gnuplot-info — For questions and discussion about gnuplot

Re: [Gnuplot-info] Print huge data into a single plot

[Matthias_D <d00...@nt...>]

Ok, I could plot everything now.

I put all the relevant data into one file before plotting.
Don't know why I couldn't plot many files into one plot.
I used "load", not "call" for the script file input.
I changed the variables of the file before each run.
-- 
View this message in context: http://old.nabble.com/Print-huge-data-into-a-single-plot-tp32578285p32594333.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] question on "parametric" data style

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

Thanks both Bernhard and Thomas,
this did exactly what I wanted to achieve - now without an additional file. 

And, yes, I had had been playing with set logscale t without success, as 
Bernhard readily pointed out.

Sometimes one just doesn't see the forest due to all those trees around ...
Take care



Dieter Jurzitza

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

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

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

[Gnuplot-info] How to fill a triangle with a gradient?

[ju4nk4 <ju...@gm...>]

Hej, my name is Juan Carlos Araujo from Uppsala University in Sweden. I study
a MSc in Scientific Computing here in Uppsala and currently I am trying to
parse all my numerics to c++ while developing an interface with c++ to
connect with Gnuplot. My main topic of research is Unstructured Grid methods
for solving PDEs. Some of them based on triangles and others in
quadrilaterals.

I find GNUPLOT a great tool for my research but I would appreciate som
insight on the problems below. If possible it would be great to have the
shell code.

1. For an arbitrary triangle with nodes (x1,y1), (x2,y2) and (x3,y3) and
colors c1, c2 and c3 respectively fill the triangle with the interpolated
gradient as shown in the figure, by using GNUPLOT.

2. Defined a function f(x,y) describing the color field for a point (x,y),
fill a triangle with such gradient.

check the attached figure, or see the problem here
http://juancarlosaraujo.sytes.net/Contenido/triangle.htm
http://old.nabble.com/file/p32586035/VertexColors.png 
-- 
View this message in context: http://old.nabble.com/How-to-fill-a-triangle-with-a-gradient--tp32586035p32586035.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] Can't read data file

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

The original message never made it to the list (or if it did, my system ate
it as spam).

Gnuplot will read data files from the path you specify, or its current
working directory if not specified, just like any other program. If you
haven't configured the shortcut or gnuplot.ini differently, on Windows, the
CWD will be the directory where the binary lives. You can change this by
creating a shortcut, and altering the starting directory under properties.

Or you can simply specify the entire path to your data file(s), e.g.
	plot "d:/software/path with spaces/file.dat"

Or you can change gnuplot's CWD before plotting, with the cd command,
e.g.:
	cd "d:/software/data"
	plot "file.dat"

Note that if you're using backslashes in your path names, you need to
either use non-interpolating single quotes, like 'c:\temp\path with
spaces\data.dat' or you need to escape the backslashes, as otherwise
interpolation will treat them as an escape sequence, like "c:\\temp\\path
to\\data.dat". For an interpolating, double-quoted string like "c:\temp"
the "\t" will be converted to a tab character.



Matthias_D <d00...@nt...> said (on 2011/10/02):
> I am also quite new to GNUplot, but I put the datafiles into the same
> directory
> as the program gnuplot.exe. Otherwise you have to set the path of the file
> (I don't know how).
> If you set the path, it maybe should not contain spaces, this is a linux
> manner.
> 
> In my case the path is:    D:\Software\gp443win32\gnuplot\binary\
> 
> Put the files here, than it should work.
> 
> Lboro B wrote:
> > Hi,
> > 
> > I have just installed gnuplot 4.0, running on Windows 7. I have been
> > trying some of the tutorials for plotting from a .dat file but I
> > constantly get the following error: can't read data file "filename.dat"
> > 
> > I have no idea what I am doing wrong here. I create the .dat file in
> > notepad and have it saved on my desktop. Is there a specific folder I
> > should save the .dat file in?
> > 
> > Hope someone can help with this.
> > 
> > Thanks

Re: [Gnuplot-info] Use variable from datafile in plot

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

without knowing how your data files look like, here's a simple
example:

datafile:

1 1 a
2 2 b
3 3 c
4 4 d

we want to use the "a" in the 1st line as a key title,
so we plot nothing to get the "a", store in "var1", and
then plot the data:

plot "datafile" using 1:(var1=stringcolumn(3), NaN) every ::0::0 notitle, \
"" using 1:2 with lines title var1


ftpronk wrote:
> 
> Dear All,
> 
> I would like to use a variable from a data file as title for a plot. 
> Something like:
> 
> var1 = './data_2' index 1000 using 4
> plot './data_1' index 1000 using 1:2 title var1
> 
> Well, obviously, the example I just gave doesnt work.  But I would like to
> know whether it is possible to do something similar in gnuplot?  Maybe
> using a syntax similar to:
> plot [...]  title sprintf("%d",  myVar)  ?
> 
> Thank you for your help!
> 
> Looking forward to your reply,
> 
> Francois. 
> 

-- 
View this message in context: http://old.nabble.com/Use-variable-from-datafile-in-plot-tp32577747p32581554.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] question on

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

Dr. Ing. Dieter Jurzitza <dieter.jurzitza <at> t-online.de> writes:

> ...
> I have a function (polar diagram) covering many decades of values. I would 
> like to have a constant number of values per decade rather than one with 
> constant spacing (configured via "set samples VALUE")
> ...
> set parametric
> set trange [0.01:1000000]
> plot RE(t),IM(t)
> ...

you may transfer a linear trange into an exponential one via '10**t':

set trange [-2:6]
set samples 81
plot RE(10**t),IM(10**t)

Re: [Gnuplot-info] question on "parametric" data style

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

On 03.10.2011 10:42, Dr. Ing. Dieter Jurzitza wrote:

> I have a function (polar diagram) covering many decades of values. I would
> like to have a constant number of values per decade rather than one with
> constant spacing (configured via "set samples VALUE")

So what you're trying to do is, essentially, a log-scaled 't' axis. 
Unfortunately, gnuplot doesn't do that by itself.  It accepts the 
command "set log t", but doesn't do anything with it.

But read on.

> I. e. I have two functions RE(x) and IM(x), and rather than performing a
>
> set parametric
> set trange [0.01:1000000]
> plot RE(t),IM(t)

Just transform 't' as you go:

set trange [-2, 6]
plot RE(10.0**t), IM(10.0**t)

Re: [Gnuplot-info] [Non linear regression] How to store the standard uncertainty of a parameter in a variable ?

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

David Kremer <david.kremer.dk <at> gmail.com> writes:

> ...
> When you do a fit, if everything goes fine, you should end up with some
> lines like :
> ...
> 
>  Final set of parameters            Asymptotic Standard Error
>  =======================            ==========================
> 
>  BG              = 0.141662         +/- 0.002977     (2.101%)
>
> My question would be, how to catch the number the asymptotic standard
> error (that is to say the number 0.002977) in a var, to be able then to
> print this number out.

"set fit errorvariables"

then do the fit and look for "<variablename>_err" in

"show variables all"

(here: BG_err)

and print the variable.

[Gnuplot-info] question on "parametric" data style

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

Dear listmembers,
I have a function (polar diagram) covering many decades of values. I would 
like to have a constant number of values per decade rather than one with 
constant spacing (configured via "set samples VALUE")

I helped myself by creating a file "sourcefile" containing a correspondingly 
configured set of values and using those for my plot.

I. e. I have two functions RE(x) and IM(x), and rather than performing a

set parametric
set trange [0.01:1000000]
plot RE(t),IM(t)

I did a

plot "sourcefile" using RE($1):IM($1)

where "sourcefile" contains a set of values like

0.01,0.02 ...0.1,0.2, say 10 per decade.

This works fine for me but seems somewhat much effort. Is there a more 
"intelligent" way to tell gnuplot to use I. e. a constant number of values per 
decade in parametric mode? Otherwise I have to increase the number of samples 
(nearly ..) beyond all bounds what increases the computation time drastically.

I could not find a corresponding hint neither in "help set samples" nor in 
"help set parametric"

Thank you very much for your efforts,
take care




Dieter Jurzitza

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

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

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

Re: [Gnuplot-info] Print huge data into a single plot

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

this 'additional stuff' can be removed by

set key off

the error messages

" ... line xx: ';' expected"

are due to wrong syntax. it should be

" ... with points pt 1 ps 1"

note the space between "pt" or "ps" and the number.

to get black points you need to specify a linetype or
line color in addition:

" ... with points pt 1 ps 1 lt 0"
" ... with points pt 1 ps 1 lc rgb 'black'"

are you using your plot commands file via "load" or via "call"?


Matthias_D wrote:
> 
> I can plot every file now, but only one file per plot!
> 
> Before it worked for two or three files, not more.
> Is it because I use Windows? I don't understand this behaviour of gnuplot.
> 
> I didn't know how to set a path for the files, so I put all the files into
> the
> directory with gnuplot.exe. But this shouldn't be the problem.
> 
> "1:(2*$2)" doesn't work with me, too. But I already changed the values
> in the datafiles, so it's ok for me to use "1:2", "1:3".
> _____________________________________________________
> I finally found a way to plot it.
> I copied all the columns into one file, then plot:
> plot for [i=2:60:2] "001-030.txt" using 1:i with lines,\
> for [i=3:61:2] "001-030.txt" using 1:i with points pt1 ps1
> 
> Looks like:
> 
>  http://old.nabble.com/file/p32580806/Output2.jpg 
> 
> Now only two problems left:
> The point marker should not change but stay at "+" (black).
> I don't want these additional stuff in the upper right corner.
> 
> 
> Thomas Sefzick wrote:
>> 
>> there should be no problem to use many datafiles and
>> assign an individual y-value to each data set, e.g.
>> 
>> plot "datafile_1" using 1:2 with lines,\
>> "" using 1:3 with points pt1 ps1,\
>> "datafile_2" using 1:(2*$2) with lines,\
>> "" using 1:(2*$3) with points pt1 ps1,\
>> "datafile_3" using 1:(3*$2) with lines,\
>> "" using 1:(3*$3) with points pt1 ps1
>> 
>> 
> 

-- 
View this message in context: http://old.nabble.com/Print-huge-data-into-a-single-plot-tp32578285p32581304.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

[Gnuplot-info] [Non linear regression] How to store the standard uncertainty of a parameter in a variable ?

[David K. <dav...@gm...>]

Hello,

I often use GNUplot to make regression using the "fit" function because
I find it useful and somehow easy to use. Since you define the
parameters as variables before to launch the fit, you can then print
out the values of the parameters you obtained after the fit.

My question is : how can you do the same with uncertainty errors ?

When you do a fit, if everything goes fine, you should end up with some
lines like :

 After 3 iterations the fit converged.
 final sum of squares of residuals : 7.15422
 rel. change during last iteration : -2.85539e-15

 degrees of freedom    (FIT_NDF)                        : 898
 rms of residuals      (FIT_STDFIT) = sqrt(WSSR/ndf)    : 0.0892571
 variance of residuals (reduced chisquare) = WSSR/ndf   : 0.00796684
 
 Final set of parameters            Asymptotic Standard Error
 =======================            ==========================
 
 BG              = 0.141662         +/- 0.002977     (2.101%)

My question would be, how to catch the number the asymptotic standard
error (that is to say the number 0.002977) in a var, to be able then to
print this number out.

Thank you very much.

David

Re: [Gnuplot-info] Print huge data into a single plot

[Matthias_D <d00...@nt...>]

I can plot every file now, but only one file per plot!

Before it worked for two or three files, not more.
Is it because I use Windows? I don't understand this behaviour of gnuplot.

I didn't know how to set a path for the files, so I put all the files into
the
directory with gnuplot.exe. But this shouldn't be the problem.

"1:(2*$2)" doesn't work with me, too. But I already changed the values
in the datafiles, so it's ok for me to use "1:2".


Thomas Sefzick wrote:
> 
> there should be no problem to use many datafiles and
> assign an individual y-value to each data set, e.g.
> 
> plot "datafile_1" using 1:2 with lines,\
> "" using 1:3 with points pt1 ps1,\
> "datafile_2" using 1:(2*$2) with lines,\
> "" using 1:(2*$3) with points pt1 ps1,\
> "datafile_3" using 1:(3*$2) with lines,\
> "" using 1:(3*$3) with points pt1 ps1
> 
> 

-- 
View this message in context: http://old.nabble.com/Print-huge-data-into-a-single-plot-tp32578285p32580806.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] Print huge data into a single plot

[Matthias_D <d00...@nt...>]

I inserted " ? " instead of "[]" in the files.
But there is still no usable output. 
My plotfile looks like this:

set datafile missing "?"
set   autoscale                        # scale axes automatically
unset log                              # remove any log-scaling
unset label                            # remove any previous labels
unset title		   # remove any previous titles
set ytics
(1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30)
set xtics (0,2000,4000,6000,8000,10000,12000,14000,16000,17544)
set title "Monitoring Data for stations 1 to 30"
set xlabel "time [hours]"
set ylabel "Stationnumber"
set xr [-300:18450]
set yr [-1.5:31.5]          #I want to print 30 stations on one page
plot "1.txt" using 1:2 with lines,\
"" using 1:3 with points pt1 ps1,\
"2.txt" using 1:(2*$2) with lines,\
"" using 1:(2*$3) with points pt1 ps1,\
"3.txt" using 1:(3*$2) with lines,\
"" using 1:(3*$3) with points pt1 ps1

But I still get the message "plot_final.p", line xx: ';' expected   (xx) is
the last line!
I also don't like the data in the upper right corner.

Output like this:
http://old.nabble.com/file/p32580666/Output1.jpg 




Thomas Sefzick wrote:
> 
> it doesn't seem to be a good idea to use brackets (=2 characters) for
> missing data, better would be to specify something like
> 
> set datafile missing "?"
> 
> and use a question mark for missing data.
> 
> there should be no problem to use many datafiles and
> assign an individual y-value to each data set, e.g.
> 
> plot "datafile_1" using 1:2 with lines,\
> "" using 1:3 with points pt1 ps1,\
> "datafile_2" using 1:(2*$2) with lines,\
> "" using 1:(2*$3) with points pt1 ps1,\
> "datafile_3" using 1:(3*$2) with lines,\
> "" using 1:(3*$3) with points pt1 ps1
> 
> please explain what you mean by 'output becomes miserable'?
> 
> 
> Matthias_D wrote:
>> 
>> Hi,
>> I'm quite new to use gnuplot. Very complicated, I think. But hope I can
>> learn it.
>> 
>> My problem (I found a complicated sollution already, but maybe there is a
>> simple one):
>> 
>> There are many temperatur monitoring stations. The time of their activity
>> shall be plotted
>> as a horizontal line (x axis is the time [hours]) as well as the hours of
>> their malfunktion 
>> (set a "+" on this line).
>> The x axis is the time [hours] from 1 to 17544 (for 2 years).
>> 
>> Now I arranged the data in one file for each station. Looks like this:
>> 
>> First column is the time (hours), second column shows the beginning and
>> end, third column shows the errors.
>> 
>> 567 1 []
>> 678 [] 1
>> 679 [] 1
>> ....
>> 17543 [] 1
>> 17544 1 []
>> 
>> I inserted the brackets to tell gnuplot that here is no value to plot.
>> Actually the data is logical. For various stations I choose to replace
>> the "1" with "2", "3", .....
>> Then the y axis will be 1, 2, 3  and so on and the data will be plotted
>> horizontally for each station.
>> This works fine with one file. 
>> 
>> plot "001.txt" using 1:2 w lines,\
>>      "001.txt" using 1:3 w points pt1 ps1,\
>>      "002.txt" u 1:2 w lines,\
>>      "002.txt" u 1:3 w points pt1 ps1
>> 
>> But when I want to plot more files the output becomes miserable.
>> 
>> I expected to add as many files as I want, but this didn't work.
>> 
>> I also tried to put all the data into a single file with many columns,
>> but it doesn't get better.
>> 
>> Often comes the message "  ';' expected. "
>> 
>> Thanks for every hint! 
>> 
> 
> 

-- 
View this message in context: http://old.nabble.com/Print-huge-data-into-a-single-plot-tp32578285p32580666.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] Print huge data into a single plot

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

it doesn't seem to be a good idea to use brackets (=2 characters) for
missing data, better would be to specify something like

set datafile missing "?"

and use a question mark for missing data.

there should be no problem to use many datafiles and
assign an individual y-value to each data set, e.g.

plot "datafile_1" using 1:2 with lines,\
"" using 1:3 with points pt1 ps1,\
"datafile_2" using 1:(2*$2) with lines,\
"" using 1:(2*$3) with points pt1 ps1,\
"datafile_3" using 1:(3*$2) with lines,\
"" using 1:(3*$3) with points pt1 ps1

please explain what you mean by 'output becomes miserable'?


Matthias_D wrote:
> 
> Hi,
> I'm quite new to use gnuplot. Very complicated, I think. But hope I can
> learn it.
> 
> My problem (I found a complicated sollution already, but maybe there is a
> simple one):
> 
> There are many temperatur monitoring stations. The time of their activity
> shall be plotted
> as a horizontal line (x axis is the time [hours]) as well as the hours of
> their malfunktion 
> (set a "+" on this line).
> The x axis is the time [hours] from 1 to 17544 (for 2 years).
> 
> Now I arranged the data in one file for each station. Looks like this:
> 
> First column is the time (hours), second column shows the beginning and
> end, third column shows the errors.
> 
> 567 1 []
> 678 [] 1
> 679 [] 1
> ....
> 17543 [] 1
> 17544 1 []
> 
> I inserted the brackets to tell gnuplot that here is no value to plot.
> Actually the data is logical. For various stations I choose to replace the
> "1" with "2", "3", .....
> Then the y axis will be 1, 2, 3  and so on and the data will be plotted
> horizontally for each station.
> This works fine with one file. 
> 
> plot "001.txt" using 1:2 w lines,\
>      "001.txt" using 1:3 w points pt1 ps1,\
>      "002.txt" u 1:2 w lines,\
>      "002.txt" u 1:3 w points pt1 ps1
> 
> But when I want to plot more files the output becomes miserable.
> 
> I expected to add as many files as I want, but this didn't work.
> 
> I also tried to put all the data into a single file with many columns, but
> it doesn't get better.
> 
> Often comes the message "  ';' expected. "
> 
> Thanks for every hint! 
> 

-- 
View this message in context: http://old.nabble.com/Print-huge-data-into-a-single-plot-tp32578285p32579103.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] Can't read data file

[Matthias_D <d00...@nt...>]

I am also quite new to GNUplot, but I put the datafiles into the same
directory
as the program gnuplot.exe. Otherwise you have to set the path of the file
(I don't know how).
If you set the path, it maybe should not contain spaces, this is a linux
manner.

In my case the path is:    D:\Software\gp443win32\gnuplot\binary\

Put the files here, than it should work.




Lboro B wrote:
> 
> Hi,
> 
> I have just installed gnuplot 4.0, running on Windows 7. I have been
> trying some of the tutorials for plotting from a .dat file but I
> constantly get the following error: can't read data file "filename.dat"
> 
> I have no idea what I am doing wrong here. I create the .dat file in
> notepad and have it saved on my desktop. Is there a specific folder I
> should save the .dat file in?
> 
> Hope someone can help with this.
> 
> Thanks
> 

-- 
View this message in context: http://old.nabble.com/Can%27t-read-data-file-tp32503824p32578304.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

[Gnuplot-info] Print huge data into a single plot

[Matthias_D <d00...@nt...>]

Hi,
I'm quite new to use gnuplot. Very complicated, I think. But hope I can
learn it.

My problem (I found a complicated sollution already, but maybe there is a
simple one):

There are many temperatur monitoring stations. The time of their activity
shall be plotted
as a horizontal line (x axis is the time [hours]) as well as the hours of
their malfunktion 
(set a "+" on this line).
The x axis is the time [hours] from 1 to 17544 (for 2 years).

Now I arranged the data in one file for each station. Looks like this:

First column is the time (hours), second column shows the beginning and end,
third column shows the errors.

567 1 []
678 [] 1
679 [] 1
....
17543 [] 1
17544 1 []

I inserted the brackets to tell gnuplot that here is no value to plot.
Actually the data is logical. For various stations I choose to replace the
"1" with "2", "3", .....
Then the y axis will be 1, 2, 3  and so on and the data will be plotted
horizontally for each station.
This works fine with one file. 

plot "001.txt" using 1:2 w lines,\
     "001.txt" using 1:3 w points pt1 ps1,\
     "002.txt" u 1:2 w lines,\
     "002.txt" u 1:3 w points pt1 ps1

But when I want to plot more files the output becomes miserable.

I expected to add as many files as I want, but this didn't work.

I also tried to put all the data into a single file with many columns, but
it doesn't get better.

Often comes the message "  ';' expected. "

Thanks for every hint! 
-- 
View this message in context: http://old.nabble.com/Print-huge-data-into-a-single-plot-tp32578285p32578285.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] Double Y data

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

Gabor Urban <urbangabo <at> gmail.com> writes:

> I have a problem for which I could not find solution. I am to make
> plots from data file containing emasurment - calculated data in form:

> x, y1, y2

> where y1 and y2 values belong to the same x value.

> I want to have all point on the very same plot but multiplot is not
> what I require. Is there an idea how can I do it?

plot 'datafilename' using 1:2, '' using 1:3

[Gnuplot-info] Double Y data

[Gabor U. <urb...@gm...>]

Hi,

I have a problem for which I could not find solution. I am to make
plots from data file containing emasurment - calculated data in form:

x, y1, y2

where y1 and y2 values belong to the same x value.

I want to have all point on the very same plot but multiplot is not
what I require. Is there an idea how can I do it?

Gabor

-- 
Linux is like a wigwam: no Gates, no Windows and an Apache inside.

Re: [Gnuplot-info] Can anyone recommend a linux shared web host provider that runs gnuplot ?

[Erich H. <eh...@mi...>]

On Thu, Sep 29, 2011 at 9:22 AM, David Lxxxx <2d...@gm...> wrote:

> Hi.  I am new to this list - but I wonder if anyone can help.
>
> I am looking for a commercial web host firm, which rents out web space on a
> shared linux server, that has gnuplot installed.
> ...


Is it absolutely necessary that it have it installed or can you just load it
yourself?  I know you can run applications you upload yourself on GoDaddy,
though you have to be careful to compile everything against the correct
libraries.

Erich Hoover
eh...@mi...

[Gnuplot-info] Can anyone recommend a linux shared web host provider that runs gnuplot ?

[David L. <2d...@gm...>]

Hi.  I am new to this list - but I wonder if anyone can help.

I am looking for a commercial web host firm, which rents out web space on a
shared linux server, that has gnuplot installed.

There are loads of firms on the web renting out linux web space, but when I
contact them to enquire if they have gnuplot installed, the result almost
invariably comes back 'no'.

If anyone is using web space rented from a commercial firm that provides
access to gnuplot - I would be grateful were you to post the firm's URL to
this list.

Thanks for any help that people may be able to offer.

Best regards,


David ('Ed')

[Gnuplot-info] Fwd: Re: Fwd: Missing line segments using hidden3d

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

[forgot to CC the list...]
On 28.09.2011 20:01, Christian Schroeder wrote:

> I'm curious if there's any way to "improve" on areas of splots that
> seem to have "missing" line segments. For example, in the plot
[...]

> around (-15,0) and (5,-15) (approximately) the plot has these little
> white patches where I would have expected there to be black.

The problem there is not with the plot; it's with the expectation.

You should not expect that kind of line-drawing to _fill_ any area with
colour.  You will get Moiré patterns that way, and those _will_
sometimes have holes.

About the best way of getting rid of that effect is to reduce the
isosamples setting enough for individual lines to become visible again.

If you wanted filled areas, I trust you know where to find the pm3d
plotting style.

Re: [Gnuplot-info] Fewer xticks in date mode

[Biloti <bi...@gm...>]

Thanks Thomas. It worked perfectly.


Thomas Sefzick wrote:
> 
> set the xtics to appear e.g. every 2 years:
> 
> seconds_per_year=365*24*60*60
> set xtics 2*seconds_per_year
> 
> 

-- 
View this message in context: http://old.nabble.com/Fewer-xticks-in-date-mode-tp32538776p32549610.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

[Gnuplot-info] Fwd: Missing line segments using hidden3d

[Christian S. <csc...@uo...>]

Hi,

Thanks for reading my inquiry.

I'm curious if there's any way to "improve" on areas of splots that seem to have "missing" line segments. For example, in the plot 

set terminal pdfcairo
set output "gaussian.pdf"
set key off
set hidden3d 
set isosamples 101
set style line 1 linetype 1 linecolor rgb "black" lw 1
set xrange [-25:25]
set xlabel 'x' 
set yrange [-25:25]
set ylabel 'y' 
f(x,y) = exp(-k*(x**2+y**2))
k=0.01
splot f(x,y) with lines ls 1

around (-15,0) and (5,-15) (approximately) the plot has these little white patches where I would have expected there to be black.

Similarly, with

set terminal pdfcairo
set output "osc.pdf"
set key off
set hidden3d 
set isosamples 101
set style line 1 linetype 1 linecolor rgb "black" lw 1
set xrange [-25:25]
set xlabel 'x'
set ylabel 'y'
set yrange [-25:25]
f(x,y) = ((x**2+y**2)!=0)? 2*pi*a*besj1(sqrt(x**2+y**2)*a)/sqrt(x**2+y**2) : pi*a
a=1
splot f(x,y) with lines ls 1

around (-10,-10) (again, approximately) there is a similar patch of missing line segments.

Is there any way I can improve on this or eliminate this entirely?

Thanks for your help,

Christian

Re: [Gnuplot-info] Fewer xticks in date mode

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

set the xtics to appear e.g. every 2 years:

seconds_per_year=365*24*60*60
set xtics 2*seconds_per_year


Biloti wrote:
> 
> The graph bellow was produced by the following gnuplot command. However,
> the x tick labels are too dense. How could I tell to gnuplot to use fewer
> ticks in this "date mode" for x axis?
> 
> set xlabel "Year"
> set key off
> set timefmt "%Y-%m-%d"
> set format x "%Y"
> set xdata time
> set mxtics 2
> set ytics 0.5
> 
> set multiplot layout 2,2 
> 
> plot "dengue.csv" u 1:3 w l, "" u 1:2 w l ls 1
> plot "dengue.csv" u 1:3 w l, "" u 1:4 w l ls 1
> plot "dengue.csv" u 1:3 w l, "" u 1:5 w l ls 1
> plot "dengue.csv" u 1:3 w l, "" u 1:6 w l ls 1
> 
> unset multiplot
> 
>  http://old.nabble.com/file/p32538776/screenshot.png 
> 
>  http://old.nabble.com/file/p32538776/dengue.csv dengue.csv 
> 

-- 
View this message in context: http://old.nabble.com/Fewer-xticks-in-date-mode-tp32538776p32549607.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] multiple axis with constraint

[Martin G. <mg...@lb...>]

Hi Hernán:

Thanks for your help. I could finally manage to "map" x1 and x2 axis by putting the numbers "by hand" along x2 axis:
set xlabel 'i1'
set xtics nomirror
set x2label 'i2'
set x2tics("i2_value" i1_value_such_as_c(i1_value,i2_value)=0, ...)
These graphs are not easy to do, but they rock like crazy! Thanks again for your help!

Martin.





On 09/27/2011 10:59 AM, Hernán Gonzalo Asorey wrote:
> Dear Martin,
>
> On Tue, Sep 27, 2011 at 14:48, Martin Genet <mg...@lb... <mailto:mg...@lb...>> wrote:
>
>     Hi Hernán:
>
>     First of all, many thanks for your help.
>
>     On 09/26/2011 06:19 PM, Hernán Gonzalo Asorey wrote:
>
>         On Mon, Sep 26, 2011 at 21:30, Martin Genet <mg...@lb... <mailto:mg...@lb...> <mailto:mg...@lb... <mailto:mg...@lb...>>> wrote:
>
>             I have a system with two inputs (i1 & i2), and one output (o).
>             The two parameters are linked by some kind of constraint (c(i1, i2) = 0).
>
>             I have data in the following format: #i1 i2 o.
>             Of course, I can easily plot o(i1) and o(i2).
>             I would like to create a plot where I have i1 in x1 axis, and corresponding i2 in x2 axis, while having o in y1 axis.
>
>         perhaps this:
>         set x2tics
>         set xtics nomirror
>         set xlabel "i1"
>         set x2label "i2"
>         set ylabel "o"
>         plot 'myfile' u 1:3 w l t "o(i1)", '' u 2:3 axes x2y1 t "o(i2)"
>
>
>     However, the problem I see here is that there is no "mapping" between x1 et x2 axis, though x2 axis should be a transformation of x1 axis through c constraint. Is there a way to put, for each label of x1 axis, a corresponding label in x2 axis?
>
>
> As far as I know, you can replace (or also add) major tics manually. For example (from gnuplot help xtics manual page):
>
> set x2tics (1,2,4,8,16,32,64,128,256,512,1024)
>
> will put x2tics at 2^n, for n=0..10, but you should add them manually. You can even do this:
>
> f(x)=2**x
> set x2tics (f(0), f(1), f(2), f(3))
>
> and so on, but I don't know if there is a way to force the x2tics (or any other axes tics) directly from a function without specify them in some sense.
>
> Cheers,
>
> Hernán
>
>
>     Thanks again,
>     Martin.
>
>
>
>
>
>
>