Numismatic and History Discussions > Coin Photography, Conservation and Storage

Choice of aperture


Here's another interesting comparison to go with the "choice of exposure" thread.  These are strips across a coin, each one taken at a different aperture, from f2.8 to f45, as shown on the  image.  The focus was set once at the start, and was not adjusted between these shots.

Depth of field is very important.  At f2.8, there is very little tolerance.  You can see that here, either the coin or the camera is not positioned perfectly level.  You'll never get an ancient coin to be perfectly level, nor of course perfectly flat.  So in this case, at f2.8, the entire right side of the photo is blurred.

Reducing the aperture improves depth of field, until at f8 and f16, the result is pretty good.  This also has the advantage of directing the light only through the part of the lens which is most likely to be very accurately made, and which should have the least optical or chromatic aberration.

Reducing the aperture further blurs the whole of the coin.  At f45, the smallest aperture my camera can do, this is very noticeable.  This effect is casued by an optical effect called diffraction.

For a 35mm digital camera, f16 is about the best option for sharp results.  If you are using a smaller camera, the optimum might be f8 or even f4.

It isn't the choice of F-stop that is blurring the image, it is the camera changing the shutter speed/ exposure timt to compensate. You could use twice as much light [read intensity]  and f45 and get the same picture you have at f16. Each change is f-stop requires twice as mich light to have the same exposure value, or you must compensate with a changed shutter speed or exposurer duration. Many modern lenses have a peak true sharpness, measure in line pairs per mm, around f22, but each lense would have to be tested, but there is usually data, but that is a bit hardcore and not really needed. Try the same experiment, and inrease the strength of your light.


In the last set, the light source was 16 cm from the coin.  In this set, it's 4 cm from the coin.  The results are just the same - somewhere between f8 and f16 there is a crossover point and the picture starts to become more blurred.

I have observed this effect in the past in general photography.  A trawl around the web suggests that f8 to f16 is a pretty typical point for this sort of effect to begin.  here's a typical quote:

"As two examples, the Canon EOS 20D begins to show diffraction at around f/11, whereas the Canon PowerShot G6 (compact camera) begins to show its effects at only about f/4.0-5.6."  Here's the link, for those who are interested - it's a tutorial page on diffraction in digital photography.  It points out that the lens isn't the only factor in diffraction effects.  There is also a relationship to the resolving ability of the camera sensor.

Having run these tests, I am standardising with this camera on f11.

That makes sense. Traditional photographic lense testing showed many lenses had a maximum sharpness at f22, and that would accord with what you  are showing.



[0] Message Index

Go to full version