hello SARG :-) the following code fails to compile, however if you comment-out either sub cadd or sub csub then it compiles and runs OK type double_complex re as double im as double end type sub cadd naked cdecl(byref result as double_complex, byref x as double_complex, byref y as double_complex) as...

try using redim instead of dim

@MrSwiss, what's the expected output?
I have a 3 year old version of FB and it compiles your code with no problem

@badidea yes, plotting arrays looks interesting dim as string plot_command const max=20 plot_command=chr(34)+"array A["+str(max)+"]=[" for i as long=1 to max plot_command+=str(i) if i<max then plot_command+=", " end if next plot_command+="];" plot_command+=&qu...

you are right, it's just a guess how long it took, being a VM I was doing other things in the host OS

the dependencies of Librsvg are high, you would probably windup with 20+ dll's and the header translation may not be so simple, have not tried with fbfrog yet.
headers and examples for Cairo should be in the FB distribution

yes, it's 1903, that was about 3 or 4 weeks ago.

unfortunately I did not time it, all I have is HDD and it did not take that long, less than one hour.

@fxm, I have it installed in one of my VM's, but I spend very little time on Windows nowadays

hello cbruce I have exactly the same function coded in FreeBASIC that is in a paper I'm working from where the function was coded in MatLab. Here are the results of the two: FreeBASIC result: z[n+1] = 0.007268783820012809 + 0.05340708046876657 * i MatLab result: z[n+1] = 0.0072687838200128062515 + 0...

hello cbruce
I suspect that it could be a precision difference in the complex math calculations between that of FB and Matlab, that is, one or the other does the complex math more accurately, unfortunately I can't verify this.

my apologies cbruce, I edited my post after searching the web.

MathWorks wrote:MATLAB constructs the double-precision (or double) data type according to IEEE® Standard 754 for double precision.

so does FB, please show some math calculations (not jus the results) in MATLAB and the same in FB

in dislin.bi inclib probably was including the wrong or missing lib
btw, this thread has gone way off topic, one reason I have not participated, even this post is off topic.

Tourist Trap wrote:Giac is a C++ library, it is the CAS computing kernel
I'm not good with external libraries linking and using from inside FB. And here it's some C++ stuff, it may be difficult to interface.

right, only a heavy-weight expert could tackle that one.