gnuplot-info — For questions and discussion about gnuplot

Re: [Gnuplot-info] Problem with yerrorbars in histogram plot

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

transparent fill styles will be a new feature of the coming version 4.4.
but you can use this feature now by downloading gnuplot_4.3-cvs.


abhishek2301 wrote:
> 
> Hello,
> I have the following script to generate a histogram:
> 
> #set terminal png giant enhanced font 'Helvetica' 20
> #set output 'test.png'
> set boxwidth 0.9 absolute
> set style fill pattern 0
> set style data histogram
> set style histogram cluster gap 1
> set style histogram errorbars gap 2 lw 2 
> set auto x
> set yrange [0:100]
> set ylabel "Percentage of segments from Server"
> set xlabel "Overlay Size"
> set key box
> plot 'ServerLoadGraph.txt' using 2:3:xtic(1) t 'b=4',
> 'ServerLoadGraph.txt' using 4:5 t 'b=8', 'ServerLoadGraph.txt' using 6:7 t
> 'b=16'
> 
> The problem is that, the lower part of the errorbar is not visible. I have
> read some solution as to superimpose the pattern plot with another empty
> plot.
> But, how to achieve this??? Because, whenever I superimpose, the
> background patterned plot is no more visible and only the empty plot is
> visible.
> 
> Thank You!!!
> 

-- 
View this message in context: http://www.nabble.com/Problem-with-yerrorbars-in-histogram-plot-tp25391202p25414318.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] data operations on multiple data files in gnuplot

[Hans-Bernhard B. <HBB...@t-...>]

baalkikhaal wrote:

> i want to perform a summation of column 2 of each of given set of data files
> and then plot the sum against the common column 1

It may be possible to do such things inside gnuplot these days (using 
the CVS version, or 4.4 once that becomes a reality), but I think you 
should do this outside.  Command-line tools like 'paste', 'join' and 
'awk' should work just fine here.  Basically:

	while there are more than 1 files left:
	  paste lastfile nexttolastfile | \
           awk '{print $1, "\t", $2+$4}' > new_lastfile
           mv new_lastfile nexttolastfile

With a more complex script, awk could do the entire job.

[Gnuplot-info] data operations on multiple data files in gnuplot

[baalkikhaal <col...@gm...>]

hello,
i want to perform a summation of column 2 of each of given set of data files
and then plot the sum against the common column 1

any suggestions as to how to do this would be grateful
thnks in advance

sreekar
-- 
View this message in context: http://www.nabble.com/data-operations-on-multiple-data-files-in-gnuplot-tp25392977p25392977.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

[Gnuplot-info] Don't let gnuplot exit after "singular matrix" message?

[leo m. <llu...@go...>]

Dear all,

is it possible to set up gnuplot in a way that it won't exit after a
"singular matrix"
message during fitting?

I have a script like this...

----x------>

#!/usr/bin/gnuplot

set fit errorvariables

FIT_LIMIT = 1e-6
FIT_MAXITER = 80

f(x) = a * exp(b*x + c) + d
fit f(x) "exp.dat" using 2:3 via a, b, c, d
fx_err = a_err + b_err + c_err + d_err

g(x) = e + f*x + g*x**2 + h*x**3 + i*x**4 + j*x**5
fit g(x) "exp.dat" using 2:3 via e, f, g, h, i, j
gx_err = e_err + f_err + g_err + h_err + i_err + j_err

[...]

<-----x-------

After fitting to different models i'll choose the one with the
smallest residual and
let it plot to a png file.

Is there a way to "hack" this immediate exit?

Best regards

[Gnuplot-info] why are data values truncated to int values when using "set xdata time"

[Gautam T. <gau...@lm...>]

I plot a lot of data in which the x axis data is time. Input is "%s", meaning 
just time as floating point seconds is input. When I use

set format x "%M:%S"

i notice that in my graphs truncate the seconds value and thus the fraction of 
seconds are lost. This leads to loss of information when one is looking at the 
graph. As an example, consider data file "foo" and the attached "1.gplot"

doing

gnuplot 1.gplot -

shows the graph 1.png, but in fact I would like the fractional parts of the 
seconds to be not dropped and point should be draw in between the integral 
second values.

Any comments/hints welcome. Thanks. I have used gnuplot for a long time, but 
may be I am missing some trick here.

Gautam

[Gnuplot-info] millisecond time resolution

[Andrew H. <and...@gm...>]

Can someone please tell me what is the current status of integrating
millisecond resolution into gnuplot?  I can't seem to find much current
information on this topic, except that as of a year and a half ago, this
feature still didn't work despite being one of the most longstanding
requests.  Suggestions for a workaround to attain millisecond resolution
will also be most appreciated, as any "solutions" have since disappeared
from their original internet homes.

Thanks
-a

Re: [Gnuplot-info] Plot middle part of data file

[Hans-Bernhard B. <HBB...@t-...>]

babelproofreader wrote:
> I know I can plot the beginning and end of a data file using the <head &
> <tail commands, 

Not quite.  The commands are 'head' and 'tail', without the  '<'.  '<' 
is just the marker for gnuplot to know that you wanted input from a 
command's output, rather than a simple file.

And generally you shouldn't need 'head' or 'tail' --- gnuplot has the 
'every' option to the plot command to take care of those.

 > but what syntax do I use to plot a section from the middle?

'every', or a combination of 'head' and 'tail':

	plot '< head -100 file | tail -50'

[Gnuplot-info] Plot middle part of data file

[babelproofreader <bab...@gm...>]

I know I can plot the beginning and end of a data file using the <head &
<tail commands, but what syntax do I use to plot a section from the middle?
I've tried using various combinations of <head & <tail together but none of
them seem to work. Any suggestions?
-- 
View this message in context: http://www.nabble.com/Plot-middle-part-of-data-file-tp25332290p25332290.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] Extracting Data from Bezier Smoothing

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

have a look at 'set table':
http://www.gnuplot.info/docs/node267.html


ranlot wrote:
> 
> I have a file with 2 columns x and y for 2d plot.  I am smoothing these
> points with Bezier.  Is there a way to extract a datafile containing
> points along the smoothed curve?
> 

-- 
View this message in context: http://www.nabble.com/Extracting-Data-from-Bezier-Smoothing-tp25326599p25327875.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] Clustered Histogram with embedded data

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

using the same data with '-' as filename several times requires to give the
date
several times:
http://www.gnuplot.info/docs/node131.html


wdezell wrote:
> 
> I'm using Gnuplot as a visualization tool for a Qt app.  To streamline the
> process my Qt app generates a text file containing all Gnuplot setup,
> embeds the plot data in the file, and executes an external Gnuplot process
> with the script file passed as a parameter.  A simple working example is
> as follows:
> 
> --- begin script file contents ---
>     reset
>     set grid
>     set auto x
>     set yrange [0:]
>     set boxwidth 0.9 relative
>     set style fill solid 1.00 border -1
>     set style histogram gap 1 title offset character 0, 0, 0
>     set style data histogram 
>     set title "BLUE Recognition By Range
>     set ylabel "Count"  
>     unset key
>     plot '-' index 0 using 2:xtic(1) ti col  ls 11 
> 
>     xx  0
>     "0 <500m"  1
>     "500 <1Km"  1
>     "1K <2Km"  21
>     "2K <3Km"  42
>     "3K <4Km"  22
>     "4K <5Km"  15
>     ">5Km"  9
>     e
> 
> 
>     pause mouse keypress
> --- end script file contents ---
> 
> Now for the guidance I desperately need -- I need to do the same thing for
> a "Clustered Histogram".  The second example on the demo page at 
> http://gnuplot.sourceforge.net/demo_4.2/histograms.html
> http://gnuplot.sourceforge.net/demo_4.2/histograms.html  is basically what
> I need to accomplish.  However, I cannot figure out a correct "plot"
> syntax that will correctly support the right "using" syntax that will grab
> data from multi-column embedded data within the same "index".
> 
> As a start, I am trying to adapt the provided example to work in a
> single-file style.  I would think that the following "plot" syntax might
> work but it does not:
> 
>  plot '-' index 0 using 6:xtic(1) ti col, '-' index 0 u 12 ti col, \
>     '-' index 0 u 13 ti col, '-' index 0 u 14 ti col
> 
> I have included commands and data from the online example below packaged
> together as example of what I am trying to achieve.  How must I modify the
> "plot" line to make this work as a single script?
> 
> Any and all guidance is deeply appreciated.
> 
> Regards,
> Bill
> 
> ------ begin sample script from example ---------------
> # set terminal png transparent nocrop enhanced font arial 8 size 500,350 
> # set output 'histograms.2.png'
> set bar 1.000000
> set boxwidth 0.9 absolute
> set style fill  solid 1.00 border -1
> set style rectangle back fc lt -3 fillstyle  solid 1.00 border -1
> set key inside right top vertical Right noreverse enhanced autotitles
> columnhead nobox
> set style histogram clustered gap 1 title  offset character 0, 0, 0
> set datafile missing '-'
> set style data histograms
> set xtics border in scale 1,0.5 nomirror rotate by -45  offset character
> 0, 0, 0
> #set xtics  norangelimit
> set xtics   ("1891-1900" 0.00000 -1, "1901-1910" 1.00000 -1, "1911-1920"
> 2.00000 -1, "1921-1930" 3.00000 -1, "1931-1940" 4.00000 -1, "1941-1950"
> 5.00000 -1, "1951-1960" 6.00000 -1, "1961-1970" 7.00000 -1)
> set title "US immigration from Northern Europe\nPlot selected data columns
> as histogram of clustered boxes" 
> set rrange [ * : * ] noreverse nowriteback  # (currently [0.00000:10.0000]
> )
> set trange [ * : * ] noreverse nowriteback  # (currently
> [-5.00000:5.00000] )
> set urange [ * : * ] noreverse nowriteback  # (currently
> [-5.00000:5.00000] )
> set vrange [ * : * ] noreverse nowriteback  # (currently
> [-5.00000:5.00000] )
> set ylabel  offset character 0, 0, 0 font "" textcolor lt -1 rotate by 90
> set y2label  offset character 0, 0, 0 font "" textcolor lt -1 rotate by 90
> set yrange [ 0.00000 : 300000. ] noreverse nowriteback
> set cblabel  offset character 0, 0, 0 font "" textcolor lt -1 rotate by 90
> set locale "C"
> plot 'immigration.dat' using 6:xtic(1) ti col, '' u 12 ti col, '' u 13 ti
> col, '' u 14 ti col
> 
> # IMMIGRATION BY REGION AND SELECTED COUNTRY OF LAST RESIDENCE
> #
> Region  Austria Hungary Belgium Czechoslovakia  Denmark France  Germany
> Greece  Ireland Italy   Netherlands     Norway  Sweden  Poland  Portugal       
> Romania Soviet_Union    Spain   Switzerland     United_Kingdom  Yugoslavia     
> Other_Europe    TOTAL   
> 1891-1900       234081  181288  18167   -       50231   30770   505152 
> 15979   388416  651893  26758   95015   226266  96720   27508   12750  
> 505290  8731    31179   271538  -       282     3378014 
> 1901-1910       668209  808511  41635   -       65285   73379   341498 
> 167519  339065  2045877 48262   190505  249534  -       69149   53008  
> 1597306 27935   34922   525950  -       39945   7387494 
> 1911-1920       453649  442693  33746   3426    41983   61897   143945 
> 184201  146181  1109524 43718   66395   95074   4813    89732   13311  
> 921201  68611   23091   341408  1888    31400   4321887 
> 1921-1930       32868   30680   15846   102194  32430   49610   412202 
> 51084   211234  455315  26948   68531   97249   227734  29994   67646  
> 61742   28958   29676   339570  49064   42619   2463194 
> 1931-1940       3563    7861    4817    14393   2559    12623   144058 
> 9119    10973   68028   7150    4740    3960    17026   3329    3871   
> 1370    3258    5512    31572   5835    11949   377566  
> 1941-1950       24860   3469    12189   8347    5393    38809   226578 
> 8973    19789   57661   14860   10100   10665   7571    7423    1076   
> 571     2898    10547   139306  1576    8486    621147  
> 1951-1960       67106   36637   18575   918     10984   51121   477765 
> 47608   43362   185491  52277   22935   21697   9985    19588   1039   
> 671     7894    17675   202824  8225    16350   1325727 
> 1961-1970       20621   5401    9192    3273    9201    45237   190796 
> 85969   32966   214111  30606   15484   17116   53539   76065   3531   
> 2465    44659   18453   213822  20381   11604   1124492 
> e
> 
> 
> pause mouse keypress
> ------ end sample script from example -----------------
> 
> 
> 

-- 
View this message in context: http://www.nabble.com/Clustered-Histogram-with-embedded-data-tp25244114p25317387.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] problem specifying fill color for boxes

[Corey H. <bug...@fa...>]

Thomas Sefzick wrote:
> it's an error in 'graphics.c'
> 
> you can see the difference when plotting without color specification:
> plot '-' using 1:2
> there everthing is ok.
> 
> i'll upload a patch on sorceforge.

I never thanked you for fixing this.

So... thank you! Sorry the thanks took so long. I hadn't realized the
fix was committed to CVS. I tested today's CVS and it works great.

-Corey

[Gnuplot-info] Interactive plot manipulation

[CLOSE D. <Dav...@us...>]

Is there a way to allow interactive plot manipulation without going into
interactive command mode? This is in an X11 environment (on Linux).

If I write, "gnuplot file -", the commands in file are executed and,
since they include a plot command, a plot window appears. I can then
interact with that plot window to scale it or select a portion for
closer examination. When I type "q", the plot window closes but the
gnuplot command interface is still active. I must then also type "exit"
(or ^D) to terminate the program.

If I write, "gnuplot file", the commands in file are executed and the
plot window briefly appears. But it immediately closes and the program
terminates.

What I want to accomplish is a mixture of the two results. I want the
plot window to stay open until I type "q", but when I do I also want the
entire program to end automatically. Is that possible?

[Gnuplot-info] Clustered Histogram with embedded data

[wdezell <Bi...@CL...>]

I'm using Gnuplot as a visualization tool for a Qt app.  To streamline the
process my Qt app generates a text file containing all Gnuplot setup, embeds
the plot data in the file, and executes an external Gnuplot process with the
script file passed as a parameter.  A simple working example is as follows:

--- begin script file contents ---
    reset
    set grid
    set auto x
    set yrange [0:]
    set boxwidth 0.9 relative
    set style fill solid 1.00 border -1
    set style histogram gap 1 title offset character 0, 0, 0
    set style data histogram 
    set title "BLUE Recognition By Range
    set ylabel "Count"  
    unset key
    plot '-' index 0 using 2:xtic(1) ti col  ls 11 

    xx  0
    "0 <500m"  1
    "500 <1Km"  1
    "1K <2Km"  21
    "2K <3Km"  42
    "3K <4Km"  22
    "4K <5Km"  15
    ">5Km"  9
    e


    pause mouse keypress
--- end script file contents ---

Now for the guidance I desperately need -- I need to do the same thing for a
"Clustered Histogram".  The second example on the demo page at 
http://gnuplot.sourceforge.net/demo_4.2/histograms.html
http://gnuplot.sourceforge.net/demo_4.2/histograms.html  is basically what I
need to accomplish.  However, I cannot figure out a correct "plot" syntax
that will correctly support the right "using" syntax that will grab data
from multi-column embedded data within the same "index".

As a start, I am trying to adapt the provided example to work in a
single-file style.  I would think that the following "plot" syntax might
work but it does not:

 plot '-' index 0 using 6:xtic(1) ti col, '-' index 0 u 12 ti col, \
    '-' index 0 u 13 ti col, '-' index 0 u 14 ti col

I have included commands and data from the online example below packaged
together as example of what I am trying to achieve.  How must I modify the
"plot" line to make this work as a single script?

Any and all guidance is deeply appreciated.

Regards,
Bill

------ begin sample script from example ---------------
# set terminal png transparent nocrop enhanced font arial 8 size 500,350 
# set output 'histograms.2.png'
set bar 1.000000
set boxwidth 0.9 absolute
set style fill  solid 1.00 border -1
set style rectangle back fc lt -3 fillstyle  solid 1.00 border -1
set key inside right top vertical Right noreverse enhanced autotitles
columnhead nobox
set style histogram clustered gap 1 title  offset character 0, 0, 0
set datafile missing '-'
set style data histograms
set xtics border in scale 1,0.5 nomirror rotate by -45  offset character 0,
0, 0
#set xtics  norangelimit
set xtics   ("1891-1900" 0.00000 -1, "1901-1910" 1.00000 -1, "1911-1920"
2.00000 -1, "1921-1930" 3.00000 -1, "1931-1940" 4.00000 -1, "1941-1950"
5.00000 -1, "1951-1960" 6.00000 -1, "1961-1970" 7.00000 -1)
set title "US immigration from Northern Europe\nPlot selected data columns
as histogram of clustered boxes" 
set rrange [ * : * ] noreverse nowriteback  # (currently [0.00000:10.0000] )
set trange [ * : * ] noreverse nowriteback  # (currently [-5.00000:5.00000]
)
set urange [ * : * ] noreverse nowriteback  # (currently [-5.00000:5.00000]
)
set vrange [ * : * ] noreverse nowriteback  # (currently [-5.00000:5.00000]
)
set ylabel  offset character 0, 0, 0 font "" textcolor lt -1 rotate by 90
set y2label  offset character 0, 0, 0 font "" textcolor lt -1 rotate by 90
set yrange [ 0.00000 : 300000. ] noreverse nowriteback
set cblabel  offset character 0, 0, 0 font "" textcolor lt -1 rotate by 90
set locale "C"
plot 'immigration.dat' using 6:xtic(1) ti col, '' u 12 ti col, '' u 13 ti
col, '' u 14 ti col

# IMMIGRATION BY REGION AND SELECTED COUNTRY OF LAST RESIDENCE
#
Region  Austria Hungary Belgium Czechoslovakia  Denmark France  Germany
Greece  Ireland Italy   Netherlands     Norway  Sweden  Poland  Portugal       
Romania Soviet_Union    Spain   Switzerland     United_Kingdom  Yugoslavia     
Other_Europe    TOTAL   
1891-1900       234081  181288  18167   -       50231   30770   505152 
15979   388416  651893  26758   95015   226266  96720   27508   12750  
505290  8731    31179   271538  -       282     3378014 
1901-1910       668209  808511  41635   -       65285   73379   341498 
167519  339065  2045877 48262   190505  249534  -       69149   53008  
1597306 27935   34922   525950  -       39945   7387494 
1911-1920       453649  442693  33746   3426    41983   61897   143945 
184201  146181  1109524 43718   66395   95074   4813    89732   13311  
921201  68611   23091   341408  1888    31400   4321887 
1921-1930       32868   30680   15846   102194  32430   49610   412202 
51084   211234  455315  26948   68531   97249   227734  29994   67646  
61742   28958   29676   339570  49064   42619   2463194 
1931-1940       3563    7861    4817    14393   2559    12623   144058  9119   
10973   68028   7150    4740    3960    17026   3329    3871    1370    3258   
5512    31572   5835    11949   377566  
1941-1950       24860   3469    12189   8347    5393    38809   226578  8973   
19789   57661   14860   10100   10665   7571    7423    1076    571     2898   
10547   139306  1576    8486    621147  
1951-1960       67106   36637   18575   918     10984   51121   477765 
47608   43362   185491  52277   22935   21697   9985    19588   1039    671    
7894    17675   202824  8225    16350   1325727 
1961-1970       20621   5401    9192    3273    9201    45237   190796 
85969   32966   214111  30606   15484   17116   53539   76065   3531    2465   
44659   18453   213822  20381   11604   1124492 
e


pause mouse keypress
------ end sample script from example -----------------


-- 
View this message in context: http://www.nabble.com/Clustered-Histogram-with-embedded-data-tp25244114p25244114.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

[Gnuplot-info] Gnuplot Book: Finally (finally!) in print!

[Philipp K. J. <ja...@ie...>]

After what at times seemed like an interminable
delay, the Gnuplot book is finally available in 
printed form!

This means that anybody who pre-ordered the 
book should by now already have their copy or
receive it within the next few days.

I wanted to thank all of you who contributed 
questions and comments, or pointed out errors in 
the draft version of the manuscript. It really made 
a difference!

More details are available on the publisher's site:
	www.manning.com/janert

and the book is available from any bookseller,
including Amazon:
	http://www.amazon.com/gp/product/1933988398

I wanted to thank everyone for their patience and
also wanted to apologize again for the delay. I hope 
you will find that it was worth it.

Best,

		Ph.

Re: [Gnuplot-info] split y-axis in intervals

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

- you could subtract e.g. '20' from the 2nd data set to
  bring the data into the same range (assuming that the
  datafile has the format "x y1 y2"):
  plot 'datafile' using 1:2, '' using 1:($3-20.0)
  but you lose the original data values

- you could use the right side y-axis for the other plot
  range:
  set yrange [0:2] nomirror
  set y2range [20:22]
  plot 'datafile' using 1:2, '' using 1:3 axes x1y2


dmainzer wrote:
> 
> Hi,
> 
> i want to plot a data file but there is a big different in there values.
> So there is a lot of space between the two curves. Now i want
> to split the yrange for example into inverals of [0:2] and [20:22]; 
> is this possible ?
> 
> There is no demo and i take a look in the help page but there is nothing
> written about that.
> 
> Best,
> David
> 
> 
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008
> 30-Day 
> trial. Simplify your report design, integration and deployment - and focus
> on 
> what you do best, core application coding. Discover what's new with 
> Crystal Reports now.  http://p.sf.net/sfu/bobj-july
> _______________________________________________
> Gnuplot-info mailing list
> Gnu...@li...
> https://lists.sourceforge.net/lists/listinfo/gnuplot-info
> 
> 

-- 
View this message in context: http://www.nabble.com/split-y-axis-in-intervals-tp25148845p25149626.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

[Gnuplot-info] split y-axis in intervals

[david m. <dm...@tu...>]

Hi,

i want to plot a data file but there is a big different in there values.
So there is a lot of space between the two curves. Now i want
to split the yrange for example into inverals of [0:2] and [20:22]; 
is this possible ?

There is no demo and i take a look in the help page but there is nothing
written about that.

Best,
David

Re: [Gnuplot-info] stacked charts

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

yes, it is a version thing - you need at least version 4.2
have a look at the FAQ:
http://www.gnuplot.info/faq/faq.html#SECTION00040000000000000000


bistritapcv wrote:
> 
> When I try 
> 
>     set style data histogram
> 
> I get an error that it expected a bunch of stuff (points, lines, histeps,
> etc) but would not take histograms.
> 
> Is this some version thing?  I have version 4.0
> 
> Do I need to update?
> 
> What is the easiest way to update?
> 
> 
> -----Original Message-----
> From: Thomas Sefzick [mailto:t.s...@fz...]
> Sent: Saturday, August 22, 2009 09:55 AM
> To: gnu...@li...
> Subject: Re: [Gnuplot-info] stacked charts
> 
> 
> have a look at:
> http://gnuplot.sourceforge.net/demo/histograms.html
> 
> 
> bistritapcv wrote:
>> 
>> In another message I mentioned I made stacked charts.  Here is my
>> question
>> about them.
>> It does not seem there is a way to stack charts.  If you plot 2 bars,
>> they
>> will both start at 0
>> and one will overwrite part (or all of) the other one.
>> 
>> I had 5 fields to plot (say 2, 3, 4, 5, 6) with the x value as field 1
>> (1,
>> 2, 3, ... 24).  What I end up doing is
>> 
>> plot "file" using 1:($2+$3+$4+$5+$6) with boxes title "field5",
>>       "file" using 1:($2+$3+$4+$5) with boxes title "field4",
>>       "file" using 1:($2+$3+$4) with boxes title "field3",
>>       "file" using 1:($2+$3) with boxes title "field2",
>>      "file" using i:2 with boxes title "field1"
>> 
>> but it seems like there should be a simpler way to stack them than that. 
>> Is there?
>> 
>> 
> 
> -- 
> View this message in context:
> http://www.nabble.com/stacked-charts-tp25085687p25094517.html
> Sent from the Gnuplot - User mailing list archive at Nabble.com.
> 
> 
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008
> 30-Day 
> trial. Simplify your report design, integration and deployment - and focus
> on 
> what you do best, core application coding. Discover what's new with 
> Crystal Reports now.  http://p.sf.net/sfu/bobj-july
> _______________________________________________
> Gnuplot-info mailing list
> Gnu...@li...
> https://lists.sourceforge.net/lists/listinfo/gnuplot-info
> 
> 
> 
> 
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008
> 30-Day 
> trial. Simplify your report design, integration and deployment - and focus
> on 
> what you do best, core application coding. Discover what's new with 
> Crystal Reports now.  http://p.sf.net/sfu/bobj-july
> _______________________________________________
> Gnuplot-info mailing list
> Gnu...@li...
> https://lists.sourceforge.net/lists/listinfo/gnuplot-info
> 
> 

-- 
View this message in context: http://www.nabble.com/stacked-charts-tp25085687p25103128.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] stacked charts

[<ton...@le...>]

When I try 

    set style data histogram

I get an error that it expected a bunch of stuff (points, lines, histeps,
etc) but would not take histograms.

Is this some version thing?  I have version 4.0

Do I need to update?

What is the easiest way to update?


-----Original Message-----
From: Thomas Sefzick [mailto:t.s...@fz...]
Sent: Saturday, August 22, 2009 09:55 AM
To: gnu...@li...
Subject: Re: [Gnuplot-info] stacked charts


have a look at:
http://gnuplot.sourceforge.net/demo/histograms.html


bistritapcv wrote:
> 
> In another message I mentioned I made stacked charts.  Here is my question
> about them.
> It does not seem there is a way to stack charts.  If you plot 2 bars, they
> will both start at 0
> and one will overwrite part (or all of) the other one.
> 
> I had 5 fields to plot (say 2, 3, 4, 5, 6) with the x value as field 1 (1,
> 2, 3, ... 24).  What I end up doing is
> 
> plot "file" using 1:($2+$3+$4+$5+$6) with boxes title "field5",
>       "file" using 1:($2+$3+$4+$5) with boxes title "field4",
>       "file" using 1:($2+$3+$4) with boxes title "field3",
>       "file" using 1:($2+$3) with boxes title "field2",
>      "file" using i:2 with boxes title "field1"
> 
> but it seems like there should be a simpler way to stack them than that. 
> Is there?
> 
> 

-- 
View this message in context: http://www.nabble.com/stacked-charts-tp25085687p25094517.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.


------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Gnuplot-info mailing list
Gnu...@li...
https://lists.sourceforge.net/lists/listinfo/gnuplot-info

Re: [Gnuplot-info] stacked charts

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

have a look at:
http://gnuplot.sourceforge.net/demo/histograms.html


bistritapcv wrote:
> 
> In another message I mentioned I made stacked charts.  Here is my question
> about them.
> It does not seem there is a way to stack charts.  If you plot 2 bars, they
> will both start at 0
> and one will overwrite part (or all of) the other one.
> 
> I had 5 fields to plot (say 2, 3, 4, 5, 6) with the x value as field 1 (1,
> 2, 3, ... 24).  What I end up doing is
> 
> plot "file" using 1:($2+$3+$4+$5+$6) with boxes title "field5",
>       "file" using 1:($2+$3+$4+$5) with boxes title "field4",
>       "file" using 1:($2+$3+$4) with boxes title "field3",
>       "file" using 1:($2+$3) with boxes title "field2",
>      "file" using i:2 with boxes title "field1"
> 
> but it seems like there should be a simpler way to stack them than that. 
> Is there?
> 
> 

-- 
View this message in context: http://www.nabble.com/stacked-charts-tp25085687p25094517.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] tables in graphs

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

have a look at the 5th example in:
http://gnuplot.sourceforge.net/demo/datastrings.html


bistritapcv wrote:
> 
> I made a nice stacked bar chart with 5 different bars.  The x axis (i.e
> xtics is hour number from 1-24 (rather than
> from 0-23 since the person providing the data uses 1-24).  The Y axis has
> the 5 bars, each measuring a different
> quantity, though most are time measurements like number of seconds calls
> were made, etc).
> 
> The Y axis sometimes goes in increments of 5000, so you could see a bar
> somewhere halfway between 5000 and 10000.  The problem is, the user just
> has to kind of guess.  Somewhere between 5000 and 10000 could be 7000,
> 7500, 8000, etc.  If it is part way between the two it is even harder to
> know.
> 
> I want to be able also to print the actual value each hour (5 values) at
> each hour.  Maybe in the form of a table under each tic or something, but
> on the same graph as where the chart is.
> 
> I have searched gnuplot info but haven't found anything.  Can anyone help?
> 
> Thanks
> 

-- 
View this message in context: http://www.nabble.com/tables-in-graphs-tp25085603p25094501.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

[Gnuplot-info] stacked charts

[<ton...@le...>]

In another message I mentioned I made stacked charts.  Here is my question about them.
It does not seem there is a way to stack charts.  If you plot 2 bars, they will both start at 0
and one will overwrite part (or all of) the other one.

I had 5 fields to plot (say 2, 3, 4, 5, 6) with the x value as field 1 (1, 2, 3, ... 24).  What I end up doing is

plot "file" using 1:($2+$3+$4+$5+$6) with boxes title "field5",
      "file" using 1:($2+$3+$4+$5) with boxes title "field4",
      "file" using 1:($2+$3+$4) with boxes title "field3",
      "file" using 1:($2+$3) with boxes title "field2",
     "file" using i:2 with boxes title "field1"

but it seems like there should be a simpler way to stack them than that.  Is there?

[Gnuplot-info] data values in graphs

[<ton...@le...>]

I made a nice stacked bar chart with 5 different bars.  The x axis (i.e xtics is hour number from 1-24 (rather than
from 0-23 since the person providing the data uses 1-24).  The Y axis has the 5 bars, each measuring a different
quantity, though most are time measurements like number of seconds calls were made, etc).

The Y axis sometimes goes in increments of 5000, so you could see a bar somewhere halfway between 5000 and 10000.  The problem is, the user just has to kind of guess.  Somewhere between 5000 and 10000 could be 7000, 7500, 8000, etc.  If it is part way between the two it is even harder to know.

I want to be able also to print the actual value each hour (5 values) at each hour.  Maybe in the form of a table under each tic or something, but on the same graph as where the chart is.

I have searched gnuplot info but haven't found anything.  Can anyone help?

Thanks
Tony

Re: [Gnuplot-info] Progressive plots

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

write the 'plot' command into a file, followed by
a 'pause' and a 'reread' command, and then 'load'
the file.
see:
http://www.gnuplot.info/docs/node99.html
http://www.gnuplot.info/docs/node151.html
http://www.gnuplot.info/docs/node97.html


simulatorp wrote:
> 
> Hi 
> 
> I have a simulation running, and would like to see the progress of the
> solution live. Like using tail -f log to view the values, I would like to
> see the values update through the plot so that I know how the numbers
> behave. Is there a way?
> 
> Right now I am using >plot 'log' using 1:2, 'log' using '1:3' with lines
> 
> Thanks 
> 
> Prapanj.
> 

-- 
View this message in context: http://www.nabble.com/Progressive-plots-tp25041726p25041958.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] Graph coincides with axes and disappears

[Torquil M. S. <to...@gm...>]

Hans-Bernhard Bröker wrote:
> Torquil Macdonald Sørensen wrote:
> 
>> Is it possible to specify that the autoscale function should add a 
>> small amount, e.g. 5%, so that the whole graph is visible? 
> 
> Not quite, but close enough.  What you're looking for is explained in 
> 'help offsets'.
> 

Thanks Hans-Bernard, that was just the ticket! I will use

set offsets 0, 0, graph 0.05, graph 0.05

in all my plots.

Best regards
Torquil Sørensen
Norway

Re: [Gnuplot-info] Graph coincides with axes and disappears

[Hans-Bernhard B. <HBB...@t-...>]

Torquil Macdonald Sørensen wrote:

> Is it possible to specify that the autoscale function should add a small 
> amount, e.g. 5%, so that the whole graph is visible? 

Not quite, but close enough.  What you're looking for is explained in 
'help offsets'.