Pinhole camera optimal resolution?

[Artlav]

For a given pinhole size in a camera, and camera size, what would be the optimal pixel size to get the maximum resolution at maximum light intensity?

i.e. the larger the hole, the more light enters through it, but eventually the sharpness would drop below the resolution. Conversely, the smaller the pixel, the sharper the image, but eventually it would get smaller than the sharpness of the image due to the hole size, reducing sensitivity at no improvement.

Naively, at infinite distance the rays would enter the pinhole parallel, forming a point the size of the pinhole, and the the pixel should be the same size.

Practically, it does not work out - the image from large hole/small pixel is worse than the image from small hole/large pixel, with "small" and "large" for both being of the same size.
Both cases are worse than large/large due to less light, and no sharper.
And small/small is somewhat sharper, but not bright enough to really tell.

So, how does it work?
Should the pixel be the size of the hole, or something less trivial is at work here?

 

[mfb]

Smaller pixels should not decrease the quality of the image. If they do not help, just reduce the resolution afterwards by adding several pixels each to get larger ones.

I think pixels which are a bit smaller than the hole are fine. If the hole is too small, you'll get additional effects from diffraction.

 

[Khashishi]

I have not studied this, but I suppose it depends on the size of hole, and magnification (distance from hole to screen/distance from hole to object). With a point light source, your features will be blurred to the size of the hole*max(magnification,1/magnification). So that's going to limit your resolution.