The resolution data in the table are not much more than a first guess (sorry, dreifish, for not stating this so far): the resolution of the opticallimits-tests (linewith/picture heigth) scaled to the size of the MFT sensor. So when a FF lens is used on MFT the resolution drops to approx 50%, since the picture heigth is approx. only half. With the speed booster the resolution should be better, as a larger area of the original sensor is projected on the MFT sensor (this was also taken into account for the calculated resolution). This calculations are problematic, since even with the same sensor size the data are different with different sensors (e.g. Mpixels) and it is even more problematic to compare different sensors sizes so directly.
But since this was the only data I found on that issue, I thought it may be good to have at least this estimate...
The resolution data for the Zuiko 8mm are even more problematic, as I could find only a test, where the numbers where not given, but read the approx. values out from the graph (that was obviously produced by the same method and even software as in the optical limits tests).
=> At least for the comparison Zuiko 8mm vs. Tokina/0,71x speedbooster, however, it turns out that this estimation was not so wrong (if you have a look at the dpreview link above). The real live comparison is clearly in favour of the Zuiko, the difference probably substantially bigger as in the table, Tokina really looks old...
=> This is discouraging, but this all is valid for photography above the surface only. Let us see how the Tokina/Booster combo works behind a dome (I will be able in few weeks to test it behind the Zen DP-170 and compare to the Zuiko 8mm that I use behind the same dome - In case the Tokina/booster combo performs equally bad as on the surface in comparison, I fear I will not use it so often...).
Edited by Architeuthis, 24 March 2018 - 12:49 PM.