![]() Are you able to pair X's, Y's first and then associate the Z's ?. If you had collected the data yourself, you should be able to collect them again or have them collected so you can process visibly. The proof of that is: the (X,Y,Z) plot has no apparent interpolation and it should if the data set would be true. Bivariate can only represent surface of bivariate format, and format is conclusive of expertise and or multiple experiments. You have about 240 values, that is would fill about a 15 x 15 square array, that you could spline and get a very faithful surface plot that you could further exploit. I hope you are fully aware of the pitfalls of high order polynomial fits.Īt this point now, my opinion is that the project is wrong from the data collection. You are getting the contours for the fitted surface, not the actual data points. You just have to scale those indices to actual data values - you already did such a scaling in creating the x and y arrays. As the contour routines only get the matrix of values, the contours are in terms of the indices for those matrices. Since you didn't include what you had done with the contours, I don't know what your scaling problem is. That results in a quite compact size (commonly even smaller than the MC11 format). Use the compressed XMCDZ format, had set the XML options (in the file properties) to no generated images). Just don't use the default XMCD format - that is bloated and often much too big for comfortable uploading and downloading. :=.= is one) feel free to post in MC13 or MC14 format. If you are using MC14 (or MC13) features that don't save in MC11 format (the combined assign/display idiom. While Jean insists on MC11 format, I am much more flexible. Mayhap this approach can be used to constrain the limits of your data so that the Contour Intervals may be of greater help. I don't have experience with this method. In the thread Tom mentioned, there is a contour3d.zip that suggests the use of min and max coords. In the contour plots example (attached), which is simply a scratchpad of work, I have explored further this question, but note that the plot derives from an equation, not data, so there are features of the plot (QuickPlot Data) available that apparently are not for data. It is an iterative process, and not guaranteed to produce an exact result. Note in Isentropic (attached) that I have adjusted the number of contour intervals to get one that is close to the value you wanted. I have tinkered with this in the past because I wanted the same thing from Mathcad. ![]() ![]() ![]() That would get you the specifics that you desire. The problem is that the contour plot is not as flexible as one in, say, MATLAB, where one may specify the beginning and ending values for the contour region, as well as the contour interval. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |