Is this a fair comparison, an image stabilised zoom against a manual focus prime? Well the zoom was designed and manufactured in the 21st century and the old primes in the 1960s. Possibly 50 years between them.
Judge for yourself.
The overall view taken at a focal length of 50mm
Below are 100% crops of the sign to the right of the lion.
Sony 16-50 at f5.6
Zuiko 50mm f1.8 at f5.6
Further testing - to investigate the influence of auto focus error on the results of the tests I conducted further experiments with the following parameters. Manual exposure, camera on tripod with 2 sec shutter delay. 3 shots for each situation, best chosen for comparison.
1) Sony at f8 autofocus
2) Sony at f8 manual focus at f5.6 ( wide open)
3) Zuiko at f8 manual focus at f5.6 ( to match Sony)
Default processing in LR and PS. No adjustments made. Actual pixel crops from top left hand corner of image.
Sony f8 - manual focus (137kb)
Sony f8 Autofocus (137 kb)
Zuiko f8 (196 kb)
The manually focused Sony result is slightly better than that achieved by autofocus, but the Zuiko is clearly superior, both in terms of the JPG size, but also visually. Actually you don't need to do any of this testing as the Zuiko looks far better in the EVF!
Overall view ( Canon 24-105 on 5DII)
Actual pixel crops of top left corner. (Click for full size)
Pentax 50mm on NEX (JPG size 187 KB)
Sony 16-50 on NEX (JPG size 142 KB)
Canon 24-105 on 5DII (JPG size 179 KB)
My pecking order would be Pentax 50mm f1,7, followed by Canon 24-105 with the Sony a poor third, but judge for yourself.
This test demonstrates quite nicely why I prefer to carry a lightweight Sony NEX 6 rather than have the Albatross of a Canon 5D11 around my neck!