/o is supported any longer in scilab v5.x (I am using v4.1)
>
> // Using fortran i/o
>
> // Open the file for operations. Delete existing file if required
> f=file('open',"junk.dat","unknown")
> // write out a header
> header=["First line of header";"Second line of header"];
> write(f,header,'(a)')
> // write out the matrix
> A=[1 2 3;4 5 6;7 8 9];
> write(f,A,'(3(1x,1p,g15.8))');
> // close the file
> file('close',f);
>
> // open the file for reading
> f=file('open',"junk.dat","old")
>
> // read in the header (need to know number of lines)
> header1=read(f,2,1,"(a)");
>
> // read in the matrix 
> B=read(f,-1,3,"(3g16.0)");
>
> // close the file
> file('close',f);
>
> // print out header and matrix
> disp(header1)
> disp(B);
>
> Regards
>
> hari ramachandran
>
>
>
>   
Thanks Hari,
It works fine.
Rudy


-- 
-------------------------
Rudy Magne
SHOM/HOM/REC        
Vagues et domaine littoral  
13 rue du Chatellier, 29609 BREST Cedex
Tel: (33) 2 98 149 936 
Fax: (33) 2 98 221 864 



From vaborg at gmail.com  Wed Mar 18 16:59:26 2009
From: vaborg at gmail.com (Ivan Maximov)
Date: Wed, 18 Mar 2009 16:59:26 +0100
Subject: code which does not work
Message-ID: <20090318165926.68e90b17.vaborg@gmail.com>

Hi!

I have an example of code with very strange behavior.
I clear what I mean. The code does not work if to run it it also does not work
if to run in debuger. But what is very strange it perfectly works if I add one
breakpoint to the source. 
The error message is invalid index. The same code is works good in Matlab.
It has no sense to write it here I can send you if you are interesting.

The question is following is it my stupid problem (it seems not looking at the
Matlab) or it is a bug in Scilab?

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



From fvogelnew1 at free.fr  Wed Mar 18 17:31:20 2009
From: fvogelnew1 at free.fr (fvogelnew1 at free.fr)
Date: Wed, 18 Mar 2009 17:31:20 +0100
Subject: [scilab-Users] code which does not work
In-Reply-To: <20090318165926.68e90b17.vaborg@gmail.com>
References: <20090318165926.68e90b17.vaborg@gmail.com>
Message-ID: <1237393880.49c121d8308ea@imp.free.fr>

Selon Ivan Maximov <vaborg at gmail.com>:
> It has no sense to write it here I can send you if you are interesting.

Sure. I don't think any useful answer to your problem can be proposed without
seeing your code.

Please provide it along with detailed explanations about:
  - how you run it
  - in what Scilab version
  - on what platform
  - where you set the breakpoint and how you do it (from the debugger or
directly using setbpt?)
  - and any other detail you think could be relevant.

Thanks,
Francois




From fvogelnew1 at free.fr  Wed Mar 18 17:59:57 2009
From: fvogelnew1 at free.fr (fvogelnew1 at free.fr)
Date: Wed, 18 Mar 2009 17:59:57 +0100
Subject: [scilab-Users] code which does not work
In-Reply-To: <20090318175247.3316ab55.vaborg@gmail.com>
References: <20090318165926.68e90b17.vaborg@gmail.com> <1237393880.49c121d8308ea@imp.free.fr> <20090318175247.3316ab55.vaborg@gmail.com>
Message-ID: <1237395597.49c1288d7863d@imp.free.fr>

OK, thanks. I'll have a deeper look but first thing staring me in the face is
the pause instruction.

Please try to remove it and tell me the result.

Francois



From vaborg at gmail.com  Wed Mar 18 17:52:47 2009
From: vaborg at gmail.com (Ivan Maximov)
Date: Wed, 18 Mar 2009 17:52:47 +0100
Subject: [scilab-Users] code which does not work
In-Reply-To: <1237393880.49c121d8308ea@imp.free.fr>
References: <20090318165926.68e90b17.vaborg@gmail.com>
	<1237393880.49c121d8308ea@imp.free.fr>
Message-ID: <20090318175247.3316ab55.vaborg@gmail.com>

Hi Francois,

>   - how you run it
by ctrl+l in editor

>   - in what Scilab version
4.2.1 it gives error message
5.1 it kills Scilab at all

>   - on what platform
Fedore 10,  x86

>   - where you set the breakpoint and how you do it (from the debugger or
> directly using setbpt?)
from debugger
breakpoint in string 7 function krotov_csa.sci

below is code
__________________________________________________________________________________________________________
main script
__________________________________________________________________________________________________________
// INEPT and CSA

clearglobal;
funcprot(0);

global path dbg

path='/home/vaborg/scilab/csa/';
cd(path);
// Debug info
dbg=1;


exec(path+'system.sce');
exec(path+'H1.sci');
exec(path+'penalty.sci');
exec(path+'krotov_csa.sci');


disp('START GENERAL');
tic();

// ITERATION LOOP
w1=rf*rand(1,N);  w2=rf*rand(1,N);
w3=rf*rand(1,N);  w4=rf*rand(1,N);


[eff,r1,r2,r3,r4]=krotov_csa(w1,w2,w3,w4);

ti=toc();
disp('STOP GENERAL');
printf("Time is: %i\n",ti);
__________________________________________________________________________________________________________
system.sce
__________________________________________________________________________________________________________
global dt N dp csI csS csIN csSN l1 csN Hij

err=exec(path+'pauli.sce','errcatch');
if err~=0,
  disp('ERROR: Can not find pauli.sce!');
  abort();
end

ix=x.*.u; iy=y.*.u; iz=z.*.u;
sx=u.*.x; sy=u.*.y; sz=u.*.z;
uu=u.*.u;



dp=2*%pi;

init=sx;
tini=ix;

//------------------------------
// couplings time and shifts
//------------------------------
J=140;
T=1/J;
N=200;  NIT=20;
l1=1e-5;
delta=1e-2; eta=delta;
dt=T/N; rf=1e3;

// shift arrays
csI=6:0.1:9;
csS=115:2.5:125;

wi=-500e6*dp;
ws=50.684e6*dp;

csIN=length(csI);
csSN=length(csS);
csN=csIN*csSN;

// Hamiltonians
Hamilt=dp*J*iz*sz;

Hij(:,:,csN)=uu;
U(:,:,csN)=uu;
B(:,:,csN)=uu;

for i=1:csIN,
  for j=1:csSN,
    Hij(:,:,(i-1)*csSN+j)=(-csI(i)*1e-6)*wi*iz+(-csS(j)*1e-6)*ws*sz;
    U(:,:,(i-1)*csSN+j)=uu;
    B(:,:,(i-1)*csSN+j)=uu;
  end
end

__________________________________________________________________________________________________________
krotov_csa.sci
__________________________________________________________________________________________________________
function [ef,u1,u2,u3,u4]=krotov_csa(x1,x2,x3,x4)
 
  u1=x1; u2=x2; u3=x3; u4=x4;
  t1=x1; t2=x2; t3=x3; t4=x4;
  
  PROP=uu;
  TE=uu;
  pause;
  
  for i=1:N,
    for j=1:csN,
        PROP=expm(-%i*dt*(Hij(:,:,j)+Hamilt+H1(u1(i),u2(i),u3(i),u4(i))));
        TE=PROP*U(:,:,j);
        U(:,:,j)=TE;
    end
  end

  eff=0;  pen=0; fu=0;
  
  for i=1:csN,
    eff=eff+real(trace(tini*U(:,:,i)*init*U(:,:,i)'))/csN;
  end
  
  
  pen=penalty(u1,u2,u3,u4);
  fu=1+eff-pen;
  
  if dbg==1,
    printf("EFF: %f \tF: %f \tP: %f\n",eff,fu,pen);
  end

  for i=1:csN,
    B(:,:,i)=U(:,:,i)'+init*U(:,:,i)'*tini;
  end
  
  
  for i=N:(-1):1,
    for j=1:csN,
      PROP=expm(-%i*dt*(Hij(:,:,j)+Hamilt+H1(t1(i),t2(i),t3(i),t4(i))));
      B(:,:,j)=B(:,:,j)*PROP;
    end
  end

//-----------------------------------------------
// ITERATION
//-----------------------------------------------
  for iter=1:NIT,

      for i=1:csN,
        U(:,:,i)=uu;
      end
      
      PROP=zeros(4,4);
      for i=1:csN,
        PROP=PROP+B(:,:,i);
      end
      
      
      u1(1)=(1-delta)*t1(1)+delta/l1*imag(trace(ix*PROP));
      u2(1)=(1-delta)*t2(1)+delta/l1*imag(trace(iy*PROP));
      u3(1)=(1-delta)*t3(1)+delta/l1*imag(trace(sx*PROP));
      u4(1)=(1-delta)*t4(1)+delta/l1*imag(trace(sy*PROP));
      
      for i=2:N,
        for j=1:csN,
          PROP=expm(-%i*dt*(Hij(:,:,j)+Hamilt+H1(u1(i-1),u2(i-1),u3(i-1),u4
(i-1)))); U(:,:,j)=PROP*U(:,:,j);
          PROP=expm(%i*dt*(Hij(:,:,j)+Hamilt+H1(t1(i-1),t2(i-1),t3(i-1),t4
(i-1)))); B(:,:,j)=B(:,:,j)*PROP;
        end
          
        PROP=zeros(4,4);  
        for j=1:csN,
          PROP=PROP+U(:,:,j)*B(:,:,j);
        end
      
          u1(i)=(1-delta)*t1(i)+delta/l1*imag(trace(ix*PROP));
          u2(i)=(1-delta)*t2(i)+delta/l1*imag(trace(iy*PROP));
          u3(i)=(1-delta)*t3(i)+delta/l1*imag(trace(sx*PROP));
          u4(i)=(1-delta)*t4(i)+delta/l1*imag(trace(sy*PROP));
      end
  
      for j=1:csN,
          PROP=expm(-%i*dt*(Hij(:,:,j)+Hamilt+H1(u1(N),u2(N),u3(N),u4(N))));
          U(:,:,j)=PROP*U(:,:,j);
      end
      
      temp=0;
      for j=1:csN,
        temp=temp+real(trace(tini*U(:,:,j)*init*U(:,:,j)'))/csN;
      end
      
      eff=cat(1,eff,temp);
      pen=cat(1,pen,penalty(u1,u2,u3,u4));
      fu=cat(1,fu,1+eff(iter+1)-pen(iter+1));
  
      if dbg==1,
        printf("EFF: %f \tF: %f \tP: %f\n",eff(iter+1),fu(iter+1),pen(iter+1));
      end
  
      if (fu(iter+1)-fu(iter))<0,
        disp('ERROR: DISCONVERGENCY');
        break;
      elseif (fu(iter+1)-fu(iter))>0 & (fu(iter+1)-fu(iter))<1e-4,
        disp('CONVERGENT!');
        break;
      end
 
      for j=1:csN,
        B(:,:,j)=U(:,:,j)'+init*U(:,:,j)'*tini;
      end
      
      PROP=zeros(4,4);
      for j=1:csN,
        PROP=PROP+U(:,:,j)*B(:,:,j);
      end
            
      t1(N)=(1-eta)*u1(N)+eta/l1*imag(trace(ix*PROP));
      t2(N)=(1-eta)*u2(N)+eta/l1*imag(trace(iy*PROP));
      t3(N)=(1-eta)*u3(N)+eta/l1*imag(trace(sx*PROP));
      t4(N)=(1-eta)*u4(N)+eta/l1*imag(trace(sy*PROP));

      for i=(N-1):(-1):1,
        for j=1:csN,
          PROP=expm(%i*dt*(Hij(:,:,j)+Hamilt+H1(u1(i+1),u2(i+1),u3(i+1),u4(i
+1)))); U(:,:,j)=PROP*U(:,:,j);
          PROP=expm(-%i*dt*(Hij(:,:,j)+Hamilt+H1(t1(i+1),t2(i+1),t3(i+1),t4(i
+1)))); B(:,:,j)=B(:,:,j)*PROP;
        end

        PROP=zeros(4,4);
        for j=1:csN,
          PROP=PROP+U(:,:,j)*B(:,:,j);
        end
        
          t1(i)=(1-eta)*u1(i)+eta/l1*imag(trace(ix*PROP));
          t2(i)=(1-eta)*u2(i)+eta/l1*imag(trace(iy*PROP));
          t3(i)=(1-eta)*u3(i)+eta/l1*imag(trace(sx*PROP));
          t4(i)=(1-eta)*u4(i)+eta/l1*imag(trace(sy*PROP));
      end
  
    for j=1:csN,
      PROP=expm(%i*dt*(Hij(:,:,j)+Hamilt+H1(u1(1),u2(1),u3(1),u4(1))));
      U(:,:,j)=PROP*U(:,:,j);
      PROP=expm(-%i*dt*(Hij(:,:,j)+Hamilt+H1(t1(1),t2(1),t3(1),t4(1))));
      B(:,:,j)=B(:,:,j)*PROP;
    end
  
//    disp(U);
  end


  ef=eff(length(eff));
endfunction
__________________________________________________________________________________________________________
H1.sci
__________________________________________________________________________________________________________
function [res]=H1(x1,x2,x3,x4)
  res=x1*ix+x2*iy+x3*sx+x4*sy;
endfunction
__________________________________________________________________________________________________________
penalty.sci
__________________________________________________________________________________________________________
function [res]=penalty(x1,x2,x3,x4)
  res=dt*l1*sum(x1.^2+x2.^2+x3.^2+x4.^2);
endfunction
__________________________________________________________________________________________________________
pauli.sce
__________________________________________________________________________________________________________
// Pauli matrix
x=[0,0.5;0.5,0]; 
y=[0,-0.5*%i;0.5*%i,0]; 
z=[0.5,0;0,-0.5]; 
u=eye(2,2);




From vaborg at gmail.com  Wed Mar 18 18:03:26 2009
From: vaborg at gmail.com (Ivan Maximov)
Date: Wed, 18 Mar 2009 18:03:26 +0100
Subject: [scilab-Users] code which does not work
In-Reply-To: <1237395597.49c1288d7863d@imp.free.fr>
References: <20090318165926.68e90b17.vaborg@gmail.com>
	<1237393880.49c121d8308ea@imp.free.fr>
	<20090318175247.3316ab55.vaborg@gmail.com>
	<1237395597.49c1288d7863d@imp.free.fr>
Message-ID: <20090318180326.dd012515.vaborg@gmail.com>

pause - is left after debugging.
In next loops trouble begins.

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



From sdr at durietz.se  Wed Mar 18 18:27:42 2009
From: sdr at durietz.se (Stefan Du Rietz)
Date: Wed, 18 Mar 2009 18:27:42 +0100
Subject: [scilab-Users] mfprintf and vector
In-Reply-To: <49C0CF5A.4050906@shom.fr>
References: <49BF71CD.1040006@shom.fr> <200903181145.06954.hsr@ee.iitm.ac.in> <49C0CF5A.4050906@shom.fr>
Message-ID: <49C12F0E.6010900@durietz.se>

On 2009-03-18 11:39, Rudy Magne wrote:
--------------------
> harishankar ramachandran a ?crit :
>> On Tuesday 17 March 2009 15:17, Rudy Magne wrote:
>>  
>>> Dear All,
>>>
>>> Is is possible to write in a text file a (1,nx) vector with 3 components
>>> per lines (as it is available in matlab or fortran) with a defined
>>> format (see below)?
>>>
>>> ex: fprintf(fid,"%10f.4 %10f.4 %10f.4 \n", vector(1:nx))
>>>
>>> I have tried with mfprintf or mtlb_fprintf without succes.  fprintfMat
>>> would have done the job but it is then not possible to write before and
>>> after the written matrix.
>>>
>>> Regards,
>>> Rudy
>>>     
>>
>> Here is a way of doing it with the fortran i/o commands. It is not as 
>> elegant as fprintfMat and fscanfMat, since it does not handle 
>> arbitrary number of preceding header lines. But it does what you want. 
>> Ofcourse, I don't know whether fortran i/o is supported any longer in 
>> scilab v5.x (I am using v4.1)
>>
>> // Using fortran i/o
>>
>> // Open the file for operations. Delete existing file if required
>> f=file('open',"junk.dat","unknown")
>> // write out a header
>> header=["First line of header";"Second line of header"];
>> write(f,header,'(a)')
>> // write out the matrix
>> A=[1 2 3;4 5 6;7 8 9];
>> write(f,A,'(3(1x,1p,g15.8))');
>> // close the file
>> file('close',f);
>>
>> // open the file for reading
>> f=file('open',"junk.dat","old")
>>
>> // read in the header (need to know number of lines)
>> header1=read(f,2,1,"(a)");
>>
>> // read in the matrix B=read(f,-1,3,"(3g16.0)");
>>
>> // close the file
>> file('close',f);
>>
>> // print out header and matrix
>> disp(header1)
>> disp(B);
>>
>> Regards
>>
>> hari ramachandran
>>
>>
>>
>>   
> Thanks Hari,
> It works fine.
> Rudy
> 
> 
I think you had a wrong format string in mfprintf - did you intend to 
have 10 positions and 4 decimals for each number?

In that case you should write:
"%10.4f %10.4f %10.4f \n"

Otherwise you will get 6 decimals (at least I get that) followed by 
".4". The parts of the format string that are not between "%" and "f" 
(where "f" is the conversion type) will be written exactly as in the 
format string. See help printf_conversion.

Did you first open the file to get fid?
fid = mopen(file, mode);
where file is a string with the pathname of your file and mode is a 
string which controls the access allowed to the file. See help mopen.

Regards
Stefan


From fvogelnew1 at free.fr  Wed Mar 18 23:26:11 2009
From: fvogelnew1 at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Vogel?=)
Date: Wed, 18 Mar 2009 23:26:11 +0100
Subject: [scilab-Users] code which does not work
In-Reply-To: <20090318175247.3316ab55.vaborg@gmail.com>
References: <20090318165926.68e90b17.vaborg@gmail.com>	<1237393880.49c121d8308ea@imp.free.fr> <20090318175247.3316ab55.vaborg@gmail.com>
Message-ID: <49C17503.3050501@free.fr>

Ivan Maximov said on 18/03/2009 17:52:
> krotov_csa.sci
> __________________________________________________________________________________________________________
> function [ef,u1,u2,u3,u4]=krotov_csa(x1,x2,x3,x4)
>  
>   u1=x1; u2=x2; u3=x3; u4=x4;
>   t1=x1; t2=x2; t3=x3; t4=x4;
>   
>   PROP=uu;
>   TE=uu;
>   pause;
>   
>   for i=1:N,
>     for j=1:csN,
>         PROP=expm(-%i*dt*(Hij(:,:,j)+Hamilt+H1(u1(i),u2(i),u3(i),u4(i))));
>         TE=PROP*U(:,:,j);
>         U(:,:,j)=TE;
>     end
>   end

This looks very much like a bug of Scilab when inserting a matrix in 
hypermatrices.

The problem is that after the first execution (i=1 and j=1) of:
>         PROP=expm(-%i*dt*(Hij(:,:,j)+Hamilt+H1(u1(i),u2(i),u3(i),u4(i))));
>         TE=PROP*U(:,:,j);
>         U(:,:,j)=TE;

something goes wrong in U: size(U) becomes 4x4, i.e. U becomes a 4x4 
matrix. It looses its third dimension.

The strange thing is that this does not happen whatever the context, 
and this explains why it works sometimes with the debugger (but I saw 
the problem also when debugging in Scipad). My guess is that global 
variables are involved, and perhaps execution contexts of your several 
exec instructions.

The problem can be easier seen when instrumenting the code a bit like 
this in function krotov_csa lines 10 and below:

for i=1:N,
  mprintf("i is: %d\n",i)
  for j=1:csN,
   mprintf("j is: %d\n",j)
   PROP=expm(-%i*dt*(Hij(:,:,j)+Hamilt+H1(u1(i),u2(i),u3(i),u4(i))));
   TE=PROP*U(:,:,j);
   disp("BEFORE");disp(size(U));
   U(:,:,j)=TE;
   disp("AFTER");disp(size(U))
  end
end


Now upon execution from Scilab shell:

-->exec main.sce;

  START GENERAL
i is: 1
j is: 1

  BEFORE

     4.    4.    155.

  AFTER

     4.    4.
j is: 2
  !--error 21
invalid index


My opinion is that Scilab finds out that U(:,:,j) is a matrix and no 
longer an hypermatrix (since the third dimension is exactly 1), and in 
this case it reduces the number of dimensions of U(:,:,j). The problem 
is that then Scilab erroneously affects U(:,:,j) into U, reducing the 
dimension of U in the process too.

I think that this deserves a bug report in the bugzilla:
http://bugzilla.scilab.org/

However it would help if you could trim down the problem to a simpler 
script. Thanks in advance.

Francois


From fvogelnew1 at free.fr  Wed Mar 18 23:38:53 2009
From: fvogelnew1 at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Vogel?=)
Date: Wed, 18 Mar 2009 23:38:53 +0100
Subject: [scilab-Users] code which does not work
In-Reply-To: <49C17503.3050501@free.fr>
References: <20090318165926.68e90b17.vaborg@gmail.com>	<1237393880.49c121d8308ea@imp.free.fr> <20090318175247.3316ab55.vaborg@gmail.com> <49C17503.3050501@free.fr>
Message-ID: <49C177FD.8090606@free.fr>

Fran?ois Vogel said on 18/03/2009 23:26:

By the way, an awfully ugly, slow, crappy workaround is to replace:

>         U(:,:,j)=TE;

by:

         UU=matrix(U,[1,1,-1]);
         for kkk=1:4
           UU(16*(j-1)+kkk)=TE(kkk);
         end
         U=hypermat([4 4 csN],UU);

(if I didn't mess it, please check - I have only a vague idea of what 
your script is doing, so can't say if result is good or not).

Francois



From vaborg at gmail.com  Wed Mar 18 23:43:16 2009
From: vaborg at gmail.com (Ivan Maximov)
Date: Wed, 18 Mar 2009 23:43:16 +0100
Subject: [scilab-Users] code which does not work
In-Reply-To: <49C17503.3050501@free.fr>
References: <20090318165926.68e90b17.vaborg@gmail.com>
	<1237393880.49c121d8308ea@imp.free.fr>
	<20090318175247.3316ab55.vaborg@gmail.com>
	<49C17503.3050501@free.fr>
Message-ID: <20090318234316.e0168240.vaborg@gmail.com>

Dear Fran?ois
I guess your suggestion concerning the global variables is true.
I will try to localize and simplify this problem and will do a bug report.

But what is surprising if I set the breakpoint before these loops
typically U variable keeps its dimension.

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



From fvogelnew1 at free.fr  Wed Mar 18 23:54:48 2009
From: fvogelnew1 at free.fr (=?UTF-8?B?RnJhbsOnb2lzIFZvZ2Vs?=)
Date: Wed, 18 Mar 2009 23:54:48 +0100
Subject: [scilab-Users] code which does not work
In-Reply-To: <20090318234316.e0168240.vaborg@gmail.com>
References: <20090318165926.68e90b17.vaborg@gmail.com>	<1237393880.49c121d8308ea@imp.free.fr>	<20090318175247.3316ab55.vaborg@gmail.com>	<49C17503.3050501@free.fr> <20090318234316.e0168240.vaborg@gmail.com>
Message-ID: <49C17BB8.2070500@free.fr>

Ivan Maximov said on 18/03/2009 23:43:
> But what is surprising if I set the breakpoint before these loops
> typically U variable keeps its dimension.

This I can't reproduce. How do you do that?

On my side, here is how I debug your code:

  - I open all your six files in Scipad launched from Scilab 4.1.2
  - I comment the pause instruction in function krotov_csa (because 
the debugger would not happily live with it)
  - I place (F9) a breakpoint on line 7 (the one that reads   TE=uu; 
in function krotov_csa)
  - I display the main .sce file in Scipad and put Scipad's focus in it
  - I hit F10 to run the debug (click OK in the dialog that pops up)
  - I hit F11 to run until the breakpoint gets hit
  - From there I hit F8 a few times to execute the code step by step, 
and the 9th time I get the invalid index error, which is exactly what 
is expected.

So, how are you doing this without triggering the error?

Francois



From fvogelnew1 at free.fr  Thu Mar 19 00:34:04 2009
From: fvogelnew1 at free.fr (=?UTF-8?B?RnJhbsOnb2lzIFZvZ2Vs?=)
Date: Thu, 19 Mar 2009 00:34:04 +0100
Subject: [scilab-Users] code which does not work
In-Reply-To: <20090318234316.e0168240.vaborg@gmail.com>
References: <20090318165926.68e90b17.vaborg@gmail.com>	<1237393880.49c121d8308ea@imp.free.fr>	<20090318175247.3316ab55.vaborg@gmail.com>	<49C17503.3050501@free.fr> <20090318234316.e0168240.vaborg@gmail.com>
Message-ID: <49C184EC.1010006@free.fr>

Ivan Maximov said on 18/03/2009 23:43:
> I guess your suggestion concerning the global variables is true.
> I will try to localize and simplify this problem and will do a bug report.

Hmm, second thoughts, and hopefully better advices.

Thinking about global variables and scoping, what you see becomes 
obvious and the fix as well.

If you add:

  global U

at the beginning of function krotov_csa, your code will work.

You should be able to understand why by reading this:
http://wiki.scilab.org/howto/global_and_local_variables

Francois


From sylvestre.ledru at scilab.org  Thu Mar 19 10:53:35 2009
From: sylvestre.ledru at scilab.org (Sylvestre Ledru)
Date: Thu, 19 Mar 2009 10:53:35 +0100
Subject: Scilab & Google summer of code 2009
Message-ID: <1237456415.21269.5650.camel@korcula.inria.fr>

Hello,

I am glad to inform you that Scilab has been accepted as organisation by
Google in the context of the Summer of Code 2009 [1] [2].

A list of ideas which could be done in the context of the Gsoc 2009 is
available here:
http://wiki.scilab.org/Ideas_of_development_for_Scilab
you are welcome if you want to add your ideas.

Btw, we already have some students who want to apply!

Don't hesitate if you have any questions,
Sylvestre

[1] http://code.google.com/soc/
[2] http://en.wikipedia.org/wiki/Google_Summer_of_Code



From mathieu.dubois at limsi.fr  Thu Mar 19 13:12:34 2009
From: mathieu.dubois at limsi.fr (Mathieu Dubois)
Date: Thu, 19 Mar 2009 13:12:34 +0100 (CET)
Subject: [scilab-Users] Generating Wav files in Stereo
In-Reply-To: <001501c9a6e8$a7819280$f684b780$@org>
References: <001501c9a6e8$a7819280$f684b780$@org>
Message-ID: <50916.82.123.12.81.1237464754.squirrel@keo.limsi.fr>

Hi Evan,

To produce a stereo file you have to pass a 2*N matrix to savewave.

Try the attached script which produce a stereo file (one sinus at 440Hz
and one at 460Hz) and then plot the power spectrum (showing 2 peaks). On
my computer I can read the file.

Another possibility is that you have a sound problem under scilab. Can you
read a wav file with playsnd?

Hope that helps,
Mathieu

> Hi all,
>
> Is it possible to generate .wav files, with a different sin wav in each
> channel? I want to demonstrate out-of-phase sound. I use the following to
> generate a basic wave in at 440 Hz:
>
> fsampling = 44100;                   	// sampling frequency
> t = 0 : 1/fsampling : 4;             	// 4 seconds at this frequency
> x = 2 * %pi * 440 * t;
> sinwav = sin (x);    				// a wave at 440 Hz - A
>
> savewave ('beep_L.wav',sinwav,fsampling);
>
> I have found this on the net: sinwav_add = [sinwav ; sinwav_oof]; -
> however,
> no sound is generated using this method.
>
> Regards,
> Evan
>
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: test_wave.sce
Type: application/octet-stream
Size: 592 bytes
Desc: not available
URL: <https://lists.scilab.org/pipermail/users/attachments/20090319/3c1d3843/attachment.obj>

From spintronics at qq.com  Thu Mar 19 14:35:03 2009
From: spintronics at qq.com (LJ)
Date: Thu, 19 Mar 2009 21:35:03 +0800
Subject: How to change  font in  the help browser
Message-ID: <49C24A07.2060700@qq.com>

Hello all,
I have just compiled scilab 5.1 successfully under my linux, but the
words in help browser looks too small. I tried to sellect larger size in
preference menu, but it only affects the console, is there any way to
specify the font of the help browser?

Thanks very much

LJ


From elisabeth.fournier at renault.com  Fri Mar 20 17:32:48 2009
From: elisabeth.fournier at renault.com (FOURNIER Elisabeth)
Date: Fri, 20 Mar 2009 17:32:48 +0100
Subject: Problem with Sgrayplot/plot
Message-ID: <016601c9a979$86619990$1b74dc0a@corp.noxiane.net>

Hello,


I have a problem with Sgrayplot and plot.

1. With Sgrayplot, I obtain a map for the data (carto.dat)
2. I want to plot just one point with a color. (Point [80,80]) with the plot 
or plot2d function.

Resultat : The point appears in color, but also a isoline ([x~10-20, 
y~10-20]) appears with the same color.

Why? What to do?
Thanks for your help

Elisabeth Fournier




-- Disclaimer ------------------------------------
Ce message ainsi que les eventuelles pieces jointes constituent une correspondance privee et confidentielle a l'attention exclusive du destinataire designe ci-dessus. Si vous n'etes pas le destinataire du present message ou une personne susceptible de pouvoir le lui delivrer, il vous est signifie que toute divulgation, distribution ou copie de cette transmission est strictement interdite. Si vous avez recu ce message par erreur, nous vous remercions d'en informer l'expediteur par telephone ou de lui retourner le present message, puis d'effacer immediatement ce message de votre systeme.
***
This e-mail and any attachments is a confidential correspondence intended only for use of the individual or entity named above. If you are not the intended recipient or the agent responsible for delivering the message to the intended recipient, you are hereby notified that any disclosure, distribution or copying of this communication is strictly prohibited. If you have received this communication in error, please notify the sender by phone or by replying this message, and then delete this message from your system.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Resultat.png
Type: image/png
Size: 17717 bytes
Desc: not available
URL: <https://lists.scilab.org/pipermail/users/attachments/20090320/d9c6d443/attachment.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: carto.zip
Type: application/x-zip-compressed
Size: 4733435 bytes
Desc: not available
URL: <https://lists.scilab.org/pipermail/users/attachments/20090320/d9c6d443/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Test_plot.sce
Type: application/octet-stream
Size: 264 bytes
Desc: not available
URL: <https://lists.scilab.org/pipermail/users/attachments/20090320/d9c6d443/attachment.obj>

From fvogelnew1 at free.fr  Sat Mar 21 09:16:17 2009
From: fvogelnew1 at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Vogel?=)
Date: Sat, 21 Mar 2009 09:16:17 +0100
Subject: [scilab-Users] Problem with Sgrayplot/plot
In-Reply-To: <016601c9a979$86619990$1b74dc0a@corp.noxiane.net>
References: <016601c9a979$86619990$1b74dc0a@corp.noxiane.net>
Message-ID: <49C4A251.6060404@free.fr>

FOURNIER Elisabeth said on 20/03/2009 17:32:
> Resultat : The point appears in color, but also a isoline ([x~10-20, 
> y~10-20]) appears with the same color.

plot(80,80,'*g')   seems to have an influence on the colormap (it adds 
the selected color, here green, at the end of the colormap). The 
isoline you're seeing is just the line having the 'max' color before 
the plot() instruction

You can use plot2d(80,80,style=0)  that does not have this behavior 
(make experiments with style=.. to fit your needs).

Francois


> -- Disclaimer ------------------------------------
> Ce message ainsi que les eventuelles pieces jointes constituent une 
> correspondance privee et confidentielle a l'attention exclusive du 
> destinataire designe ci-dessus. Si vous n'etes pas le destinataire du 
> present message ou une personne susceptible de pouvoir le lui delivrer, 
> il vous est signifie que toute divulgation, distribution ou copie de 
> cette transmission est strictement interdite. Si vous avez recu ce 
> message par erreur, nous vous remercions d'en informer l'expediteur par 
> telephone ou de lui retourner le present message, puis d'effacer 
> immediatement ce message de votre systeme.

------
D?charge de responsabilit?
Nous ne savons pas utiliser les fonctions de notre serveur de courriel 
relatives au cryptage et ? la s?curit?. Nous sommes tellement 
perplexes que nous ne pouvons pas m?me ?tre certains que ce courriel a 
?t? envoy? ? la bonne personne. Ainsi, si vous recevez cet email ? 
cause de notre incomp?tence, veuillez le supprimer et en informer 
l'exp?diteur d?s que possible. Veuillez remarquer que ceci est r?dig? 
dans un style juridique aussi effrayant que possible de mani?re ? ce 
que personne ne remarque combien nous sommes incomp?tents.


> ***
> This e-mail and any attachments is a confidential correspondence 
> intended only for use of the individual or entity named above. If you 
> are not the intended recipient or the agent responsible for delivering 
> the message to the intended recipient, you are hereby notified that any 
> disclosure, distribution or copying of this communication is strictly 
> prohibited. If you have received this communication in error, please 
> notify the sender by phone or by replying this message, and then delete 
> this message from your system.

------
Disclaimer
We don't know how to use the security and encryption features of our 
e-mail server. We are so confused that we cannot even be certain that 
this e-mail was delivered to the proper person. So, if you received 
this e-mail because of our incompetence, please delete it and notify 
us as soon as possible. Notice that this is being written in the most 
scary legalistic way, so no one notices how incompetent we are.


From rcotza at mac.com  Sat Mar 21 10:26:59 2009
From: rcotza at mac.com (Raffaele Cotza)
Date: Sat, 21 Mar 2009 10:26:59 +0100
Subject: MESSAGE
Message-ID: <E1D29DF6-873C-40B7-B894-7C865874045F@mac.com>

Scilab TCL/TK module not installed.
at line      16 of function restore called by :
line   543 of function scicos called by :
line    16 of function scicos_demostration called by :
"/Simple_Demo.cosf")
line     9 of exec file called by :
ec(script_path,-1);;if exists("%oldgcbo")
while executing a callback

I HAVE RECEIVED THIS MESSAGE HAW I CAN INSTALL TCL/TK MODULE


From sylvestre.ledru at scilab.org  Sat Mar 21 14:42:36 2009
From: sylvestre.ledru at scilab.org (Sylvestre Ledru)
Date: Sat, 21 Mar 2009 14:42:36 +0100
Subject: Tcl not working under Mac OS X [was: MESSAGE]
In-Reply-To: <E1D29DF6-873C-40B7-B894-7C865874045F@mac.com>
References: <E1D29DF6-873C-40B7-B894-7C865874045F@mac.com>
Message-ID: <1237642956.4656.78.camel@zlarin>

Le samedi 21 mars 2009 ? 10:26 +0100, Raffaele Cotza a ?crit :
> Scilab TCL/TK module not installed.
> at line      16 of function restore called by :
> line   543 of function scicos called by :
> line    16 of function scicos_demostration called by :
> "/Simple_Demo.cosf")
> line     9 of exec file called by :
> ec(script_path,-1);;if exists("%oldgcbo")
> while executing a callback
> 
> I HAVE RECEIVED THIS MESSAGE HAW I CAN INSTALL TCL/TK MODULE

Hello Raffaele,

Looking at your email address, I guess you are running Mac OS X. Tcl/Tk
is not available.
See:
http://www.scilab.org/download/index_download.php?page=RELEASE_NOTES_5.1

Sylvestre




From whalesuit at gmail.com  Mon Mar 23 23:14:59 2009
From: whalesuit at gmail.com (Eric E)
Date: Mon, 23 Mar 2009 18:14:59 -0400
Subject: Problem with function calls in 5.1 alpha for MacOS Leopard
Message-ID: <7426fe270903231514u3fe7530an540ecdec0651e98d@mail.gmail.com>

Howdy Sylvestre,
  I'm having some trouble making and stepping through function calls
in Scilab 5.1 alpha for MacOS.  What's the best place to report and
discuss these problems?

Many thanks,

Eric


From sylvestre.ledru at scilab.org  Mon Mar 23 23:18:42 2009
From: sylvestre.ledru at scilab.org (Sylvestre Ledru)
Date: Mon, 23 Mar 2009 23:18:42 +0100
Subject: [scilab-Users] Problem with function calls in 5.1 alpha for
	MacOS Leopard
In-Reply-To: <7426fe270903231514u3fe7530an540ecdec0651e98d@mail.gmail.com>
References: <7426fe270903231514u3fe7530an540ecdec0651e98d@mail.gmail.com>
Message-ID: <1237846722.26960.156.camel@zlarin>

Salut Eric,

Le lundi 23 mars 2009 ? 18:14 -0400, Eric E a ?crit :
> Howdy Sylvestre,
>   I'm having some trouble making and stepping through function calls
> in Scilab 5.1 alpha for MacOS.  
That is not surprising. It is just an alpha.

> What's the best place to report and discuss these problems?
Well, if you have been able to reproduce a bug, please submit it on 
http://bugzilla.scilab.org/ with the piece of code to reproduce it. I
will have a look.

Sylvestre




From hron at gmx.de  Wed Mar 25 19:59:30 2009
From: hron at gmx.de (ron)
Date: Wed, 25 Mar 2009 19:59:30 +0100
Subject: in16 matrix
Message-ID: <1238007570.6037.8.camel@stargate>

Hello,
I am wondering if it is possible to create an int16 type matrix. This
would save stack memory in my application since the matrix is huge.
How can I do this?
thank you.
ron

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Filtered-With-Copfilter: Version 0.84beta4 (ProxSMTP 1.8)
Copfilter-Virus-Scanned: ClamAV 0.94.2/9167/Wed Mar 25 19:06:32 2009
Copfilter-Virus-Scanned: AVG 7.5.52, engine 442 270.11.28/2022  2009-03-25
Copfilter-Virus-Scanned: F-PROT 6.2.1.4252 - Engine 4.4.4.56 - Virusdatabase 2009-03-24


From sdr at durietz.se  Thu Mar 26 14:30:33 2009
From: sdr at durietz.se (Stefan Du Rietz)
Date: Thu, 26 Mar 2009 14:30:33 +0100
Subject: [scilab-Users] in16 matrix
In-Reply-To: <1238007570.6037.8.camel@stargate>
References: <1238007570.6037.8.camel@stargate>
Message-ID: <49CB8379.40707@durietz.se>

I quote from Scilab help:

  y=int16(X)

Regards
Stefan

On 2009-03-25 19:59, ron wrote:
--------------------
> Hello,
> I am wondering if it is possible to create an int16 type matrix. This
> would save stack memory in my application since the matrix is huge.
> How can I do this?
> thank you.
> ron


From Francis.Drossaert at pgs.com  Thu Mar 26 15:14:22 2009
From: Francis.Drossaert at pgs.com (Francis Drossaert)
Date: Thu, 26 Mar 2009 14:14:22 -0000
Subject: [scilab-Users] in16 matrix
In-Reply-To: <49CB8379.40707@durietz.se>
References: <1238007570.6037.8.camel@stargate> <49CB8379.40707@durietz.se>
Message-ID: <6D1269A4C2D21240807B91421B187255068DF65B@lonms21.onshore.pgs.com>

Can I follow up on this with a question?

After a matrix conversion with uint8 or unint16 or uint32 I can't use tabul(y) anymore and plot(y) does not work anymore either. Although plot2d(y) seems to work, it seems that there are Scilab function which can't handle integer matrices. Is there a work-around? I typically work with vectors with 2e7 elements which 1.6e8 bytes so some memory savings would be welcome.

Francis

-----Original Message-----
From: Stefan Du Rietz [mailto:sdr at durietz.se] 
Sent: 26 March 2009 13:31
To: users at lists.scilab.org
Subject: Re: [scilab-Users] in16 matrix

I quote from Scilab help:

  y=int16(X)

Regards
Stefan

On 2009-03-25 19:59, ron wrote:
--------------------
> Hello,
> I am wondering if it is possible to create an int16 type matrix. This
> would save stack memory in my application since the matrix is huge.
> How can I do this?
> thank you.
> ron

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

From sdr at durietz.se  Thu Mar 26 15:39:00 2009
From: sdr at durietz.se (Stefan Du Rietz)
Date: Thu, 26 Mar 2009 15:39:00 +0100
Subject: [scilab-Users] in16 matrix
In-Reply-To: <6D1269A4C2D21240807B91421B187255068DF65B@lonms21.onshore.pgs.com>
References: <1238007570.6037.8.camel@stargate> <49CB8379.40707@durietz.se> <6D1269A4C2D21240807B91421B187255068DF65B@lonms21.onshore.pgs.com>
Message-ID: <49CB9384.2060906@durietz.se>

Are you plotting or using all the data at once? I use double(data(index)) as input of part of data to functions that cannot handle integer type matrices. E.g. I don't plot all my data at once.

If someone has another solution I would also appreciate to get it!

Stefan


On 2009-03-26 15:14, Francis Drossaert wrote:
--------------------
> Can I follow up on this with a question?
> 
> After a matrix conversion with uint8 or unint16 or uint32 I can't use tabul(y) anymore and plot(y) does not work anymore either. Although plot2d(y) seems to work, it seems that there are Scilab function which can't handle integer matrices. Is there a work-around? I typically work with vectors with 2e7 elements which 1.6e8 bytes so some memory savings would be welcome.
> 
> Francis
> 
> -----Original Message-----
> From: Stefan Du Rietz [mailto:sdr at durietz.se] 
> Sent: 26 March 2009 13:31
> To: users at lists.scilab.org
> Subject: Re: [scilab-Users] in16 matrix
> 
> I quote from Scilab help:
> 
>   y=int16(X)
> 
> Regards
> Stefan
> 
> On 2009-03-25 19:59, ron wrote:
> --------------------
>> Hello,
>> I am wondering if it is possible to create an int16 type matrix. This
>> would save stack memory in my application since the matrix is huge.
>> How can I do this?
>> thank you.
>> ron
> 
> 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



From Francis.Drossaert at pgs.com  Thu Mar 26 16:02:25 2009
From: Francis.Drossaert at pgs.com (Francis Drossaert)
Date: Thu, 26 Mar 2009 15:02:25 -0000
Subject: [scilab-Users] in16 matrix
In-Reply-To: <49CB9384.2060906@durietz.se>
References: <1238007570.6037.8.camel@stargate> <49CB8379.40707@durietz.se> <6D1269A4C2D21240807B91421B187255068DF65B@lonms21.onshore.pgs.com> <49CB9384.2060906@durietz.se>
Message-ID: <6D1269A4C2D21240807B91421B187255068DF65C@lonms21.onshore.pgs.com>

No, Scilab can't plot so much data; even though I have 8GB memory (Scilab is not particularly memory efficient). In my case it would not make much sense either. I need to use Boolean masks to pick out the data point belonging to certain sequences and sensor numbers.

I guess I could use double(data) as input. Probably it will save some memory but I doubt that it will speed up things.

Francis


-----Original Message-----
From: Stefan Du Rietz [mailto:sdr at durietz.se] 
Sent: 26 March 2009 14:39
To: users at lists.scilab.org
Subject: Re: [scilab-Users] in16 matrix

Are you plotting or using all the data at once? I use double(data(index)) as input of part of data to functions that cannot handle integer type matrices. E.g. I don't plot all my data at once.

If someone has another solution I would also appreciate to get it!

Stefan


On 2009-03-26 15:14, Francis Drossaert wrote:
--------------------
> Can I follow up on this with a question?
> 
> After a matrix conversion with uint8 or unint16 or uint32 I can't use tabul(y) anymore and plot(y) does not work anymore either. Although plot2d(y) seems to work, it seems that there are Scilab function which can't handle integer matrices. Is there a work-around? I typically work with vectors with 2e7 elements which 1.6e8 bytes so some memory savings would be welcome.
> 
> Francis
> 
> -----Original Message-----
> From: Stefan Du Rietz [mailto:sdr at durietz.se] 
> Sent: 26 March 2009 13:31
> To: users at lists.scilab.org
> Subject: Re: [scilab-Users] in16 matrix
> 
> I quote from Scilab help:
> 
>   y=int16(X)
> 
> Regards
> Stefan
> 
> On 2009-03-25 19:59, ron wrote:
> --------------------
>> Hello,
>> I am wondering if it is possible to create an int16 type matrix. This
>> would save stack memory in my application since the matrix is huge.
>> How can I do this?
>> thank you.
>> ron
> 
> 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


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

From hron at gmx.de  Thu Mar 26 19:04:37 2009
From: hron at gmx.de (ron)
Date: Thu, 26 Mar 2009 19:04:37 +0100
Subject: [scilab-Users] in16 matrix
In-Reply-To: <49CB8379.40707@durietz.se>
References: <1238007570.6037.8.camel@stargate>  <49CB8379.40707@durietz.se>
Message-ID: <1238090677.6007.10.camel@stargate>

Thanks Stefan,
but this is not what I want to do. 
y=int16(X)
makes a copy of 'X' and converts to int16. The result is even more used memory.

I figured out in the meantime that I can create int16-matrices using the file read-Function mget:
A=mgeti(10000,'us',fd);
this implicitly creates the vector A with int16 values.
Doing this in a loop using 
A(:,i)=...
does what I want.

The intention was to make nice plots of the data, but as I see in the other responses here this will not always work on integers.
ron


On Thu, 2009-03-26 at 14:30 +0100, Stefan Du Rietz wrote:
> I quote from Scilab help:
> 
>   y=int16(X)
> 
> Regards
> Stefan
> 
> On 2009-03-25 19:59, ron wrote:
> --------------------
> > Hello,
> > I am wondering if it is possible to create an int16 type matrix. This
> > would save stack memory in my application since the matrix is huge.
> > How can I do this?
> > thank you.
> > ron
> 


=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Filtered-With-Copfilter: Version 0.84beta4 (ProxSMTP 1.8)
Copfilter-Virus-Scanned: ClamAV 0.94.2/9170/Thu Mar 26 16:37:38 2009
Copfilter-Virus-Scanned: AVG 7.5.52, engine 442 270.11.29/2024  2009-03-26
Copfilter-Virus-Scanned: F-PROT 6.2.1.4252 - Engine 4.4.4.56 - Virusdatabase 2009-03-26


From sdr at durietz.se  Fri Mar 27 00:46:49 2009
From: sdr at durietz.se (Stefan Du Rietz)
Date: Fri, 27 Mar 2009 00:46:49 +0100
Subject: [scilab-Users] in16 matrix
In-Reply-To: <1238090677.6007.10.camel@stargate>
References: <1238007570.6037.8.camel@stargate>  <49CB8379.40707@durietz.se> <1238090677.6007.10.camel@stargate>
Message-ID: <49CC13E9.2060402@durietz.se>

But you can of course do
X = int16(X)
to avoid keeping the old X.

Stefan


On 2009-03-26 19:04, ron wrote:
--------------------
> Thanks Stefan,
> but this is not what I want to do. 
> y=int16(X)
> makes a copy of 'X' and converts to int16. The result is even more used memory.
> 
> I figured out in the meantime that I can create int16-matrices using the file read-Function mget:
> A=mgeti(10000,'us',fd);
> this implicitly creates the vector A with int16 values.
> Doing this in a loop using 
> A(:,i)=...
> does what I want.
> 
> The intention was to make nice plots of the data, but as I see in the other responses here this will not always work on integers.
> ron
> 
> 
> On Thu, 2009-03-26 at 14:30 +0100, Stefan Du Rietz wrote:
>> I quote from Scilab help:
>>
>>   y=int16(X)
>>
>> Regards
>> Stefan
>>
>> On 2009-03-25 19:59, ron wrote:
>> --------------------
>>> Hello,
>>> I am wondering if it is possible to create an int16 type matrix. This
>>> would save stack memory in my application since the matrix is huge.
>>> How can I do this?
>>> thank you.
>>> ron
> 
> 
> =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
> Filtered-With-Copfilter: Version 0.84beta4 (ProxSMTP 1.8)
> Copfilter-Virus-Scanned: ClamAV 0.94.2/9170/Thu Mar 26 16:37:38 2009
> Copfilter-Virus-Scanned: AVG 7.5.52, engine 442 270.11.29/2024  2009-03-26
> Copfilter-Virus-Scanned: F-PROT 6.2.1.4252 - Engine 4.4.4.56 - Virusdatabase 2009-03-26
>