Re: GoSim - simulink variant that uses Octave as math engine

[CROZIER Richard]

On 23/10/18 07:37, Andreas Weber wrote:
> Am 22.10.18 um 20:28 schrieb GoSim GoSim:
>> Hello, I will not write my name as I want to be anonymous and I read 
>> that you back up all mails.
>> I have written a program that I call GoSim that is a simple simulink 
>> variant that uses Octave as mathematical engine. The basics is done 
>> but I need to fix the library...in short, not completely done but it 
>> works. It definately needs more blocks....
>> My goal is to make it part of Octave but of course you have to 
>> approve, as a plugin is enough for me.
>> I am asking you to try the program, just unzip and run the jar without 
>> changing the catalogue structure. There is a file pid.gsm that have a 
>> model containing two controllers which can have the desired value 
>> changed in real time with the "real time constant block". It's intuitive.
> 
> You can't expect that we (or at least me) run binary files from an 
> anonymous source. Please upload your code to a public repo and provide a 
> link.
> 
> I had a quick look at the .m files in your archive and most of them look 
> like smoke-balls, for example
> 
> function [a]=Subtract(i,h,stop,count,sn,in1, in2)
> a=in1-in2;
> 
> or
> 
> function [param_out, out1]=Derivative(param_in,i,h,stop,count,sn, in1)
> 
> if (count==1) %för att undvika spik i första samplen
>    out1=0;
>    param_out=in1;
> else
>    out1=(in1-param_in)/h;
>    param_out=in1;
> end
> 
> Oh and SplitVector.m can be done much easier.
> 
> At this point I've quit reading and my personal impression is that 
> someone tries to install a trojan horse on my computer. Please apologize 
> if this is not the case.
> 
> -- Andy
> 

I think this is harsh, I doubt this is a virus or malware. It could be a 
very useful project!

Actually these functions look exactly like the kind of thing one 
requires in a simulink-like environment, being functions that perform 
basic block tasks like subtraction and derivative calculation but with a 
standard interface that accepts global simulation parameters.

It will be a lot easier though for people to try it out if the code is 
open source. This is because it will *never* be accepted into Octave if 
it is not open source, and I personally would not want to invest time in 
it while it's closed source in case you later decided never to open 
source it. Some screenshots would be helpful though.

Regards,

Richard




-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.