From: Brian E. <be...@me...> - 2005-08-02 15:03:48
|
Brian Elmegaard <be...@me...> writes: > > adding CRT_fp8.o to your list of objects. > > This did. Thanks a lot. I have now tried all the solutions that have been suggested. I am able to make one call to a simple routine give exactly the same results with the two ways of running the program when I use crt_fp8.o. However, when I run the full code in any of the suggested ways, the same happens. The standalone program runs fluently, but the mex version run within matlab gets stuck. I am getting to the conclusion that I probably was a little imprecise when I said the same code is used in both cases. This is in fact not true. I do a few things differently (however, this seems to make no trouble) and I do pass fortran a few variables forth and back from matlab over c. Maybe it is just not possible to have exactly the same results when this is done? Maybe matlab and persistent mex arrays somehow are able to change some digits of the numbers? In addition the program does several nested iterations until some accuracy, so probably I will not be able to have exactly the same? regards, -- Brian (remove the sport for mail) http://www.et.web.mek.dtu.dk/Staff/be/be.html |