[stoppingdown]

After one year of a previous edition of the test with the a6000 and the Sigma 150-600mm C, I did it again with the a6300, to see whether things have improved with a better autofocusing system.

 

They did, definitely.

 

Test conditions: sunny day, hand-held, 1/1000 sec, Æ’/8, auto ISO; focals from 480mm to 550mm. I classified shots in three classes: ok! (perfect sharpness), decent (acceptable sharpness), ko. The focusing mode was "lock-on", that is I first focused using the center sensor, then the camera tracked the subject (so it took into account the fact that I couldn't precisely track the subject, and/or I could decide to place it into another position) and I was able to see the active sensors around the subject; so I could visually confirm where the camera was focusing. When the tracking was lost, I restarted the process.

 

The original test set was composed of 210 shots:

 

50 (24%) ok!
53 (25%) decent
107 (51%) ko
 
But I must take into account a strange thing: starting from a certain point (the latest 67 shots), auto-focus always failed - got it near, actually, but far from decency. I don't know what happened; perhaps I was tired of hand-holding (but the problem is not motion blur, is mis-focus)... or ... who knows? I recall I could confirm that the camera had a harder time putting the highlighted sensors around the subject.
 
If something happened - for instance, some error of mine - not taking into accounts the latest failure sequence numbers would be:
 
143 photos
50 (35%) ok!
53 (37%) decent
40 (28%) ko!
 
That is, one third/one third/one third.
 
The camera had a harder success ratio when the background, instead of the sky with or without clouds, was the slope of surrounding mountains: sometimes the camera was fooled and tracked it. But even in this scenario, some good shots turned out.
 
All in all, that's good. It should be definitely fine for BIF with a predictable motion.