Methinks accuracy is a somewhat subjective term here.

One of the formulas often referenced is:

d = c * sqrt( f * l)

Where d is the pinhole diameter

c is a constant (that's where the trouble starts)

f is the focal length

and l is the wavelength of the light (more trouble)

Folks seem to use 0.00055 for the light wavelength (which might be a green) but the spectral sensitivity of the film could come into play. Ortho film would likely produce sharper results if one optimized for the appropriate wavelength.

The constant --- oooh --- I think Lord Rayleigh, an early tinkerer with these matters, came up with 1.9. But I've seen people using numbers as low as 1.5. Obviously this cauld bend the results quite a bit. In my quick perusal of the MrPinhole calculator, I didn't see an indication of what he used, although it may be lurking there somewhere. Pinhole Designer defaults to .00055 for the light wavelength and 1.9 for the constant, but gives a user the option to edit those values.

Edit: The 0.0366 is simply bundling numbers, the product of the constant, c, and sqrt( wavelength) for metric values; 0.0073 for inches.

When one considers the issues of fabrication, pinhole cutting, reciprocity failure and possible film flatness problems, it's probably not worth doing three decimal place calculations here.

DaveT