Yet a DA(L) 35/2.4 Review, with Crops and Samples

http://info.xitek.com/pzreview/xitektest/201011/25-54995.html

(in Chinese, English Translation Here (machine-translated))

Lens is tested at different apertures and 100% crops are shown. The lens was tested with a Pentax K-r. Bokeh are checked and wide opened shots are posted. And there are other "normal" sample shots of "real-life" as well.

In particular, look at those wide opened shots at f/2.4, and we would know why a larger aperture / faster lens is always desired if perfect bokeh is wanted:-



Note that the "circles" near the extreme boundaries of the frame (actually closer to the perimeter of the image circle) have become olive shaped owing to the blocking by the lens barrel. So, to get this effect disappeared completely, stopped down is required - typically for one to two stops, depending on the optical design. Hence, assuming that it requires 1.5 stops closed down for the aperture before perfectly round bokeh is obtained, we should not see this effect anymore at f/4, but then the DoF is greatly increased as a result.

In fact, the "blocking" effect for bokeh happens more often for zoom lenses which usually suffer more than primes, owing to a longer lens barrel for zooming which results in even more blocking for such a zooming design. It is therefore prime lenses are still required for more demanding purposes and this included, i.e., for all perfectly round bokeh for background highlights and objects. However, with small fastest apertures of many of those digital Pentax DA primes, this key prime lens advantage has been diminished, frankly speaking.