[daip] Re: TVSTAT

[Eric Greisen]

K. Dyer writes:
 > Well I have empirically found the formula:
 > 
 > 3.1416*(major/2.354)*(minor/2.354)*2/(cellsize**2)
 > 
 > but I have no idea where the 2 comes from. (2.354 is a conversion from
 > FWHM-> sigma).
 > 

    Int (exp(-(ax)^2) dx = sqrt(pi) / a

    fwhm = F  :  exp (-(ax)**2) = 1/2
                 ax = sqrt(ln(2))
                 a = 2 sqrt(ln(2)) / F
                 a =1.6651 / F

    The problem according to MR is a is not sigma.

Eric