Some views of a random bike, trying out focal lengths and viewpoints.
In the context of the assignment brief, the best images are those in which the subject almost fills the frame. Minimum distortion occurs at the long end of my standard zoom, 42mm (84mm equivalent) and a viewpoint 4-5 metres from the subject. The frame can be filled at the 14mm (28mm equivalent) end of the zoom range and about 1.5m from the subject, but there is no valid reason for introducing the associated distortions.
An alternative approach is to use the long end of the zoom to extract details.
This will be my fallback plan if I cannot make my original idea work.
Even with the bicycle filling the image frame, there is a lot of background visible. Ideally, this should be de-emphasised by differential focussing, requiring a wide aperture. My next experiments will be made using an OM-system 50mm f/1.4 lens. Initial results suggest that it is a little soft at f/1.4 so I will work with it at f/2, which is wider than anything I have in my E-system kit. However, with no auto-focus and no focus-confirmation aids (microprism or split-image) it will take some practice.
A further thought is that series of similar images, such as I intend, will be a typological exercise similar to the work of Bernd and Hille Becher, who will be the subject of further research. (example below)