erf function inputs not clear

[emnajaoua]

Hi all,

in the description in this link http://zone.ni.com/reference/en-XX/help/370051Y-01/cvi/libref/cvierf/

it is öentioned that the erf function Erf (double x, double *erf) has 2 inputs.the second input should be an array of double but what exactly represents this inpt to the function because as far as i know erf in matlab have only one input.

Any explications please.

[Barp]

You are looking in the wrong place...

This function you showed is documentation for LabWindows CVI:

http://zone.ni.com/reference/en-XX/help/370051Y-01/cvi/libref/cvierf/

However, you should be looking at this document:

http://zone.ni.com/reference/en-XX/help/373123B-01/lvtextmath/msfunc_erf/

In this documentation, the function is y= erf (x), as you would expect.

[emnajaoua]

Hi Barp,

Actually, I am working in labview communications design and not labview so the "erf" function is not defined in the C node.

I searched the erf function in this link http://www.ni.com/documentation/en/labview-comms/1.0/c-ref/analysis-lib/

and it directed me to Labwindows CVI.

as you can see the error.Any other suggestions please ?

Thanks,

[altenbach]

---

@emnajaoua wrote:

Actually, I am working in labview communications design and not labview so the "erf" function is not defined in the C node.

---

So why did you post this in the LabVIEW Mathscript RT forum? This discussion probably belongs over here.

You can ask the moderator to move this thread.

[emnajaoua]

Hi,

I have already asked the question there but no luck so far.

also the "erf" function belongs to the mathscript RT module as I know so I wanted to search for a solution.

[Barp · Accepted solution]

For this product you are using (which you did not say, but looks like LabVIEW Comms , Erf function is not supported in Mathscript or as G-code.

Now you can try this code:

 a1 =  0.254829592;
 a2 = -0.284496736;
 a3 =  1.421413741;
 a4 = -1.453152027;
 a5 =  1.061405429;
 p  =  0.3275911;

s = 1;
if (x < 0) 
   s = -1;
end
x = abs(x);

 t = 1.0/(1.0 + p*x);
 y = 1.0 - (((((a5*t + a4)*t) + a3)*t + a2)*t + a1)*t*exp(-x*x);
    
y= s*y;

I converted from here: http://www.johndcook.com/blog/cpp_erf/

By the way, this code could also be easily integrated using C-node too. 

Now, most likely Erf function will be available in future versions of LabVIEW Communications System Design.

[emnajaoua]

Hi Barp,

I am sorry for not being that clear.I have tried your solution in the C node but its outputs as not correct (I don't know why by the way).So I have tried the same script in the mathscript node using this http://www.mathworks.com/matlabcentral/fileexchange/55749-error-function-stand-alone/content/erf_sta... and it seems to work.

Thanks for the solution .I missed the idea to search for a standalone function.

Best Regards,

Emna.

[Barp]

yes... I forgot that Mathscript requires ".*" if x is a vector. This is just natural as G-code...

Anyway, I am happy you have a solution until we have this function available for you.