gnuplot-info — For questions and discussion about gnuplot

Re: [Gnuplot-info] gnuplot stuck at 100% CPU when put in background in a shell script

[Cook, R. <co...@ll...>]

Sorry, I was not clear.  This is not a desktop machine.  This is a large cluster with many users and part of my role is to make gnuplot work well on it.  Taking 100% cpu on the login node of our cluster with hundreds of users is a bad thing actually and we try to avoid it.  

I tried gnuplot 4.6.0 and it has the same behavior. 

Compiling without readline might or might not solve the problem, but creates a worse one in my view -- readline is a "necessary convenience."  :-)  I can also launch gnuplot without wrapping it in a shell script, but this again is not a desirable solution.  Or we could use Tecplot instead, but gnuplot has some really sweet features and is very popular here.    

My question is, what is it about being put into the background that causes gnuplot to spin up to 100% CPU looking for a keystroke?  

Thanks
-- Rich 

On Sep 7, 2012, at 11:41 PM, walter harms <wh...@bf...> wrote:

> hi,
> taking 100% cpu time is not a bad thing. that simply means it takes all powers
> it can get to do its job. the problems start when you can not start a second programm
> anymore because the system is so busy.
> 
> The first question is always:
> does the latest version 4.6 behave the same ? (maybe the bug is fixed already)
> 
> to establish that libreadline is the problem, can you create a version of gnuplot without
> libreadline ?
> 
> 
> re,
> wh
> 
> 
> 
> Am 08.09.2012 02:40, schrieb Cook, Rich:
>> Hello, 
>> I'm sorry for such an odd question as a newbie poster. 
>> I have built and installed gnuplot and wrapped it in a wrapper script.  The script simply makes a usage entry in a database, calls gnuplot, then notes that it exits.  I'm including the script below.  
>> 
>> Anyhow, when the user runs this script and puts it into the background, gnuplot immediately starts taking up 100% of a CPU, spinning in readline.  This only happens when gnuplot is wrapped in this way, not when it's called directly.  I would like to avoid this behavior, obviously, but I would also like to continue using a wrapper script.  What is it about being called from a script that is causing this odd behavior on gnuplot's part?  
>> 
>> Thanks.  
>> 
>> Here's a stack trace from a debugger of gnuplot when it's hammering our CPU:  
>> Stack Trace
>>     tcsetattr,           FP=7fffffffd060
>>     rl_tty_unset_default_bindings, FP=7fffffffd080
>>     rl_deprep_terminal,  FP=7fffffffd090
>>     rl_cleanup_after_signal, FP=7fffffffd0a0
>>     rl_reset_after_signal, FP=7fffffffd140
>>     rl_read_key,         FP=7fffffffd160
>>     readline_internal_char, FP=7fffffffd180
>>     readline,            FP=7fffffffd190
>> C    read_line,           FP=7fffffffd1e0
>> C    com_line,            FP=7fffffffd1f0
>> C    main,                FP=7fffffffd3e0
>>     __libc_start_main,   FP=7fffffffd4a0
>>     _start,              FP=7fffffffd4b0
>> 
>> 
>> Here's the wrapper script:  
>> 
>> #!/usr/bin/env bash
>> 
>> IMG_TRACK=${IMG_TRACK:-/usr/local/tools/imgtrack-1.0/bin/imgtrack}
>> if [ -f $IMG_TRACK ]; then 
>>    $IMG_TRACK BEGIN gnuplot-4.4.3 $0 "$@"
>> fi
>> /usr/local/tools/gnuplot-4.4.3/bin/gnuplot.real "$@"
>> _value=$?
>> if [ -f $IMG_TRACK ]; then 
>>    $IMG_TRACK END gnuplot-4.4.3 $0 "$@"
>> fi
>> exit $_value
>> 
>> 
>> 

Rich Cook

--------------------------------------
There is no real you, only neurons firing.  If you understand this, then you know you always have a choice.  Buddha says, "Wake up."  
--------------------------------------

Re: [Gnuplot-info] gnuplot stuck at 100% CPU when put in background in a shell script

[walter h. <wh...@bf...>]

hi,
taking 100% cpu time is not a bad thing. that simply means it takes all powers
it can get to do its job. the problems start when you can not start a second programm
anymore because the system is so busy.

The first question is always:
does the latest version 4.6 behave the same ? (maybe the bug is fixed already)

to establish that libreadline is the problem, can you create a version of gnuplot without
libreadline ?


re,
 wh



Am 08.09.2012 02:40, schrieb Cook, Rich:
> Hello, 
> I'm sorry for such an odd question as a newbie poster. 
> I have built and installed gnuplot and wrapped it in a wrapper script.  The script simply makes a usage entry in a database, calls gnuplot, then notes that it exits.  I'm including the script below.  
> 
> Anyhow, when the user runs this script and puts it into the background, gnuplot immediately starts taking up 100% of a CPU, spinning in readline.  This only happens when gnuplot is wrapped in this way, not when it's called directly.  I would like to avoid this behavior, obviously, but I would also like to continue using a wrapper script.  What is it about being called from a script that is causing this odd behavior on gnuplot's part?  
> 
> Thanks.  
> 
> Here's a stack trace from a debugger of gnuplot when it's hammering our CPU:  
> Stack Trace
>      tcsetattr,           FP=7fffffffd060
>      rl_tty_unset_default_bindings, FP=7fffffffd080
>      rl_deprep_terminal,  FP=7fffffffd090
>      rl_cleanup_after_signal, FP=7fffffffd0a0
>      rl_reset_after_signal, FP=7fffffffd140
>      rl_read_key,         FP=7fffffffd160
>      readline_internal_char, FP=7fffffffd180
>      readline,            FP=7fffffffd190
> C    read_line,           FP=7fffffffd1e0
> C    com_line,            FP=7fffffffd1f0
> C    main,                FP=7fffffffd3e0
>      __libc_start_main,   FP=7fffffffd4a0
>      _start,              FP=7fffffffd4b0
> 
> 
> Here's the wrapper script:  
> 
> #!/usr/bin/env bash
> 
> IMG_TRACK=${IMG_TRACK:-/usr/local/tools/imgtrack-1.0/bin/imgtrack}
> if [ -f $IMG_TRACK ]; then 
>     $IMG_TRACK BEGIN gnuplot-4.4.3 $0 "$@"
> fi
> /usr/local/tools/gnuplot-4.4.3/bin/gnuplot.real "$@"
> _value=$?
> if [ -f $IMG_TRACK ]; then 
>     $IMG_TRACK END gnuplot-4.4.3 $0 "$@"
> fi
> exit $_value
> 
> 
> 

[Gnuplot-info] gnuplot stuck at 100% CPU when put in background in a shell script

[Cook, R. <co...@ll...>]

Hello, 
I'm sorry for such an odd question as a newbie poster. 
I have built and installed gnuplot and wrapped it in a wrapper script.  The script simply makes a usage entry in a database, calls gnuplot, then notes that it exits.  I'm including the script below.  

Anyhow, when the user runs this script and puts it into the background, gnuplot immediately starts taking up 100% of a CPU, spinning in readline.  This only happens when gnuplot is wrapped in this way, not when it's called directly.  I would like to avoid this behavior, obviously, but I would also like to continue using a wrapper script.  What is it about being called from a script that is causing this odd behavior on gnuplot's part?  

Thanks.  

Here's a stack trace from a debugger of gnuplot when it's hammering our CPU:  
Stack Trace
     tcsetattr,           FP=7fffffffd060
     rl_tty_unset_default_bindings, FP=7fffffffd080
     rl_deprep_terminal,  FP=7fffffffd090
     rl_cleanup_after_signal, FP=7fffffffd0a0
     rl_reset_after_signal, FP=7fffffffd140
     rl_read_key,         FP=7fffffffd160
     readline_internal_char, FP=7fffffffd180
     readline,            FP=7fffffffd190
C    read_line,           FP=7fffffffd1e0
C    com_line,            FP=7fffffffd1f0
C    main,                FP=7fffffffd3e0
     __libc_start_main,   FP=7fffffffd4a0
     _start,              FP=7fffffffd4b0


Here's the wrapper script:  

#!/usr/bin/env bash

IMG_TRACK=${IMG_TRACK:-/usr/local/tools/imgtrack-1.0/bin/imgtrack}
if [ -f $IMG_TRACK ]; then 
    $IMG_TRACK BEGIN gnuplot-4.4.3 $0 "$@"
fi
/usr/local/tools/gnuplot-4.4.3/bin/gnuplot.real "$@"
_value=$?
if [ -f $IMG_TRACK ]; then 
    $IMG_TRACK END gnuplot-4.4.3 $0 "$@"
fi
exit $_value



-- 
✐Richard Cook   
✇ Lawrence Livermore National Laboratory
Bldg-453 Rm-4024, Mail Stop L-557        
7000 East Avenue,  Livermore, CA, 94550, USA
☎ (office) (925) 423-9605    
☎ (fax) (925) 423-6961
---
Information Management & Graphics Grp., Services & Development Div., Integrated Computing & Communications Dept.
(opinions expressed herein are mine and not those of LLNL)

Re: [Gnuplot-info] deprecated color specification; how to set light foreground color for dark background?

[Thomas O. <tho...@or...>]

Am Wed, 5 Sep 2012 16:29:30 -0700
schrieb Tait <gnu...@t4...>: 

> The font/axis/etc. colors can all be specified in their respective
> "set ..." commands.

Yeah, while it wasn't particulary elegant, it was easy with the old way
to just repeat "white" often enough till everything is visible. Now I
have to determine the name of each thing that might be missing and
issue a separate command to colorize it. It's no deal breaker, but
since my intention is basically to get inverse colors (light on dark
instead of dark on light background), it would be uber-cool to just be
able to set one "foreground" color.

I'll cope with what's available, thanks for the responses in any case


Alrighty then,

Thomas

Re: [Gnuplot-info] deprecated color specification; how to set light foreground color for dark background?

[Tait <gnu...@t4...>]

> > Anyhow: What is the current way to achieve "reversed" colors in a plot? I was unable to find that information in the docs (recent release notes, the help).
>
> Nobody? Not even someone shouting RTFM?

I can look at the old help files when I'm on a different computer, but
the first color was background color, the next the default line/font
color, and the following colors were just added to the palette or
something, I think? The background option will set the background, as you
found. The font/axis/etc. colors can all be specified in their respective
"set ..." commands. Set xtics accepts a "textcolor"; set title and set
key likewise; set grid accepts a line colorspec; I believe "set border"
will accept a line colorspec as well, although it's not documented.

The combination of options should let you control the color of every
aspect of the plot.

Re: [Gnuplot-info] Question from a novice: Drawing simple polygons using Cartesian coordinates

[Morteza J. <jal...@gm...>]

Hi

1.
plot [-1:1] [-1:1] NaN notitle

In case you just want the object. But you have to specify at least the
yrange manually so change the second '-1:1' to anything that covers your
object. You can change first '-1:1' to ':' or any range that covers your
object.

2.
unset border
unset tic

3. Use 'set output "[filename].png"' after 'set terminal' and before
'plot'. It should be in your current working directory (directory where you
executed gnuplot). You can specify directory in the filename too.

On Tue, Sep 4, 2012 at 8:25 PM, Sarah Lemaire <sle...@ve...> wrote:

> I'm sure these questions are easy, but I'm brand new to gnuplot, and the
> documentation/demos/examples are not helpful. I ordered the book, but it
> won't arrive for another two days.
>
>
> 1.      I want to draw polygons on a graph using the vertex coordinates.
> The set object command is easy, but I can't figure out how to create a
> simple plot that just contains both polygons. If I enter
>
> plot [1:5] x
>
> I get a red line going through the graph. What is the best command for
> simply plotting the polygons I've defined using set object?
>
>
> 2.      Can I create a plot with no axes at all? Ideally, I would like
> just my polygons in the image.
>
>
> 3.      I would like to generate .png files. Supposely I can use
>
> set terminal png
> plot ...
>
> And that creates a file output.png someplace, but I searched for that
> file, and it doesn't exist. How do I create that file?
>
> Thanks in advance for your help,
>
> Sarah
>
> ------------------------------------------------------------------------------
> Live Security Virtual Conference
> Exclusive live event will cover all the ways today's security and
> threat landscape has changed and how IT managers can respond. Discussions
> will include endpoint security, mobile security and the latest in malware
> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> _______________________________________________
> gnuplot-info mailing list
> gnu...@li...
> https://lists.sourceforge.net/lists/listinfo/gnuplot-info
>

Re: [Gnuplot-info] deprecated color specification; how to set light foreground color for dark background?

[Thomas O. <tho...@or...>]

Am Wed, 22 Aug 2012 13:40:35 +0200
schrieb Thomas Orgis <tho...@or...>: 

> Anyhow: What is the current way to achieve "reversed" colors in a plot? I was unable to find that information in the docs (recent release notes, the help).

Nobody? Not even someone shouting RTFM?


Norighty then,

Thomas

[Gnuplot-info] Question from a novice: Drawing simple polygons using Cartesian coordinates

[Sarah L. <sle...@Ve...>]

I'm sure these questions are easy, but I'm brand new to gnuplot, and the documentation/demos/examples are not helpful. I ordered the book, but it won't arrive for another two days.


1.      I want to draw polygons on a graph using the vertex coordinates. The set object command is easy, but I can't figure out how to create a simple plot that just contains both polygons. If I enter

plot [1:5] x

I get a red line going through the graph. What is the best command for simply plotting the polygons I've defined using set object?


2.      Can I create a plot with no axes at all? Ideally, I would like just my polygons in the image.


3.      I would like to generate .png files. Supposely I can use

set terminal png
plot ...

And that creates a file output.png someplace, but I searched for that file, and it doesn't exist. How do I create that file?

Thanks in advance for your help,

Sarah

Re: [Gnuplot-info] problems with "set arrow" command: instead single arrow it produces in addition a vertical line across the whole graph

[JamesBrown <jam...@ma...>]

it works! many thanks Thomas!



Thomas Sefzick wrote:
> 
> you need the arrows in the 1st plot only, so
> unset arrow
> before doing the 2nd plot.
> 
> 
> JamesBrown wrote:
>> 
>> Dear all,
>> 
>> I have the following problem:
>> In a script file (attached below) I set a single  arrow like
>> 
>> set arrow 1 from 24.5, -4.3015 to 23.05, -4.30605 nohead lw 2
>> 
>> but instead giving the desired arrow this command outputs an arrow plus a
>> strange vertical line across the whole graph.
>> I have a similar problem with set arrow 2, but with arrows 3, 4 and 5
>> everything is alright though the command are the same.
>> It looks  like this strange behavior for the arrows 1 and 2 strongly
>> depends on the y-coordinate of the initial point.
>> 
>> I have now idea what is happening.
>> Could please anyone help me with this problem?
>> The   script file, the file with data for plotting and the output image
>> file are attached below.
>> 
>> Thank you in advance,
>> Sincerely Yours,
>> James
>> 
>> P.S.:
>> my gnuplot version: Version 4.4 patchlevel 2
>> system details: OS:  Linux 2.6.37.6-0.11-desktop i686
>>                             System:  openSUSE 11.4 (i586)
>>                             KDE:  4.6.00 (4.6.0) "release 6"
>> computer: laptop Asus X72J
>> http://old.nabble.com/file/p34348260/UTF-8BZW5lcmd5X2NvbnZlcmdlbmNlLmpwZw.jpeg 
>> http://old.nabble.com/file/p34348260/energy_convergence.gnu
>> energy_convergence.gnu 
>> 
> 
> 

-- 
View this message in context: http://old.nabble.com/problems-with-%22set-arrow%22-command%3A-instead-single-arrow-it-produces-in-addition-a-vertical-line-across-the-whole-graph-tp34348260p34348277.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] problems with "set arrow" command: instead single arrow it produces in addition a vertical line across the whole graph

[Thomas S. <t.s...@fz...>]

you need the arrows in the 1st plot only, so
unset arrow
before doing the 2nd plot.


JamesBrown wrote:
> 
> Dear all,
> 
> I have the following problem:
> In a script file (attached below) I set a single  arrow like
> 
> set arrow 1 from 24.5, -4.3015 to 23.05, -4.30605 nohead lw 2
> 
> but instead giving the desired arrow this command outputs an arrow plus a
> strange vertical line across the whole graph.
> I have a similar problem with set arrow 2, but with arrows 3, 4 and 5
> everything is alright though the command are the same.
> It looks  like this strange behavior for the arrows 1 and 2 strongly
> depends on the y-coordinate of the initial point.
> 
> I have now idea what is happening.
> Could please anyone help me with this problem?
> The   script file, the file with data for plotting and the output image
> file are attached below.
> 
> Thank you in advance,
> Sincerely Yours,
> James
> 
> P.S.:
> my gnuplot version: Version 4.4 patchlevel 2
> system details: OS:  Linux 2.6.37.6-0.11-desktop i686
>                             System:  openSUSE 11.4 (i586)
>                             KDE:  4.6.00 (4.6.0) "release 6"
> computer: laptop Asus X72J
> http://old.nabble.com/file/p34348260/UTF-8BZW5lcmd5X2NvbnZlcmdlbmNlLmpwZw.jpeg 
> http://old.nabble.com/file/p34348260/energy_convergence.gnu
> energy_convergence.gnu 
> 

-- 
View this message in context: http://old.nabble.com/problems-with-%22set-arrow%22-command%3A-instead-single-arrow-it-produces-in-addition-a-vertical-line-across-the-whole-graph-tp34348260p34348275.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

[Gnuplot-info] problems with "set arrow" command: instead single arrow it produces in addition a vertical line across the whole graph

[JamesBrown <jam...@ma...>]

Dear all,

I have the following problem:
In a script file (attached below) I set a single  arrow like

set arrow 1 from 24.5, -4.3015 to 23.05, -4.30605 nohead lw 2

but instead giving the desired arrow this command outputs an arrow plus a
strange vertical line across the whole graph.
I have a similar problem with set arrow 2, but with arrows 3, 4 and 5
everything is alright though the command are the same.
It looks  like this strange behavior for the arrows 1 and 2 strongly depends
on the y-coordinate of the initial point.

I have now idea what is happening.
Could please anyone help me with this problem?
The   script file, the file with data for plotting and the output image file
are attached below.

Thank you in advance,
Sincerely Yours,
James

P.S.:
my gnuplot version: Version 4.4 patchlevel 2
system details: OS:  Linux 2.6.37.6-0.11-desktop i686
                            System:  openSUSE 11.4 (i586)
                            KDE:  4.6.00 (4.6.0) "release 6"
computer: laptop Asus X72J
http://old.nabble.com/file/p34348260/UTF-8BZW5lcmd5X2NvbnZlcmdlbmNlLmpwZw.jpeg 
http://old.nabble.com/file/p34348260/energy_convergence.gnu
energy_convergence.gnu 
-- 
View this message in context: http://old.nabble.com/problems-with-%22set-arrow%22-command%3A-instead-single-arrow-it-produces-in-addition-a-vertical-line-across-the-whole-graph-tp34348260p34348260.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

[Gnuplot-info] mac wxt problem

[Peter <one...@ya...>]

Hello,

I have the problem to get gnuplot 4.6 to compile with mac os x 10.6

my mac runs  mac os x 10.6 with xcode_4.2, my macport was self updated to the latest version.

Here is the command I run:

 sudo /opt/local/bin/port install cairo +no_x11 +quartz +universal
 sudo /opt/local/bin/port install pango +no_x11 +quartz +universal
  ./configure CPPFLAGS='-arch i386' LDFLAGS='-arch i386' --with-readline=builtin

  --prefix=$HOME/gnuplot-4.6_bin --without-lisp-files
 make


I got the following errors.

In file included from ./plot.h:46,

                 from wxterminal/wxt_gui.h:113,
                 from wxterminal/wxt_gui.cpp:96:
./mouse.h:189: error: variable or field 'set_ruler' declared void
./mouse.h:189: error: '_Bool' was not declared in this scope
./mouse.h:189: error: expected primary-expression before 'int'
./mouse.h:189: error: expected primary-expression before 'int'
./mouse.h:197: error: '_Bool' has not been declared

In file included from wxterminal/wxt_gui.h:113,
                 from wxterminal/wxt_gui.cpp:96:
./plot.h:53: error: '_Bool' does not name a type
./plot.h:54: error: '_Bool' does not name a type
./plot.h:55: error: '_Bool' does not name a type

how to resolve such a problem ?

Thanks. 

Peter

[Gnuplot-info] deprecated color specification; how to set light foreground color for dark background?

[Thomas O. <tho...@or...>]

Hello,

I have programs that generate gnuplot scripts, and they used to work nicely with

	set terminal png size 1024x768 font arial 21 x000000 xffffff xffffff xffffff

to get black background and white boundaries, tics, labels, etc. Now my current gnuplot complains:

	line 0: obsolete color option

... and it doesn't work anymore. Reading the help for 'png' also mentions that additional colors are deprecated. It mentions 'background xrrggbb', which is wrong, btw; it should be 'background "#rrggbb"'.

Anyhow: What is the current way to achieve "reversed" colors in a plot? I was unable to find that information in the docs (recent release notes, the help).


Alrighty then,

Thomas

[Gnuplot-info] History and interactive/batch mode

[Elias A. <eli...@if...>]

Hi,

I use gnuplot a lot, and in some quite different contexts, for which 
different settings are useful.  So I tried putting shorthands like this 
in my .bashrc:

alias gnuplot-a="gnuplot -e 'some  commands;' -"
alias gnuplot-b="gnuplot -e 'other commands;' -"

(or, equivalently, "gnuplot init_file -"); the problem is that this puts 
gnuplot into batch mode and I cannot access the history, which kind of 
defeats the whole idea of saving keystrokes.

Here are two threads that I found that deal with this problem:

https://groups.google.com/d/topic/comp.graphics.apps.gnuplot/UGwZzqL84I8/discussion

http://tech.groups.yahoo.com/group/info-gnuplot-beta/message/2929

In the first, someone had the same problem and suggested putting gnuplot 
in interactive mode when reading from '-' but got the reply: "The '-' 
command-line argument is meant for use with programs
controlling gnuplot from the outside (after having loaded some scripts
first), not for actual interactive usage by people."

I understand that, but I really feel there should be a way to do what I 
am trying, be it with a command (e.g. 'load history' as suggested in the 
second thread above) or a switch (e.g. '--interactive').

Besides, gnuplot already treats an explicit '-' differently from 
no-arguments, and in the latter case seems to detect whether it is on a 
terminal:

$ gnuplot			# real interactive
gnuplot> print 2*pi
6.28318530717959
gnuplot>
$ echo print 2*pi | gnuplot	# real batch
6.28318530717959
$ echo print 2*pi | gnuplot -	# pseudo-interactive
gnuplot> 6.28318530717959
gnuplot>

So for '-', why not also check for a terminal and load the history in 
that case?  At the same time, one could eliminate the prompts when not 
on a terminal.

Or is there a solution I overlooked?  After all, the threads are not 
exactly new.  (I realize that I could fake it by writing a .gnuplot file 
on-the-fly, but that seems needlessly complicated and error-prone.)

Sorry for the long post.

	Elias

PS: Also, sorry if this is posted to the list more than once -- I had a 
problem where it was rejected at first, now I am no longer 100% sure if 
it got through before.

Re: [Gnuplot-info] Gnuplot doesn't show me the right x-axis intersection of a function

[Jefferson S. E. P. <st...@gm...>]

Hi Luigi,

to what Dieter Jurzitza already said I only add that you'll see better
your intersection with the x-axis if you adjust the range of the plot,
for example:
  set xrange [0:10]
  set xrange [0:10]
  plot -0.9+9.638*x**(-1.394) w l lw 2
With the default ranges all you'll see is indeed that divergence of
your function as x->0.

Cheers,
Jefferson


On Sat, Aug 18, 2012 at 11:06 PM, Luigi <dri...@li...> wrote:
> Hi. I was trying to plot the following function:
>
> -0.9+9.638x^(-1.394)
>
> by using the command
>
> plot y=-0.9+9.638x**(-1.394)
>
> The problem is that the intersection with the x-axis (y=0) should be at
> around x=5.47, but the function's plot just approaches y=0 as an asymptote.
>
> When I try to plot it with Maxima, the intersection is correct.
>
> Is there something wrong in the syntax of the command? Thanks.
>
> ------------------------------------------------------------------------------
> Live Security Virtual Conference
> Exclusive live event will cover all the ways today's security and
> threat landscape has changed and how IT managers can respond. Discussions
> will include endpoint security, mobile security and the latest in malware
> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> _______________________________________________
> gnuplot-info mailing list
> gnu...@li...
> https://lists.sourceforge.net/lists/listinfo/gnuplot-info

Re: [Gnuplot-info] Gnuplot doesn't show me the right x-axis intersection of a function

[Dr. I. D. J. <die...@t-...>]

Hi Luigi,
I am sure you did something wrong. First of all, your formula is erratic as it 
should be

f(x)=-0.9+9.638*x**(-1.394)

please note the "*" between the factor (9.638) and the power of x ...

secondly the command plot y= something does not work, what you ought to do is 
either 

plot -0.9+9.638*x**(-1.394) w l lw 2

or say

f(x)=-0.9+9.638*x**(-1.394)
plot f(x) w l lw 2

to provide an example. In both cases I get the zero crossing around x=5.479

Hope this helps,
take care




Dieter Jurzitza

On Samstag, 18. August 2012 23:06:27 Luigi wrote:
> Hi. I was trying to plot the following function:
> 
> -0.9+9.638x^(-1.394)
> 
> by using the command
> 
> plot y=-0.9+9.638x**(-1.394)
> 
> The problem is that the intersection with the x-axis (y=0) should be at
**************
-- 
-----------------------------------------------------------

                               |
                                \
                 /\_/\           |
                | ~x~ |/-----\   /
                 \   /-       \_/
  ^^__   _        /  _  ____   /
 <°°__ \- \_/     |  |/    |  |
  ||  ||         _| _|    _| _|

if you really want to see the pictures above - use some font
with constant spacing like courier! :-)
-----------------------------------------------------------

[Gnuplot-info] Gnuplot doesn't show me the right x-axis intersection of a function

[Luigi <dri...@li...>]

Hi. I was trying to plot the following function:

-0.9+9.638x^(-1.394)

by using the command

plot y=-0.9+9.638x**(-1.394)

The problem is that the intersection with the x-axis (y=0) should be at 
around x=5.47, but the function's plot just approaches y=0 as an asymptote.

When I try to plot it with Maxima, the intersection is correct.

Is there something wrong in the syntax of the command? Thanks.

[Gnuplot-info] Image as background in multiplot

[bitsaint <bit...@gm...>]

Hello everyone,
im trying to have 4 x-y axis in the results of multiplot. I want to have a
standard background in each x-y axis.
Unfortunatelly im getting the error "
plot "Floor2.png" binary filetype=png w rgbimage
                                  ^
"multiex.txt", line 12: Unsupported file type

"

You can see my gnuplot script below

set style function lines
set size 1.0, 1.0
set origin 0.0, 0.0
set output "map.png"
set multiplot
set size 0.5,0.5
set origin 0.0,0.5
set grid
unset key
set angles radians
set samples 250
plot "Floor2.png" binary filetype=jpg w rgbimage


#  Plot 1
set title "Stats 1"
set xlabel "x"
set ylabel "y"
set xrange [0:46]
set yrange [0:31]

plot "stats4.dat"  w p pt 7 lc rgb"#00FF00" notitle
plot "stats.dat" w p pt 7 lc rgb"#ff0000" notitle


#  Plot 2
set size 0.5,0.5
set origin 0.0,0.0
set title "Stats 2"
set ylabel "y" 
set xrange [0:46]
set yrange [0:31]
plot "stats4.dat"  w p pt 7 lc rgb"#00FF00" notitle
plot "stats.dat" w p pt 7 lc rgb"#ff0000" notitle

#  Plot 3
set size 0.5,0.5
set origin 0.5,0.5
set title "Stats 3"
set ylabel "y" 
set xlabel "x"
set xrange [0:46]
set yrange [0:31]

plot "stats4.dat"  w p pt 7 lc rgb"#00FF00" notitle
plot "stats.dat" w p pt 7 lc rgb"#ff0000" notitle
#

#  Plot 4
set origin .5,0.
set title "Stats 4"
set xrange [0:46]
set yrange [0:31]
plot "stats4.dat"  w p pt 7 lc rgb"#00FF00" notitle
plot "stats.dat" w p pt 7 lc rgb"#ff0000" notitle
  
unset multiplot
pause mouse any "Any key or mouse will terminate"

can someone help please? 

Thanx :)
-- 
View this message in context: http://old.nabble.com/Image-as-background-in-multiplot-tp34257136p34257136.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

[Gnuplot-info] bar chart; bars are on top of each other

[Tilly <new...@go...>]

Hello everyone!

I wanted to create a bar graph from three different dat. files. For the
x-value in 1918, the numbers of all files are displayed as a bar. The
structure of the data looks like this:

# num-nat.dat
1918       1542
1919       1478
...

# num-mem.dat
1918       13542
1919       14478
...

# num-small.dat
1918       17421
1919       16278
...

In the plot the bars are on top of each other and are covered by the
largest ones.

<Code>
set yrange [0:21000]
set xrange [1917.5:1929.5]
set xtics 1

set style data histogram
set style histogram clustered
set style fill solid border lt -1
#unset style fill
set boxwidth 0.25
set bars 1

plot "anzahl-nat.dat" title "FED" with boxes, "anzahl-mem.dat" title
"Member"  with boxes, "anzahl-small.dat" title "Small" with boxes
</Code>

Can someone please give an example of the right plot command?

Thanks,
Tilly

Re: [Gnuplot-info] import external function via pipe?

[Lev A. M. <mel...@ma...>]

Walter, thank you for your reply.

First, I don't care if the data are binary or not - "external function" is 
implemented by me. I can output whatever needed, be it text, binary, 
tertiary, anything.

What I do need is probably a bidirectional interprocess communication. 
Gnuplot (splot) sends the argument list for each grid point and my program 
sends back the function evaluated for these arguments.

<a> popen() is 1-way
<b> I don't know how to invoke popen from gnuplot

The function can not be effectively precomputed because the data range 
is not specified beforehand. And actually contour plotting requires a 
grid, and if I understand correctly (?) a grid can not be stored in the 
data file, this requires explicit (resource-expensive) conversion.

For example, this works:

f(x,y)=real(system(sprintf("./func %f %f",y,x)))
splot(f(x,y))

where the external program func is below:

#include <stdio.h>
#include <stdlib.h>

main(int argc, char **argv)
{
double x,y;

if(3!=argc)
    exit(-1);

x=atof(argv[1]);
y=atof(argv[2]);

printf("%g",x*x+y*y);

return(0);
}

I only want to speed it up. Possible?

Thanks
-Lev

wh> I am not sure that i understand what you are trying to archive
wh> 
wh> Writing the data to a file and reading that as matrix should be pretty fast.
wh> I assume you are aware that you can use binary data ? (http://gnuplot.sourceforge.net/demo_4.6/binary.html)
wh> 
wh> if you can write a gnuplot script you can uses popen() that should be
wh> much better that system().
wh> 
wh> Am 30.07.2012 22:42, schrieb Lev A. Melnikovsky:
wh> > Hi,
wh> > 
wh> > the function I need to plot is a result of nontrivial computation (each data 
wh> > point is obtained by solving an ODE). I have implemented it in C, but executing 
wh> > the program with
wh> > 
wh> > real(system(sprintf("somecommand %f %f", x, y)))
wh> > 
wh> > once for each data point (this is a 3d surface) implies a significant overhead. 
wh> > My program is statically linked, still this is too slow. I need grid data (to 
wh> > create contours), so building the datafile is inefficient.
wh> > 
wh> > Ideally I'd like to see an option to open a pipe and write the arguments to it 
wh> > and read the results from it. My program would be glad to work in this mode. Is 
wh> > this possible?
wh> > 
wh> > I'm using Linux if this matters.
wh> > 
wh> 
wh> re,
wh>  wh
wh> 

[Gnuplot-info] Matrix Time Format

[nhilker <Nat...@ec...>]

Hi,

I am attempting to create an image plot of 3-dimensional matrix data in
which the x-axis is date/time data. I have had success doing this using 'set
timefmt "%s"', but this seems to fail in certain scenarios. To see what I
mean, observe the following simplified examples.

Example 1:
The plot appears to function properly here.

testdat.csv:
4,20000001,20000002,20000003,20000004
0.265,10,8,5,30000
0.290,110,7,6,5000
0.320,1200,8,7,600	
0.375,13000,9,8,70
0.450,230,25,765,4
1.000,45,1000,20,1500

http://old.nabble.com/file/p34241078/testplot1.png 


Example 2:
Simply adding a zero to each date/time observation causes the image plot to
plot along a diagonal.
Interestingly enough, if you add yet another zero to these values, the plot
will function properly again. Can anyone explain what is going on here?

testdat.csv:
4,200000010,200000020,200000030,200000040
0.265,10,8,5,30000
0.290,110,7,6,5000
0.320,1200,8,7,600	
0.375,13000,9,8,70
0.450,230,25,765,4
1.000,45,1000,20,1500

http://old.nabble.com/file/p34241078/testplot2.png 


Example 3:
Finally, this example gives me the error that I am trying to resolve:
Warning: empty x range [2e+008:2e+008], adjusting to [1.98e+008:2.02e+008]
            "nonuniform_matrix.p", line 31: warning: Image grid must be at
least 2x2.

What is causing this error and how can I fix it?

testdat.csv:
4,200000001,200000002,200000003,200000004
0.265,10,8,5,30000
0.290,110,7,6,5000
0.320,1200,8,7,600	
0.375,13000,9,8,70
0.450,230,25,765,4
1.000,45,1000,20,1500

http://old.nabble.com/file/p34241078/testplot3.png 


Script (nonuniform_matrix.p):
set terminal png size 1024,768
set datafile separator ","
set output "testplot.png"
set xlabel "Date/Time"
set ylabel "Counts (Median Bin Diameter [µm])"

set xdata time
set timefmt "%s"
set format x "%Y-%m-%d\n%H:%M:%S"
set yrange[0.1915:1.0735]
set ytics("0.265" 0.265, "0.290" 0.412, "0.320" 0.559, "0.375" 0.706,
"0.450" 0.853, "1.000" 1)
		  
set palette rgbformulae 0, 13, 10
set palette negative

set logscale cb
set cbrange [1:100000]

set view map

splot "testdat.csv" nonuniform matrix with image failsafe
-- 
View this message in context: http://old.nabble.com/Matrix-Time-Format-tp34241078p34241078.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] import external function via pipe?

[walter h. <wh...@bf...>]

Am 30.07.2012 22:42, schrieb Lev A. Melnikovsky:
> Hi,
> 
> the function I need to plot is a result of nontrivial computation (each data 
> point is obtained by solving an ODE). I have implemented it in C, but executing 
> the program with
> 
> real(system(sprintf("somecommand %f %f", x, y)))
> 
> once for each data point (this is a 3d surface) implies a significant overhead. 
> My program is statically linked, still this is too slow. I need grid data (to 
> create contours), so building the datafile is inefficient.
> 
> Ideally I'd like to see an option to open a pipe and write the arguments to it 
> and read the results from it. My program would be glad to work in this mode. Is 
> this possible?
> 
> I'm using Linux if this matters.
> 

I am not sure that i understand what you are trying to archive

Writing the data to a file and reading that as matrix should be pretty fast.
I assume you are aware that you can use binary data ? (http://gnuplot.sourceforge.net/demo_4.6/binary.html)

if you can write a gnuplot script you can uses popen() that should be
much better that system().

re,
 wh

[Gnuplot-info] import external function via pipe?

[Lev A. M. <mel...@ma...>]

Hi,

the function I need to plot is a result of nontrivial computation (each data 
point is obtained by solving an ODE). I have implemented it in C, but executing 
the program with

real(system(sprintf("somecommand %f %f", x, y)))

once for each data point (this is a 3d surface) implies a significant overhead. 
My program is statically linked, still this is too slow. I need grid data (to 
create contours), so building the datafile is inefficient.

Ideally I'd like to see an option to open a pipe and write the arguments to it 
and read the results from it. My program would be glad to work in this mode. Is 
this possible?

I'm using Linux if this matters.

Thanks in advance
-L

[Gnuplot-info] pm3d plots only triangles

[Christian a. B. <sch...@gm...>]

Hi folks,

I want to plot a surface plot of the following file:
http://old.nabble.com/file/p34223810/population.txt population.txt

It is sorted by the first column and there is a line break whenever the
value in the first column changes. The format is:
x y f(x,y)
A normal splot yields reasonable results in 3D, but for better readability I
just want a surface map using pm3d. However, this yields only weird
triangles. Any ideas?

Thanks a lot in advance

[Gnuplot-info] pm3d plots only triangles

[whining_dwarf <ma...@ii...>]

Hi folks,

I want to plot a surface plot of the following file:
http://old.nabble.com/file/p34223810/population.txt population.txt 

It is sorted by the first column and there is a line break whenever the
value in the first column changes. The format is:
x y f(x,y)
A normal splot yields reasonable results in 3D, but for better readability I
just want a surface map using pm3d. However, this yields only weird
triangles. Any ideas?

Thanks a lot in advance

-- 
View this message in context: http://old.nabble.com/pm3d-plots-only-triangles-tp34223810p34223810.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.