gnuplot development

Suman Dhayal wrote:

> But right now, I am need to combine two plots
> together whose x-coordinate scale factor is the same but not the
> y-coordinate. I used the multiplot command to do this 

Multiplot is almost never the right technique for that kind of job. 
Secondary y axes would be a lot easier to use.

> and since I needed to change the font also for some labels and
> titles, I used the enhanced terminal type.

Nitpick: "enhanced" is not a terminal type.  It's a common terminal 
option.  What actually was your terminal type?

> not use the terminal type "enhanced", the "~" label works but when I
> use it along with this terminal,

Read "help enhanced" again, slowly.  Find the '~' and learn what you 
just did, and how not to do it.

hi,

Thank you for the quick response.. my terminal type is postscript. But I
don't understand how should i use secondary y-axes?? 


Thanking you,

Suman


Hans-Bernhard Bröker-2 wrote:
> 
> Suman Dhayal wrote:
> 
>> But right now, I am need to combine two plots
>> together whose x-coordinate scale factor is the same but not the
>> y-coordinate. I used the multiplot command to do this 
> 
> Multiplot is almost never the right technique for that kind of job. 
> Secondary y axes would be a lot easier to use.
> 
>> and since I needed to change the font also for some labels and
>> titles, I used the enhanced terminal type.
> 
> Nitpick: "enhanced" is not a terminal type.  It's a common terminal 
> option.  What actually was your terminal type?
> 
>> not use the terminal type "enhanced", the "~" label works but when I
>> use it along with this terminal,
> 
> Read "help enhanced" again, slowly.  Find the '~' and learn what you 
> just did, and how not to do it.
> 
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's
> challenge
> Build the coolest Linux based applications with Moblin SDK & win great
> prizes
> Grand prize is a trip for two to an Open Source event anywhere in the
> world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________
> Gnuplot-info mailing list
> Gnuplot-info@...
> https://lists.sourceforge.net/lists/listinfo/gnuplot-info
> 
> 

-- 
View this message in context: http://www.nabble.com/Ebhanced-terminal-type-with-Multiplot-error-tp18614837p18618620.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Loïc wrote:

> set table "out" 
> set xrange [0:10] reverse
> plot x**2

[...]

> Hence, x are in increasing mode, and not decreasing as I wish.
> 
> Is it a bug ? 

No.  There's no promise of the range reversal affecting sampling order. 
  It affects the relation between coordinates and positions on a graph, 
not the order in which those positions are visited.  That ordering is 
generally unimportant, since you couldn't spot the difference in the 
resulting image anyway.

The table driver makes that a bit mor complicated, of course.  But, like 
with so many problems, parametric mode helps:

gnuplot> set table ; set parametric ; set samples 15
gnuplot> f_x(t) = 10 - t
gnuplot> plot [0:10] f_x(t), f_x(t)**2

#Curve 0 of 1, 15 points
#x y type
  10  100  i
  9.28571  86.2245  i
  8.57143  73.4694  i
  7.85714  61.7347  i
  7.14286  51.0204  i
  6.42857  41.3265  i
  5.71429  32.6531  i
  5  25  i
  4.28571  18.3673  i
  3.57143  12.7551  i
  2.85714  8.16327  i
  2.14286  4.59184  i
  1.42857  2.04082  i
  0.714286  0.510204  i
  0  0  i

Many thanks, I didn't think at a parametric curve.

Loïc

Le mercredi 23 juillet 2008 à 00:15 +0200, Hans-Bernhard Bröker a
écrit :
> Loïc wrote:
> 
> > set table "out" 
> > set xrange [0:10] reverse
> > plot x**2
> 
> [...]
> 
> > Hence, x are in increasing mode, and not decreasing as I wish.
> > 
> > Is it a bug ? 
> 
> No.  There's no promise of the range reversal affecting sampling order. 
>   It affects the relation between coordinates and positions on a graph, 
> not the order in which those positions are visited.  That ordering is 
> generally unimportant, since you couldn't spot the difference in the 
> resulting image anyway.
> 
> The table driver makes that a bit mor complicated, of course.  But, like 
> with so many problems, parametric mode helps:
> 
> gnuplot> set table ; set parametric ; set samples 15
> gnuplot> f_x(t) = 10 - t
> gnuplot> plot [0:10] f_x(t), f_x(t)**2
> 
> #Curve 0 of 1, 15 points
> #x y type
>   10  100  i
>   9.28571  86.2245  i
>   8.57143  73.4694  i
>   7.85714  61.7347  i
>   7.14286  51.0204  i
>   6.42857  41.3265  i
>   5.71429  32.6531  i
>   5  25  i
>   4.28571  18.3673  i
>   3.57143  12.7551  i
>   2.85714  8.16327  i
>   2.14286  4.59184  i
>   1.42857  2.04082  i
>   0.714286  0.510204  i
>   0  0  i
> 
>

> ...
> $ splot "radET1.hdat" u 1:2:( $4==0 ? 1/0 : 4 ) w image
>
> would do the trick, but doesn't. It always gives errors like
> "GNUPLOT (plot_image):  Image grid must be at least 2 x 2" or
> "GNUPLOT (plot_image):  Visible pixel grid has a scan line shorter than
> previous scan lines"

'help image' says:
> ... It is assumed that the image data forms an
> equidistant sampling grid in the viewing plane along two
> ... directions.

for each grid point a value is required, so exclusion is not possible.

> Mabe it isn't even possible to exclude points from grid data like this?
> Any clues??

make the unwanted points white:

set pal mod RGB funct (gray>1./400.?sqrt(gray):1), (gray>1./400.?gray**3:1),
(gray>1./400.?sin(gray*2.*pi):1)

(you may replace 1./400. by some value smaller than your valid data)
-- 
View this message in context: http://www.nabble.com/heatmap-%2B-ternary-operator-problem-tp18565447p18570092.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Thanks, after some further research I came to the same conclusion...

> set pal mod RGB funct (gray>1./400.?sqrt(gray):1),
> (gray>1./400.?gray**3:1), (gray>1./400.?sin(gray*2.*pi):1)

This is quite fancy! Amazing...
by now I changed the coloring to a simple gradient:
white,purple,blue,cyan,green,yellow,red
Which also works - if not as sophisticated...

Many thanks,
Thomas
-- 
View this message in context: http://www.nabble.com/heatmap-%2B-ternary-operator-problem-tp18565447p18574570.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

> ...
> set term post color enhanced 

it's the 'enhanced' option which positions the xlabel at a slightly
different position than 'noenhanced' mode does.
the bounding box should be adjusted differently in 'enhanced' mode,
but it isn't.

you could adjust (with a simple text editor) the bounding box from
%%BoundingBox: 50 50 410 302
to
%%BoundingBox: 50 48 410 302
in the eps-file to get the complete xlabel shown.

-- 
View this message in context: http://www.nabble.com/eps---cuts-down-the-x-label-tp18507008p18532230.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

> Many thanks, Thomas, this works! I reset the bounding box
> with pico. Is it possible to make it via gnuplot? Or is it better
> to write a little script to change the bounding boxes in all of
> my figures? I have a lot. :-) And I've never did such a thing. . .

you don't need to (i just found out how it works):
there is the 'noenhanced' option for 'set xlabel' (and other labels)
which switches the enhanced mode off for this label.
it's useful for labels which contain no special 'enhanced' sequences.

just append 'noenhanced' to your 'set xlabel' command and the xlabel
will appear at the same position as it would appear in 'noenhanced'
mode.
-- 
View this message in context: http://www.nabble.com/eps---cuts-down-the-x-label-tp18507008p18545628.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

> How can I set the label to display in the same position, i.e. Top
> of graph if the graph starts with y values from 0 to 2, but later a
> value pops in to say 50?  If i set it using coordinates, then the
> whole scale changes, the position of the labels change.  Basically
> I'd like to create a fixed label...

'help set label' gives an answer:
> ...
> The <position> is specified by either x,y or x,y,z, and may be preceded by
> `first`, `second`, `graph`, `screen`, or `character` to select the
> coordinate
> system.  See `coordinates` for details.

and 'help coordinates' tells you:
> ...
> `graph` specifies the area within the axes---0,0 is bottom
> left and 1,1 is top right

set label 1 "abc" at graph 0.5,0.95 center

will place your label centered in the middle of the x axis just below the
upper y axis.

-- 
View this message in context: http://www.nabble.com/Gnuplot-updating-graphs-in-real-time-tp18495467p18514697.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

atoms.h wrote:
> Is it possible to contrain a fit parameter?
> For example a>0 or a=2..3?

You can use a function of your choice to map the fitted parameter to the 
one used in the function, e.g.

	a(a_fit) = exp(a_fit)
	# or
	# a(a_fit) = 2+tanh(a_fit)/2

	fit f(x,a(a_fit)) 'data' via a_fit

Dear Robert,
you have to use gnuplot 4.3 from CVS in order to make this work (if there is a 
precompiled M$-binary, what I do not know).
The only other alternative for now is opensuse 11.0 because the version of 
gnuplot that is shipped with it has been patched to work as you'd expect it 
to.
Hope this helps
take care




Dieter Jurzitza

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

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

if you really want to see the pictures above - use some font
with constant spacing like courier! :-)
-----------------------------------------------------------Am Mittwoch, 2. 
Juli 2008 12:43:34 schrieb Robert Schikowski:
> Hi all,
>
> I'm using gnuplot 4.2 on Windows XP. What I want to do is to plot from a
> file containing IPA symbols encoded in UTF-8. However, if I do
>
>    set terminal png font arialuni
>    plot 'file' using 2:3:1 with labels
>
> the correct font is used, but the UTF-8 file seems to be interpreted as
> ISO- Latin-1 or whatever the default setting is (e.g. inverted e's turn to
> an accented E). I tried to do
>
>    set encoding utf8
>
> but I was told by the terminal that it was
>
>    expecting one of 'default', 'iso_8859_1', 'iso_8859_2', 'iso_8859_15',
>    'cp437', 'cp850', 'cp852', 'koi8r', 'koi8u'
>
> Is there any way to add UTF-8? Or any other way to get my IPA symbols onto
> the plot?
>
> Robert
>
>
> -------------------------------------------------------------------------
> Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
> Studies have shown that voting for your favorite open source project,
> along with a healthy diet, reduces your potential for chronic lameness
> and boredom. Vote Now at http://www.sourceforge.net/community/cca08
> _______________________________________________
> Gnuplot-info mailing list
> Gnuplot-info@...
> https://lists.sourceforge.net/lists/listinfo/gnuplot-info