Hi everyone!
I am trying to plot some fields (velocity and pressure fields) using the colormap option. The results is quite good but I need help for some improvements. Here is my code: cd output data_U = csvread('U_2014_10_05.csv'); cd .. X_min = min(data_U(:,1)); X_max = max(data_U(:,1)); Y_min = min(data_U(:,2)); Y_max = max(data_U(:,2)); data_U(:,1) = data_U(:,1)  X_min; data_U(:,2) = data_U(:,2)  Y_min; Xi = linspace(0,X_maxX_min,1000); Yi = linspace(0,Y_maxY_min,1000); [XX YY]=meshgrid(Xi,Yi); Ui = griddata(data_U(:,1), data_U(:,2), data_U(:,3), XX, YY); map=jet(16); colormap (map); surf(XX,YY,Ui); shading interp; #title("Vista frontale"); #xlabel("x"); #ylabel("y"); axis ('off', "image"); colorbar("EastOutside"); #set(gca, 'XGrid', 'off'); #set(gca, 'YGrid', 'off'); #caxis("auto"); caxis([0 93.6]); view(2) print color velocita_2014_10_05.eps My issues are: 1. The colorbar overlaps the graph even if I used the "EastOutside" option. Why? 2. Is it possible to have the maximum value (93.6) shown on the top of the colorbar? 3. Is it possible to add, instead of the two classical axis, a small reference system in the lower left corner? Something like the small set of arrows that usually show up in the lower left corner in Paraview (for the ones who are familiar with it). Maybe without color, just with black arrows and the labels for the 3 directions (x, y and z). 4. The graph is good but not correct at the moment and I need to exclude some values:  The values inside the blade (all zeros) so the blade will be white as the outside of the channel.  The values outside the pressure side of the channel: they are not real, but they have been created by the interpolation. I do not understand why only there and not outside the suction side for example (look at the two images attached for a better understanding of the problem). Can anybody help me? Thanks in advance! 
On Sun, Oct 12, 2014 at 9:45 AM, guido.bonalumi <[hidden email]> wrote: 
Hi everyone! Tell us what version of Octave and what OS. _______________________________________________ Helpoctave mailing list [hidden email] https://lists.gnu.org/mailman/listinfo/helpoctave 
Octave 3.6.1 on Ubuntu 12.04 LTS 64 bit. Sorry!

On Sun, Oct 12, 2014 at 10:24 AM, guido.bonalumi <[hidden email]> wrote: Octave 3.6.1 on Ubuntu 12.04 LTS 64 bit. Sorry! Using 3.8.1 the color bar is to the side. I think you should upgrade.  _______________________________________________ Helpoctave mailing list [hidden email] https://lists.gnu.org/mailman/listinfo/helpoctave 
Ok I will upgrade as soon as I can (I am upgrading TexLive right now).
Meanwhile any idea to solve the other problems? In particular the last one, which is quite tricky. 
This post was updated on .
I have updated Octave to the latest version (3.8.1) and this is the result:
Honestly this result is even worst than the previous one! The colorbar still overlaps the graph and now its scale is less detailed. The code is still the same, why do I get this poor result? 
On Sun, Oct 12, 2014 at 6:52 PM, guido.bonalumi <[hidden email]> wrote: I have updated Octave to the latest version (3.8.1) and this is the result: What is the output of graphics_toolkit? Have you tried close all graphics_toolkit("gnuplot") and then plot again? What about "fltk"? (I think that is the default). _______________________________________________ Helpoctave mailing list [hidden email] https://lists.gnu.org/mailman/listinfo/helpoctave 
Actually I do not know what are you talking about!
Unfortunately I am not so expert. I tried adding the two lines that you suggested but I still have the same result, with the colorbar that overlaps the graph. The only difference is that the scale now is more detailed. I do not know anything about graphics_toolkit, do you have some documentation that I can read? Anyway, later I will have to export this graph with matlab2tikz because I have to use it in LaTex. Will this graphics_toolkit option have some effects on this conversion somehow? 
Administrator

In reply to this post by guido.bonalumi
> On Oct 12, 2014, at 9:45 AM, guido.bonalumi <[hidden email]> wrote: > > Hi everyone! > I am trying to plot some fields (velocity and pressure fields) using the > colormap option. The results is quite good but I need help for some > improvements. > Here is my code: > /cd output > data_U = csvread('U_2014_10_05.csv'); Can you make "U_2014_10_05.csv" available for us to download? That way we can determine if the problem still exists in the latest development sources, and can also (more efficiently) try fixing the problem for you. Ben _______________________________________________ Helpoctave mailing list [hidden email] https://lists.gnu.org/mailman/listinfo/helpoctave 
Of course! Here it is:
U_2014_10_05.csv 
I was trying this small code that was written for me by a friend of mine:
clc close all clear all cd output a=csvread('U_2014_10_05.csv'); cd .. x=a(:,1); y=a(:,2); z=a(:,3); scatter3(x(:),y(:),z(:),[],z(:)); colormap(jet(16)); shading interp colorbar; It was written it for Matlab, and running it with Octave I receive this error: 'invalid error for color property "markeredgecolor"' The error is referred to this line: scatter3(x(:),y(:),z(:),[],z(:)); but I checked and the syntax seems right to me. Do you know some possible cause to this error? 
On Wed, Oct 15, 2014 at 1:06 PM, guido.bonalumi <[hidden email]> wrote: I was trying this small code that was written for me by a friend of mine: Does this works? x = rand(16,1); y = rand(16,1); z = rand(16,1); scatter3(x(:),y(:),z(:),[],z(:)); If not, what version of Octave? What is the output of graphics_toolkit? If I execute the rest of the code then I get an error, but a different one. _______________________________________________ Helpoctave mailing list [hidden email] https://lists.gnu.org/mailman/listinfo/helpoctave 
ok I did some tests on my 2 PCs:
1) First test: PC: 1, the one that gave me the problem Octave version: 3.6.1 graphics_toolkit: gnuplot result: your code works, there is a little error on the y axis because the values rapresented on the scale overlap all on the first value, but the graph is ok 2) Second test: PC: 2 Octave version: 3.8.1 graphics_toolkit: fltk result: your code works and the y axis scale's values now are fine. My code doesn't work! A window appears but there is no graph inside, and nothing else happens neither an error. 3) Second test: PC: 2 Octave version: 3.8.1 graphics_toolkit: gnuplot result: your code works but there is the same error with the y axis as in the first case. My code doesn't work! Nothing happens! 
On Wed, Oct 15, 2014 at 8:18 AM, guido.bonalumi <[hidden email]> wrote: ok I did some tests on my 2 PCs: I tried the code: x = rand(16,1); y = rand(16,1); z = rand(16,1); scatter3(x(:),y(:),z(:),[],z(:)); and it works correctly here. ubuntu 14.04 and octave 3.8.1 graphics_toolkit('fltk') and with graphics_toolkit( 'gnuplot') Doug _______________________________________________ Helpoctave mailing list [hidden email] https://lists.gnu.org/mailman/listinfo/helpoctave 
I checked again, adding this to lines just to have an output:
view(2) print color prova.eps and your code works fine, the output file (prova.eps) is ok, even the y axis. I do not understand why my code is not working. 
On Wed, Oct 15, 2014 at 9:16 AM, guido.bonalumi <[hidden email]> wrote:
I checked again, adding this to lines just to have an output: Are you asking about the results of the print command, or the results on the plot window???? Doug _______________________________________________ Helpoctave mailing list [hidden email] https://lists.gnu.org/mailman/listinfo/helpoctave 
Sorry but I do not understand your question. What I see in my .eps file and what I see in the plot window should be the same. Am I right?

On Wed, Oct 15, 2014 at 9:56 AM, guido.bonalumi <[hidden email]> wrote:
Sorry but I do not understand your question. What I see in my .eps file and _______________________________________________ Helpoctave mailing list [hidden email] https://lists.gnu.org/mailman/listinfo/helpoctave 
Ok now I understand. I added the two lines to the code suggested by Juan Pablo. With his code the .eps file is fine so it means that is just a problem of visualization in the plot window.
Adding the two lines at my code is useless because it crashes when the plot window shows up. The window is empty and the code seems to be stuck. 
On Wed, Oct 15, 2014 at 10:13 AM, guido.bonalumi <[hidden email]> wrote: Ok now I understand. I added the two lines to the code suggested by Juan Let us go 1 step at a time does this code 'crash' octave? x = rand(16,1); y = rand(16,1); z = rand(16,1); scatter3(x(:),y(:),z(:),[],z(:)); I just had a new idea. Did you wait long enough for the data to be processed? You have a large data set and it will take a long time to get drawn. The time obviously depends on the speed of your machine etc. Doug _______________________________________________ Helpoctave mailing list [hidden email] https://lists.gnu.org/mailman/listinfo/helpoctave 
Free forum by Nabble  Edit this page 