[50.1 PMMA @ F2]
There isn't a whole lot to say about the photo taken at F/2 which I haven't already said about photo taken at F/1. The presence of spherical aberrations still ruins the photo, contrast is still low, and it really isn't usable. The image is slightly clearer than F/1, but not by a lot.
However, one good point I can highlight is the presence of strong bokeh. While it may seem difficult to differentiate between bokeh and spherical aberrations, you can see around the contrasty edges of the bottle a strong difference between the "in-focus" area and the out of focus area behind it. This level of out of focus blur (bokeh) would be highly desirable if the lens design could be improved to remove the SA and increase sharpness of the in-focus areas. However, with the lens in its current design, the presence of strong bokeh only contributes to what looks like a completely blury image.
[50.1 PMMA @ F4]
At F/4 we are starting to finally see some worth while improvements. The image is starting to clear up and we can suddenly see the first signs of the lens becoming crisper. The bokeh which I mentioned at F/2 is more obvious here thanks the reduced SA around the centre of the image (look at the top right shoulder of the bottle; niceee). Unfortunately as the F/# increases so does the depth of field. This means that by the time the image is suitably sharp (like F/16 for example), the bokeh won't be present anymore.
The centre of the image has visibly become clearer. It is clear as to why this is when we compare the ray traces of F/2 and F/4. The extreme outter rays on the F/2 ray trace were being refracted at a bad angle, meaning they would converge ahead of the image plane. However, in the F/4 ray trace we can observe that rays in the same outter position are now blocked by the stop down ring, and only the more precise contral rays are entering through the optic and hitting the image plane. This is what is resulting in the sharper image.
The strong SA still present in the image seems to be taking on a halo effect, forming a ring around the sharper centre. Again, looking at the ray trace tells us why. The green rays are entering the optic at the most extreme angle for this focal length lens (50mm, ≈ ±20 degrees from the optical axis) and we can see that they converge way ahead of the image plane. The rays carry on past convergence and are spread apart by the time they hit the sensor, producing the blur we see in the image.
[50.1 PMMA @ F8]
The photos taken at F/8 is not what I was expecting. The image doesn't seem to fit between F/4 and F/16 in terms of sharpness. This is not the fault of the lens, but rather the fault of my sloppy testing. Although I didn't spot it at the time, it looks to me like there is motion blur in the image, as if the camera was accidentally nudged during the exposure. This has caused the image to blur just slightly and not really give an accurate representation of what the lens can do at this aperture.
Looking instead then at the ray trace, we can see that the image at this aperture should be much clearer. With only the very central part of the lens actually being used, the rays all nicely converge on the image plane when entering at 0 degrees. The green rays also look tighter, and while they still don't converge anywhere near the image plane, the smaller spread at the image sensor results in reduced SA. This gives the appearance that the image is getting clearer.
In the next post I will look at the best result of the lot, F/16.
