Puzzled Lens Testing - m43 lenses on e-m5

worthless to check such result with such test , too much criteria varies and not much as in control. First off, Ok its RAW, but how do you view it. What's the software used to de-mosaic the info ... , is software correction of any sort applied. then there's the handheld, sorry, no matter how steady you held, you are not going to beat a tripod, with live view for critical placement of plane of focus using manual focus. Then that of ISO, sorry but if you really need to test, then you need to test all of them on the same camera at same ISO setting ( which I think you are hopefully cause it sound like it might not be ). You need a target that's evenly lited ( so that there is less than say 1/10 stop of light across the target , and you need to meter that using an independent calibrated meter and shoot all of the test on that measured exposure ( so you know if the test subject actually give you the light gathering as spec )

And as for the result, let me put it this way, I do not think 100% crop viewing actually tell the story .... you got to look deep into the TEST and the RAW to check the difference ( not necessary a better or worse scenario ).

At best the way you test can be said to be a loosely sampling of field shoots and that give you some idea of how the combo works , but its perhaps not good enough for judging those lens for real. And without knowing the specifics I dare not saying why you would get those results.