Fujifilm GFX100S: Focus Accuracy and Precision with Fujifilm GF 35-70/4.5-5.6 @ 58mm
Related: Fujifilm, Fujifilm GF 35-70mm f/4.5-5.6 WR, Fujifilm GFX100S, Fujifilm Lenses, Fujifilm medium format, Fujifilm mirrorless, medium format, precision and accuracy
re: precision and accuracy
re: focusing accuracy
re: Fujifilm GFX100S
See my prior post on this matter, which prompted this controlled test: Autofocus Accuracy: Fujifilm GFX100S + Fujifilm GF 35-70mm f/4.5-5.6.
I want my work to be its sharpest possible, and that means understanding the limits of camera autofocus, and if necessary, using manual focus. Particularly for a 100MP camera with 3.76 micron pixel size, there is no margin for error.
This page assesses that focus error and variability under controlled conditions using the most favorable target one might ask AF to work on, a Siemens star. The results might cause indigestion or consternation, but they confirm what I am seeing in outdoor shooting, which is what prompted this effort.
Fujifilm GFX100S: Focus Accuracy and Precision with Fujifilm GF 35-70/4.5-5.6 @ 58mm
Includes a dozen frames versus a reference frame, discussion, mitigation advice.
It is fine to be skeptical at these results, and I expect it might even be upsetting to some readers. But the GFX100S is a 100MP camera, and there is little margin for error. Set up your own test using the protocol I describe as the best way to turn disbelief into being convinced, or to prove otherwise.
To emphasize: two GFX100S camera bodies prove that this is not a “bad camera” (a ludicrous idea, given software and PDAF). Indeed, the 2nd camera body was borrowed from a friend who has been very happy with its AF performance for people and similar. Also reproduction of similar results with three lenses on the Siemens chart and four lenses outdoors on two different targets prove that it is not a lens-specific issue.
Fujifilm GFX100S + Fujifilm GF 35-70mm f/4.5-5.6 WR @ 47.4mm equiv (57.6mm) RAW: Enhance Details
[low-res image for bot]
