I wish more lens tests included consideration of how well the lens(es) handles out-of-focus areas of the image. It's difficult to do because lenses can differ in so many ways in this respect and because there's no one simple measure of "bokeh". People don't even agree on what distinguishes good bokeh from bad. Still... for some of us the way a lens portrays a 3D world when a narrow DOF means only some of it is sharp can be more important than minor differences in corner sharpness. I've seen that one review site makes an effort nowadays to report/test bokeh - I wish more would make the effort.
It so happens that I just finished a comparative test of four lenses including the 35mm 1.8G included in this
Camera Labs group test. The others were a Nikon 35mm f/2 Ai-S, Nikon 35-70mm f/2.8 AF, and Tamron 17-50mm f/2.8 (one model older than the current one - no VR). I tested how they handle out-of-focus highlights by shooting tiny Christmas lights in a dark room, at both infinity focus and closest focus (with the lights in between at 42 inches). I compared near-center to near-edge at several aperture stops. No doubt the test could have been more complete or done in lots of different ways, but I think the results are meaningful and useful. Highlights rendered as discs of circles of confusion (or whatever you want to call those things - I like to call them bokeh-dots) are just one small part of what constitutes out-of-focus quality, but I find that differences between lenses show up well, and in my opinion ugly bokeh-dots are a real turn-off.
The test results are published as a single image at Flickr:
HERE
Please tell me what you think and/or offer ideas for improved testing/comparison of bokeh.
[this is my first-ever post here and I hope it's a welcome one]