gnuplot-info — For questions and discussion about gnuplot

[Gnuplot-info] Setting Rectangle Problems

[mikeincousa <mik...@gm...>]

Plot size: 11 x 17
Terminal:Windows
Goal:  an empty box in the lower right 

> set object 2 rectangle at 20, 0 size 10,1 fs empty border -1 lw 2 

#works, but with the dimensions determined by trial and error

x, and y for width and height do not agree with plot scale dimensions of 11
x 17

Same problem with from and to.

#interactive terminal quirk?

#something to do with coordinate system? (using default)?

#"border" is confusing. Seems that [rectangle] border as contrasted with the
plot border is actually set by choosing a default line type? I went to plot
border first.





--
View this message in context: http://gnuplot.10905.n7.nabble.com/Setting-Rectangle-Problems-tp20290.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] rotate labels by data column

[Ethan M. <eam...@gm...>]

I like the idea, and it requires only a trivial amount of code, so...
in today's CVS (version 5.1)

    http://gnuplot.sourceforge.net/demo_cvs/rotate_labels.html

using theozh's test data:

 plot $Data using 1:2:3:4 with labels rotate variable


Ethan

On Tue, Aug 2, 2016 at 12:07 AM, Grothausmann, Roman Dr. <
gro...@mh...> wrote:

> Dear mailing list members,
>
>
> Is there any way to plot labels rotated according to angles given by a
> data column?
> All I could find is:
> plot ... with labels rotate by X
> which, as I understand, only allows a fixed rotation for all labels.
>
> Any help or hints are very much appreciated
> Roman
>
>
> ------------------------------------------------------------------------------
> _______________________________________________
> gnuplot-info mailing list
> gnu...@li...
> Membership management via:
> https://lists.sourceforge.net/lists/listinfo/gnuplot-info
>

[Gnuplot-info] How to simply get contour or surface plot form datafile?

[Francesco L. R. <kec...@gm...>]

Hello,
I would like to use gnuplot to plot my 2d function ch(x,y) [obtained by a
pde solver, FreeFem++]. My datafile is structured in this way:


x1 y1 ch(x1,y1)
x2 x2 ch(x2,y2)
x3 y3 ch(x3,y3)
x1 y1 ch(x1,y1)


x4 y4 ch(x4,y4)
...
...
...


...


where I have the value of my function at the 3 vertices of every triangle
which constitutes my (triangular, not structured) mesh (see figure here
http://i65.tinypic.com/2mydkq9.jpg).
What I would like to get is two separate figures:
1. something like this for the surface, with a colorbar as a legend:
http://i68.tinypic.com/egvkzr.jpg
and
2. something like this for contour lines, with a colorbar as a legend:
http://i67.tinypic.com/msgepg.jpg


How can I get these two figures with gnuplot?
I've tried for example doing the first one :


set palette rgbformulae 33,13,10
set xrange [0: 0.25]
set yrange [0: 0.20]
set view map
splot "mydatafile.txt" w l pal


but the triangles inside are white (not filled with colors) and only the
edges of the triangles are coloured.
How about the contour lines?


Thank you in advance,


Keccogrin

Re: [Gnuplot-info] Bugs? 1) Grid ticks in logscale incorrect when range above 1e6+1 and 2) also not plotted completely

[D <dcm...@we...>]

Yes, ok, it's not a bug. But from a beginners perspective I just wanted 
to plot a graph and at least in log mode I had to "set sample <n>" for 
the graph to reach the bottom and top axis and be fully drawn. There's 
basically no fast way to know that this was necessary.
I don't know but maybe at least some helpful warning should appear when 
one uses a certain xy range, specially in log mode (I know it is not 
limited to log mode)?

Also of course the log ticks looked ok up to a certain xrange and I had 
to use:
"set mxtics 10                  # see "help set mxtics"
set mytics 10"
Seems also not beginner friendly or kinda buggy (because it works to a 
certain range and then not).

Anyways I just see possible improvements to be made here for a next version.

On 03.08.2016 00:08, Ethan Merritt wrote:
> There is no bug here - you are fighting the limits of floating point 
> precision.
> Consider the final steps that are sampled as x approaches 10000.
> For linear  sampling in steps of 100
>
> f(9800)   is 490000.000000002
> f(9900)   is 990000.00000001
> f(10000) is inf   (divide by zero at IEEE floating precision)
>
> So f(9900) is the last point that shows on the plot, and that value is 
> less
> than the y range you have chosen.  The function reaches ymax (10^7) at 
> f(9990.)
> To catch that on your plot you must sample finely enough to 
> distinguish 9990 from 10000.
>
> And that's for linear sampling. In logscale the sampling must be even 
> finer.
> The program selects even samples along the x axis (even spacing on 
> log(x)).
> So most of the samples are at low values of x.
>
> On Sun, Jul 31, 2016 at 5:16 AM, D <dcm...@we... 
> <mailto:dcm...@we...>> wrote:
>
>     Thanks. I had to 'set sample' to 500000 for gnuplot to plot the
>     whole line (slightly different xy range and the line goes to the
>     bottom). Works in linear mode without the need for any 'set
>     sample', wonder why the log mode requires that at all. Maybe you
>     should fix that for v5.1 too.
>
>
>     On 29.07.2016 20:52, Ethan Merritt wrote:
>>     1)   version 5.0:
>>           set mxtics 10                  # see "help set mxtics"
>>           set mytics 10
>>
>>          version 5.1
>>           currently a bug.  thanks for the report
>>
>>     2) set sample 10000
>>     Or else define the functions as parametric so that the sample is
>>     independent of the x/y precision.
>>     But increasing the number of samples is easier.
>>
>>     On Fri, Jul 29, 2016 at 11:28 AM, D <dcm...@we...
>>     <mailto:dcm...@we...>> wrote:
>>
>>         Hi, using:
>>
>>         set terminal pngcairo truecolor enhanced font "arial,11"
>>         fontscale 1.0
>>         size 1000,1000
>>         set output "test.png"
>>
>>         set log xy
>>         #set logscale xy
>>
>>         set xrange [100:1e7]
>>         set yrange [100:1e7]
>>
>>         set grid xtics ytics mxtics mytics lt 0 lw 1 lc rgb "#000000"
>>
>>         f(x)  = x/   (1-0.0001*x)
>>         fi(x) = x/(1/(1-0.0001*x))
>>         g(x)  = x/   (1-0.0000001*x)
>>         gi(x) = x/(1/(1-0.0000001*x))
>>         id(x) = x
>>
>>         plot f(x)  lt 2 lc rgb "#ff0000" lw 1 , \
>>               fi(x)  lt 2 lc rgb "#ff0000" lw 1  , \
>>               g(x)  lt 2 lc rgb "#ff0000" lw 1  , \
>>               gi(x)  lt 2 lc rgb "#ff0000" lw 1  , \
>>               id(x) dashtype 2 lc rgb "#000000" lw 1
>>
>>         1) the grid tick lines look either incorrect or at least not
>>         to what
>>         they look up to [100:1e6]. How can I make them work with
>>         [100:1e7] ?
>>         2) the plot of the lines does neither go to the bottom
>>         (fi,gi), nor to
>>         the top (f,g), as expected. How to fix?
>>
>>         ------------------------------------------------------------------------------
>>         _______________________________________________
>>         gnuplot-info mailing list
>>         gnu...@li...
>>         <mailto:gnu...@li...>
>>         Membership management via:
>>         https://lists.sourceforge.net/lists/listinfo/gnuplot-info
>>
>>
>
>

Re: [Gnuplot-info] rotate labels by data column

[theozh <th...@gm...>]

sorry, minor correction:

In the following line it should read of course 'a' instead of 'i' within
cos() and sin() for the position of the label.

set label i center at (5+5*cos(-a/n*360+90)),(5+5*sin(-a/n*360+90))\

Re: [Gnuplot-info] Bugs? 1) Grid ticks in logscale incorrect when range above 1e6+1 and 2) also not plotted completely

[Ethan M. <eam...@gm...>]

There is no bug here - you are fighting the limits of floating point
precision.
Consider the final steps that are sampled as x approaches 10000.
For linear  sampling in steps of 100

f(9800)   is 490000.000000002
f(9900)   is 990000.00000001
f(10000) is inf   (divide by zero at IEEE floating precision)

So f(9900) is the last point that shows on the plot, and that value is less
than the y range you have chosen.  The function reaches ymax (10^7) at
f(9990.)
To catch that on your plot you must sample finely enough to distinguish
9990 from 10000.

And that's for linear sampling. In logscale the sampling must be even finer.
The program selects even samples along the x axis (even spacing on log(x)).
So most of the samples are at low values of x.

On Sun, Jul 31, 2016 at 5:16 AM, D <dcm...@we...> wrote:

> Thanks. I had to 'set sample' to 500000 for gnuplot to plot the whole line
> (slightly different xy range and the line goes to the bottom). Works in
> linear mode without the need for any 'set sample', wonder why the log mode
> requires that at all. Maybe you should fix that for v5.1 too.
>
>
> On 29.07.2016 20:52, Ethan Merritt wrote:
>
> 1)   version 5.0:
>       set mxtics 10                  # see "help set mxtics"
>       set mytics 10
>
>      version 5.1
>       currently a bug.  thanks for the report
>
> 2) set sample 10000
> Or else define the functions as parametric so that the sample is
> independent of the x/y precision.
> But increasing the number of samples is easier.
>
> On Fri, Jul 29, 2016 at 11:28 AM, D <dcm...@we...> wrote:
>
>> Hi, using:
>>
>> set terminal pngcairo truecolor enhanced font "arial,11" fontscale 1.0
>> size 1000,1000
>> set output "test.png"
>>
>> set log xy
>> #set logscale xy
>>
>> set xrange [100:1e7]
>> set yrange [100:1e7]
>>
>> set grid xtics ytics mxtics mytics lt 0 lw 1 lc rgb "#000000"
>>
>> f(x)  = x/   (1-0.0001*x)
>> fi(x) = x/(1/(1-0.0001*x))
>> g(x)  = x/   (1-0.0000001*x)
>> gi(x) = x/(1/(1-0.0000001*x))
>> id(x) = x
>>
>> plot f(x)  lt 2 lc rgb "#ff0000" lw 1 , \
>>       fi(x)  lt 2 lc rgb "#ff0000" lw 1  , \
>>       g(x)  lt 2 lc rgb "#ff0000" lw 1  , \
>>       gi(x)  lt 2 lc rgb "#ff0000" lw 1  , \
>>       id(x) dashtype 2 lc rgb "#000000" lw 1
>>
>> 1) the grid tick lines look either incorrect or at least not to what
>> they look up to [100:1e6]. How can I make them work with [100:1e7] ?
>> 2) the plot of the lines does neither go to the bottom (fi,gi), nor to
>> the top (f,g), as expected. How to fix?
>>
>>
>> ------------------------------------------------------------------------------
>> _______________________________________________
>> gnuplot-info mailing list
>> gnu...@li...
>> Membership management via:
>> https://lists.sourceforge.net/lists/listinfo/gnuplot-info
>>
>
>
>

Re: [Gnuplot-info] rotate labels by data column

[theozh <th...@gm...>]

well, you could do it with setting labels...

However, the example below is a bit a cumbersome workaround, but it
seems to do the job in a wxt terminal.
- the first data column is used to define the position
- the second data column gives the label text
- the third data column defines the rotation
The first 3 plot command are not plotting anything but extracting
numbers and text into the variables a,b,c.

Hope this helps...
Maybe there is a more elegant version?


reset

$Data <<EOD
1 one -30
2 two -60
3 three -90
4 four -120
5 five -150
6 six -180
7 seven -210
8 eight -240
9 nine -270
10 ten -300
11 eleven -330
12 twelve -360
EOD

set xrange [0:10]
set yrange [0:10]
set size square
unset xtics
unset ytics
n = 12.   # . for floating point division
set angles degrees
do for [i=1:n] {
  plot $Data u 1:(a=$1,0/0) every 1:1:i-2:0:i-1:0 notitle;
  plot $Data u 1:(b=stringcolumn(2),0/0) every 1:1:i-2:0:i-1:0 notitle;
  plot $Data u 1:(c=$3,0/0) every 1:1:i-1:0:i-1:0 notitle;
  set label i center at (5+5*cos(-i/n*360+90)),(5+5*sin(-i/n*360+90))\
sprintf("%s",b) rotate by c font "Arial,20";
  }

  plot $Data u 1:1 notitle     # or whatever else you need to plot

[Gnuplot-info] rotate labels by data column

[Grothausmann, R. Dr. <gro...@mh...>]

Dear mailing list members,


Is there any way to plot labels rotated according to angles given by a data column?
All I could find is:
plot ... with labels rotate by X
which, as I understand, only allows a fixed rotation for all labels.

Any help or hints are very much appreciated
Roman

Re: [Gnuplot-info] Bugs? 1) Grid ticks in logscale incorrect when range above 1e6+1 and 2) also not plotted completely

[D <dcm...@we...>]

Thanks. I had to 'set sample' to 500000 for gnuplot to plot the whole 
line (slightly different xy range and the line goes to the bottom). 
Works in linear mode without the need for any 'set sample', wonder why 
the log mode requires that at all. Maybe you should fix that for v5.1 too.

On 29.07.2016 20:52, Ethan Merritt wrote:
> 1)   version 5.0:
>       set mxtics 10                  # see "help set mxtics"
>       set mytics 10
>
>      version 5.1
>       currently a bug.  thanks for the report
>
> 2) set sample 10000
> Or else define the functions as parametric so that the sample is 
> independent of the x/y precision.
> But increasing the number of samples is easier.
>
> On Fri, Jul 29, 2016 at 11:28 AM, D <dcm...@we... 
> <mailto:dcm...@we...>> wrote:
>
>     Hi, using:
>
>     set terminal pngcairo truecolor enhanced font "arial,11" fontscale 1.0
>     size 1000,1000
>     set output "test.png"
>
>     set log xy
>     #set logscale xy
>
>     set xrange [100:1e7]
>     set yrange [100:1e7]
>
>     set grid xtics ytics mxtics mytics lt 0 lw 1 lc rgb "#000000"
>
>     f(x)  = x/   (1-0.0001*x)
>     fi(x) = x/(1/(1-0.0001*x))
>     g(x)  = x/   (1-0.0000001*x)
>     gi(x) = x/(1/(1-0.0000001*x))
>     id(x) = x
>
>     plot f(x)  lt 2 lc rgb "#ff0000" lw 1 , \
>           fi(x)  lt 2 lc rgb "#ff0000" lw 1  , \
>           g(x)  lt 2 lc rgb "#ff0000" lw 1  , \
>           gi(x)  lt 2 lc rgb "#ff0000" lw 1  , \
>           id(x) dashtype 2 lc rgb "#000000" lw 1
>
>     1) the grid tick lines look either incorrect or at least not to what
>     they look up to [100:1e6]. How can I make them work with [100:1e7] ?
>     2) the plot of the lines does neither go to the bottom (fi,gi), nor to
>     the top (f,g), as expected. How to fix?
>
>     ------------------------------------------------------------------------------
>     _______________________________________________
>     gnuplot-info mailing list
>     gnu...@li...
>     <mailto:gnu...@li...>
>     Membership management via:
>     https://lists.sourceforge.net/lists/listinfo/gnuplot-info
>
>

Re: [Gnuplot-info] Weird Gnuplot problem

[Dave H. <da...@ho...>]

On Fri, 29 Jul 2016, Ethan A Merritt wrote:

[ ... ]

> > What I really want to see is the parabola crossing the straight line, 
> > around pi * 16. so where have I stuffed up?
> 
> The default range on x is [-10:10], which is what you have.
> You asked for xtics starting at 0 and going to 100, which you also have.
> IOW you got just what you asked for.
> 
> Maybe you intended "set xrange" rather than "set xtics"?

Bingo!  Many thanks.

-- 
Dave Horsfall DTM (VK2KFU)  "Those who don't understand security will suffer."

Re: [Gnuplot-info] Interactive user input with gnuplot

[Joseph G. <jos...@ya...>]

The idea is to get user input, do some calculations and use the result on a plot, right?If you use R then there are demos here on how to do that:
CRAN - Package Rgnuplot

  
|  
|   |  
CRAN - Package Rgnuplot
   |  |

  |

 
For Python, C, shell scripts, etc there is similar stuff online.
Sometimes the calculations are too complex or the data has to be prepared before the plot, fortunately gnuplot is very easy to be called from any language or environment.
      From: theozh <th...@gm...>
 To: gnu...@li... 
 Sent: Friday, July 29, 2016 4:11 PM
 Subject: [Gnuplot-info] Interactive user input with gnuplot
   
Is there a way in gnuplot to interactively enter a few values, i.e.
numbers and strings?

The idea is to enable users which are not familiar with gnuplot and do
not want to (are not able to or should not) edit a gnuplot script to
generate plots which require a few custom values which are then stored
in user variables for further use in the script.

The command "pause" allows some very limited user input

pause mouse keypress "Type a letter"
# Returning the ASCII value of the key pressed in MOUSE_KEY
or
pause -1 "Hit a key to continue"

But I am looking for a simple popup window, e.g. like:

+------------------+
| Sample: .......  |
| Start:  .......  |
| Stop:  .......  |
|                  |
|  OK    Cancel  |
+------------------+

Any ways to achieve this?


------------------------------------------------------------------------------
_______________________________________________
gnuplot-info mailing list
gnu...@li...
Membership management via: https://lists.sourceforge.net/lists/listinfo/gnuplot-info


  

Re: [Gnuplot-info] Weird Gnuplot problem

[Ethan A M. <EAM...@gm...>]

On Saturday, 30 July 2016 01:50:40 PM Dave Horsfall wrote:
> Gnuplot 5.0.4, on MacOS 10.11.6.
> 
> For reasons that are not particularly germane, I am trying to determine 
> the area of a square vs. a circle, given a piece of string of length "x".  
> I have worked it out in simple maths, but I want to plot it.
> 
> gnuplot> set xtics 0,1,100
> gnuplot> plot (x*x)/16, pi * x
> 
> But no matter what I do, I get a partial graph, no xtics on the left, 
> extending to 10 on the right, with the expected parabola and angled line 
> intersecting at x=0.
> 
> What I really want to see is the parabola crossing the straight line, 
> around pi * 16. so where have I stuffed up?

The default range on x is [-10:10], which is what you have.
You asked for xtics starting at 0 and going to 100, which you also have.
IOW you got just what you asked for.

Maybe you intended "set xrange" rather than "set xtics"?

Re: [Gnuplot-info] Gnuplot's internal sequence

[Ethan A M. <EAM...@gm...>]

On Friday, 29 July 2016 11:07:29 PM Alan Corey wrote:
> On 7/29/16, Ethan Merritt <eam...@gm...> wrote:
> > Use the "stats" command.  If you only need to know the y value range, that
> > would be
> >
> >     stats "outvals.tab" using 2
> 
> Thank you, that's close.  What it gets though are the min/max of the
> data rather than the limits of the plot.  For example 17.6 - 31.9
> rather than 16 - 32.  I imagine there's a relatively simple formula
> that sets the plot limits to round numbers?  I think those are what I
> should put through the * 1.8 + 32 conversion.  I probably can't access
> the plot limits but I can derive them over again from the data limits
> if I know the formula.

Not a formula.  By default the axis range is extended to the next tic mark.
What that is depends on "set ytics".  If you do not want the axis range
to extend beyond the actual data, you can use the keyword "noextend".
I.e.
   set yrange [*:*] noextend
In that case the axis range will exactly match the min/max of the 
data y coordinates,.

See "help noextend"

		Ethan

[Gnuplot-info] Weird Gnuplot problem

[Dave H. <da...@ho...>]

Gnuplot 5.0.4, on MacOS 10.11.6.

For reasons that are not particularly germane, I am trying to determine 
the area of a square vs. a circle, given a piece of string of length "x".  
I have worked it out in simple maths, but I want to plot it.

gnuplot> set xtics 0,1,100
gnuplot> plot (x*x)/16, pi * x

But no matter what I do, I get a partial graph, no xtics on the left, 
extending to 10 on the right, with the expected parabola and angled line 
intersecting at x=0.

What I really want to see is the parabola crossing the straight line, 
around pi * 16. so where have I stuffed up?

Thanks.

-- 
Dave Horsfall DTM (VK2KFU)  "Those who don't understand security will suffer."

Re: [Gnuplot-info] Gnuplot's internal sequence

[Alan C. <ala...@gm...>]

On 7/29/16, Ethan Merritt <eam...@gm...> wrote:
> Use the "stats" command.  If you only need to know the y value range, that
> would be
>
>     stats "outvals.tab" using 2

Thank you, that's close.  What it gets though are the min/max of the
data rather than the limits of the plot.  For example 17.6 - 31.9
rather than 16 - 32.  I imagine there's a relatively simple formula
that sets the plot limits to round numbers?  I think those are what I
should put through the * 1.8 + 32 conversion.  I probably can't access
the plot limits but I can derive them over again from the data limits
if I know the formula.

-- 
Credit is the root of all evil.  - AB1JX

Re: [Gnuplot-info] Gnuplot's internal sequence

[Ethan M. <eam...@gm...>]

Use the "stats" command.  If you only need to know the y value range, that
would be

    stats "outvals.tab" using 2

On Fri, Jul 29, 2016 at 12:59 PM, Alan Corey <ala...@gm...> wrote:

> My data is data points contained in an input file.  What I'd like to
> do is something like:
>
>   set y2range [(1.8*GPVAL_Y_MIN)+32:(1.8*GPVAL_Y_MAX)+32]
>
> But GPVAL_Y_MIN and GPVAL_Y_MAX apparently aren't defined until the
> read is complete, and putting things like y2range or y2ticks after the
> read seems to have no effect.  I tried refresh and replot after the y2
> stuff, even a 2nd plot line.
>
> Is there a way to do a dummy read that just sets up the min/max stuff?
>
> reset
> set terminal gif size 1200,800
> set output "2labc.gif"
> set xdata time
> set timefmt x "%s"
> set format x "%H:%M\n%D"
> set style data lines
> set xtics auto
> #  GPVAL_Y_MIN etc aren't defined yet
> #set y2range [(1.8*GPVAL_Y_MIN)+32:(1.8*GPVAL_Y_MAX)+32]
> set y2label "Fahrenheit"
> plot "outvals.tab" using ($1-(4*3600)):2 notitle
> # These have no effect:
> #set y2range [(1.8*GPVAL_Y_MIN)+32:(1.8*GPVAL_Y_MAX)+32]
> #set y2tics auto
> #refresh
>
>
> --
> Credit is the root of all evil.  - AB1JX
>
>
> ------------------------------------------------------------------------------
> _______________________________________________
> gnuplot-info mailing list
> gnu...@li...
> Membership management via:
> https://lists.sourceforge.net/lists/listinfo/gnuplot-info
>

[Gnuplot-info] Gnuplot's internal sequence

[Alan C. <ala...@gm...>]

My data is data points contained in an input file.  What I'd like to
do is something like:

  set y2range [(1.8*GPVAL_Y_MIN)+32:(1.8*GPVAL_Y_MAX)+32]

But GPVAL_Y_MIN and GPVAL_Y_MAX apparently aren't defined until the
read is complete, and putting things like y2range or y2ticks after the
read seems to have no effect.  I tried refresh and replot after the y2
stuff, even a 2nd plot line.

Is there a way to do a dummy read that just sets up the min/max stuff?

reset
set terminal gif size 1200,800
set output "2labc.gif"
set xdata time
set timefmt x "%s"
set format x "%H:%M\n%D"
set style data lines
set xtics auto
#  GPVAL_Y_MIN etc aren't defined yet
#set y2range [(1.8*GPVAL_Y_MIN)+32:(1.8*GPVAL_Y_MAX)+32]
set y2label "Fahrenheit"
plot "outvals.tab" using ($1-(4*3600)):2 notitle
# These have no effect:
#set y2range [(1.8*GPVAL_Y_MIN)+32:(1.8*GPVAL_Y_MAX)+32]
#set y2tics auto
#refresh


-- 
Credit is the root of all evil.  - AB1JX

Re: [Gnuplot-info] Bugs? 1) Grid ticks in logscale incorrect when range above 1e6+1 and 2) also not plotted completely

[Ethan M. <eam...@gm...>]

1)   version 5.0:
      set mxtics 10                  # see "help set mxtics"
      set mytics 10

     version 5.1
      currently a bug.  thanks for the report

2) set sample 10000
Or else define the functions as parametric so that the sample is
independent of the x/y precision.
But increasing the number of samples is easier.

On Fri, Jul 29, 2016 at 11:28 AM, D <dcm...@we...> wrote:

> Hi, using:
>
> set terminal pngcairo truecolor enhanced font "arial,11" fontscale 1.0
> size 1000,1000
> set output "test.png"
>
> set log xy
> #set logscale xy
>
> set xrange [100:1e7]
> set yrange [100:1e7]
>
> set grid xtics ytics mxtics mytics lt 0 lw 1 lc rgb "#000000"
>
> f(x)  = x/   (1-0.0001*x)
> fi(x) = x/(1/(1-0.0001*x))
> g(x)  = x/   (1-0.0000001*x)
> gi(x) = x/(1/(1-0.0000001*x))
> id(x) = x
>
> plot f(x)  lt 2 lc rgb "#ff0000" lw 1 , \
>       fi(x)  lt 2 lc rgb "#ff0000" lw 1  , \
>       g(x)  lt 2 lc rgb "#ff0000" lw 1  , \
>       gi(x)  lt 2 lc rgb "#ff0000" lw 1  , \
>       id(x) dashtype 2 lc rgb "#000000" lw 1
>
> 1) the grid tick lines look either incorrect or at least not to what
> they look up to [100:1e6]. How can I make them work with [100:1e7] ?
> 2) the plot of the lines does neither go to the bottom (fi,gi), nor to
> the top (f,g), as expected. How to fix?
>
>
> ------------------------------------------------------------------------------
> _______________________________________________
> gnuplot-info mailing list
> gnu...@li...
> Membership management via:
> https://lists.sourceforge.net/lists/listinfo/gnuplot-info
>

[Gnuplot-info] Bugs? 1) Grid ticks in logscale incorrect when range above 1e6+1 and 2) also not plotted completely

[D <dcm...@we...>]

Hi, using:

set terminal pngcairo truecolor enhanced font "arial,11" fontscale 1.0 
size 1000,1000
set output "test.png"

set log xy
#set logscale xy

set xrange [100:1e7]
set yrange [100:1e7]

set grid xtics ytics mxtics mytics lt 0 lw 1 lc rgb "#000000"

f(x)  = x/   (1-0.0001*x)
fi(x) = x/(1/(1-0.0001*x))
g(x)  = x/   (1-0.0000001*x)
gi(x) = x/(1/(1-0.0000001*x))
id(x) = x

plot f(x)  lt 2 lc rgb "#ff0000" lw 1 , \
      fi(x)  lt 2 lc rgb "#ff0000" lw 1  , \
      g(x)  lt 2 lc rgb "#ff0000" lw 1  , \
      gi(x)  lt 2 lc rgb "#ff0000" lw 1  , \
      id(x) dashtype 2 lc rgb "#000000" lw 1

1) the grid tick lines look either incorrect or at least not to what 
they look up to [100:1e6]. How can I make them work with [100:1e7] ?
2) the plot of the lines does neither go to the bottom (fi,gi), nor to 
the top (f,g), as expected. How to fix?

[Gnuplot-info] Documentation [was Re: Multiple sources on a common X axis]

[zowie <def...@bo...>]

Documentation in general is hard, and presenting information in a way that end-users
can read is a skill that is quite different from presenting it in a way that 
works for experts.

My favorite example of the pitfalls was the three-line instruction manual that
used to appear on emacs’ front page by default, back in the 1990s.  It read 
something like this:

   The expression “C-“ means to hold down control while pressing the key. 
   To load a file, type “C-X F”.  For a self-help tutorial, type “C-H T”.  
   To quit, type “C-X C-C”.

I encountered that text as a graduate student in the Physics Dept. at 
Stanford University in 1990.  For the next few years, I would tell new 
students in my research group to “just fire up emacs, and follow the 
directions to the tutorial”.  Out of about 8-10 people to whom I told that 
(remember, these were pretty smart people) zero of them actually found
the self-help tutorial the first try, without being told to read EVERY 
WORD of what was on the screen.

It turns out that it’s very hard for people to find text in the middle of other
text — we’re not natural greppers.

The gnuplot documentation suffers from the same sort of fallacy, in spades.
As an expert I still have trouble finding items in it.  For example, I recently
had trouble with single vs. double quotes in our Perl front-end module
(PDL::Graphics::Gnuplot), because despite multiple reads it was not obvious to me
from the documentation that those two types of quote are interchangeable and 
behave slightly differently within gnuplot.

That’s not to chastise the team — Ethan and Hans-Bernhard and everyone else
have created (and continue to maintain) something phenomenally useful and 
flexible.  Just to acknowledge that many of the support issues on this list
(and tons of “invisible” ones that never get mailed here) reflect a need
to improve the documentation, rather than a failure of the user base
or holes in the code itself.

Cheers,
Craig


> On Jul 29, 2016, at 7:30 AM, Alan Corey <ala...@gm...> wrote:
> 
>> It took me quite some time to figure out a few things. This might keep
>> some people away from using gnuplot.
> 
> Who takes time to read documentation?  I mean, I have, but my best
> resource for looking up how to do something is to look at how I did it
> a year ago.  That's true not only for Gnuplot.
> 
> One thing that might help is to somewhere on a web page have tiny
> (320x200 or so) thumbnails of different plot types that are linked to
> maybe example plot scripts and/or into the documentation where it
> talks about using some feature.  I thought Gnuplot had a "concept
> index", something I use does, in addition to a normal index.
> 
> But that's part of my point, if you use a few dozen different programs
> and each one has hundreds of pages of documentation there isn't time
> in a lifetime to read it all.  It's good that the documentation
> exists, but having maybe a few different ways of finding what you're
> looking for in it is good too.  And this seems to be more of a problem
> in things written by committees or multiple people or at different
> times.  Lazarus for example is potentially a great program with
> thousands of pages of documentation spanning a couple decades but you
> can't find anything in it.
> 
> So I look up how I did it last time.  I use the PDFs and online
> documentation too.
> 
> ------------------------------------------------------------------------------
> _______________________________________________
> gnuplot-info mailing list
> gnu...@li...
> Membership management via: https://lists.sourceforge.net/lists/listinfo/gnuplot-info
> 

Re: [Gnuplot-info] Multiple sources on a common X axis

[sfeam <sf...@us...>]

On Friday, 29 July 2016 09:30:30 AM Alan Corey wrote:
> > It took me quite some time to figure out a few things. This might keep
> > some people away from using gnuplot.
> 
> Who takes time to read documentation?  I mean, I have, but my best
> resource for looking up how to do something is to look at how I did it
> a year ago.  That's true not only for Gnuplot.
> 
> One thing that might help is to somewhere on a web page have tiny
> (320x200 or so) thumbnails of different plot types that are linked to
> maybe example plot scripts and/or into the documentation where it
> talks about using some feature.  I thought Gnuplot had a "concept
> index", something I use does, in addition to a normal index.

      http://gnuplot.sourceforge.net/demo/


> 
> But that's part of my point, if you use a few dozen different programs
> and each one has hundreds of pages of documentation there isn't time
> in a lifetime to read it all.  It's good that the documentation
> exists, but having maybe a few different ways of finding what you're
> looking for in it is good too.  And this seems to be more of a problem
> in things written by committees or multiple people or at different
> times.  Lazarus for example is potentially a great program with
> thousands of pages of documentation spanning a couple decades but you
> can't find anything in it.
> 
> So I look up how I did it last time.  I use the PDFs and online
> documentation too.
> 
> ------------------------------------------------------------------------------
> _______________________________________________
> gnuplot-info mailing list
> gnu...@li...
> Membership management via: https://lists.sourceforge.net/lists/listinfo/gnuplot-info

Re: [Gnuplot-info] Multiple sources on a common X axis

[Alan C. <ala...@gm...>]

> It took me quite some time to figure out a few things. This might keep
> some people away from using gnuplot.

Who takes time to read documentation?  I mean, I have, but my best
resource for looking up how to do something is to look at how I did it
a year ago.  That's true not only for Gnuplot.

One thing that might help is to somewhere on a web page have tiny
(320x200 or so) thumbnails of different plot types that are linked to
maybe example plot scripts and/or into the documentation where it
talks about using some feature.  I thought Gnuplot had a "concept
index", something I use does, in addition to a normal index.

But that's part of my point, if you use a few dozen different programs
and each one has hundreds of pages of documentation there isn't time
in a lifetime to read it all.  It's good that the documentation
exists, but having maybe a few different ways of finding what you're
looking for in it is good too.  And this seems to be more of a problem
in things written by committees or multiple people or at different
times.  Lazarus for example is potentially a great program with
thousands of pages of documentation spanning a couple decades but you
can't find anything in it.

So I look up how I did it last time.  I use the PDFs and online
documentation too.

[Gnuplot-info] Interactive user input with gnuplot

[theozh <th...@gm...>]

Is there a way in gnuplot to interactively enter a few values, i.e.
numbers and strings?

The idea is to enable users which are not familiar with gnuplot and do
not want to (are not able to or should not) edit a gnuplot script to
generate plots which require a few custom values which are then stored
in user variables for further use in the script.

The command "pause" allows some very limited user input

pause mouse keypress "Type a letter"
# Returning the ASCII value of the key pressed in MOUSE_KEY
or
pause -1 "Hit a key to continue"

But I am looking for a simple popup window, e.g. like:

+------------------+
| Sample: .......  |
| Start:  .......  |
| Stop:   .......  |
|                  |
|   OK    Cancel   |
+------------------+

Any ways to achieve this?

Re: [Gnuplot-info] Multiple sources on a common X axis

[theozh <th...@gm...>]

well, certainly almost all the information you need is somewhere in
documentation, manual and demos.

However, as a beginner, if you stumble across the command "multiplot"
you may believe that this could do the job, although for most cases
"plot" is sufficient.


When I started with gnuplot I was missing comprehensive tutorials (let
me know if I missed one).
A tutorial is not the same as a manual.
Although, there are a lot of introductory tutorials around, most of them
were too basic and soon not sufficient or satisfying anymore. If you
have reached a certain level, you mostly get along with the manual.

More explicit, I missed e.g.
- an overview plot illustrating and explaining the nomenclature of the
axis, labels, tics, coordinates, margins; as well as names and concepts
e.g. for frame, border, margin, canvas, etc. If you do not now how an
object, setting or parameter is named, it is difficult to search for it.
- many tips and tricks... and "how to's"... (which of course is the
intention of the demos).
- updated tips and tricks based on the questions which are asked and
answered e.g. in this newsgroup
- known limits and difficulties what can be done with gnuplot and what
not (and workarounds if there is one).

It took me quite some time to figure out a few things. This might keep
some people away from using gnuplot.

I am wondering whether it is generally possible to write a tutorial
which makes it unnecessary to search the web or to ask (FA)Qs in forums
or newsgroups. Assuming of course that people are reading and following
the tutorial ;-).
This might be a task to linguists to funnel and direct all possible
terms and questions to the right information and instruction in the
tutorial/manual.

I can imagine that is a lot of work to compile the manual. Thanks to all
who have been involved! A comprehensive and understandable step by step
tutorial might be a project on its own.

Re: [Gnuplot-info] Multiple sources on a common X axis

[Dave H. <da...@ho...>]

On Thu, 28 Jul 2016, Hans-Bernhard Bröker wrote:

> I keep wondering how so many people arrive at the belief that they 
> should be using multiplot for so many jobs it's the completely wrong 
> tool for.  Is the concept of
> 
> 	plot 'file1', 'file2'
> 
> really that hard to pick up from the documentation and demos?

Could it be that there is something wrong with the documentation?

-- 
Dave Horsfall DTM (VK2KFU)  "Those who don't understand security will suffer."