A Second Look at Image Sharpness

Yesterday I discussed a test I made to compare image sharpness when using various tripod mounted techniques. I compared 60 images and ranked them by sharpness.  While some were clearly sharper than others, the process became very difficult with many of the images.  But after a couple of hours comparing photos, I knew the best and worst of the five techniques.

This morning, I cleaned up my directories and consolidated the test images.  I usually shoot in RAW so these JPEG files I made for the test were smaller than normal.  Out of curiosity, I sorted the images by size.  To my surprise the five biggest files corresponded to the 5 images I had ranked as sharpest.  The bottom 5 images were also the at the bottom of my list.  In less than a second, Windows 7 had ranked all 60 files by sharpness.

Sorting by size will not normally rank the images by sharpness.  These images were different.  They were all of exactly the same subject and cropped the same.  Shutter speed and aperture were adjusted so every image received the same exposure.  All the images were ISO 100. When the JPEG compression occurred the image should all have been the same size. The reason that they weren't is because the extra detail in the sharpest images could not be compressed as much.  More detail equals larger files.