EDIT [9th December 2018]: ZEISS has informed that they are working on a software update that will allow users to work within a larger focus range with a wide-open aperture. This means that when working at close distances the aperture will close down less than what is described here, reducing also the visibility of nonagons. The update will be available to download for free from the ZEISS website in early 2019. ZEISS will notify all registered customers via email. In others words, the information presented here is a bit outdated and the Batis 2/40 CF will work a bit differently with a new FW-update available in early 2019.
With the previous posts I've explored the Close Focus ability of the Batis 2/40 CF. This means that the lens can focus very close and optically speaking it has an outstanding performance. In fact, the Batis 2/40 CF has a floating focus unit and it offers a very high contrast and resolution over the complete focus range, all the way down to the minimum focusing distance of 0.24m (0.78ft). This is of course great news, but there are also some limitations that this design brings to the table. One of the two limitations is related to nonagons and the other is about depth of field. Now, if you are wondering if the 'nonagon' is even an English word I can confess that I had to check it: it is. The term nonagon describes a nine-sided polygon and is actually an accurate and perfect term for the situation (my credits to whoever coined up the term to describe this issue in the first place).
So, this is how it works. When working close to minimum focusing distance the lens closes down the aperture a bit. This starts at distances below 1m (39 in) and happens gradually (see the table below). Because the aperture is closed a bit it doesn't stay fully round even at maximum aperture of f/2. And because the aperture mechanism of the Batis 2/40 CF has nine blades the highlight bokeh balls are sometimes rendered as nine-sided polygons; also known as nonagons. This happens by lens design, ie. it's a design choice, and stopping down probably ensures the high optical performance at close distances. I should also mention that this sort of design is used in many other lenses too, but obviously there are also lenses which stay fully open even at minimum focusing distance (dedicated macro lenses, for example, where this kind of functionality is essential).
Set aperture | Focus distance | 'True' aperture
f/2 > 1m (39in) f/2
f/2 1 m (39in) ~ f/2
f/2 0,4 m (15.7in) ~ f/2.8
f/2 0,24 m (9.4in) ~ f/4
So, using the Batis 2/40 CF you might introduce some nonagons into your pictures. However, this only happens at specific circumstances, so before jumping into conclusions some clearing up needs to be done. First (1), the described aperture behavior only affects close up distances below 1 m (39 in). Above 1 m (39 in) the aperture works just like in normal lens which means that f/2 stays fully open and leads to fully round bokeh balls (but just like in any other lens, once you close down the aperture the polygon shape starts to emerge gradually). Second (2), the actual appearance of nonagons also depends on many other factors than just the position of the aperture. If the light is strong enough the background highlights are rendered as nonagons indeed, but in medium light highlights retain their round shape easier. There also needs to be some clear highlight points at the background in appropriate distance for nonagons to take their shape clearly. And in general, the appearance of nonagons really depends on background and its distance. In many scenes the background doesn't include these elements which means that the nonagons doesn't appear.
Now, I hope I have described the aperture mechanism and nonagons as clearly as possible. The obvious question is: can this design choice limit your photography? I can see two potential issues. First one (a) is the possible appearance nonagons. Many photographers prefer round shapes for bokeh balls and nonagons differ from this preference. The second on (b) is that the aperture mechanism limits the use of shallow depth of field, as the depth of field is not as shallow at f/4 as it is at f/2.
For the nonagons I want to stress that their appearance is not deterministic in a way that every picture taken at close focus range would contain them. Quite the contrary in fact, because I've found them pretty difficult to produce. I've come up with some pictures where the bokeh balls are kind of taking their shape as nonagons, but without someone explicitly pointing them at me I wouldn't have noticed them at all (please see the provided examples, and see couple of more here). Not claiming they don't exist, because they do and there are some pictures in the net that exemplify them better, but like I said my experience is that they are (in their clear geometric form) quite rare. Of course it would possible to use something bright like Christmas lights to agitate them at close focus distances, but it wouldn't represent normal use of the lens. In real life, they happen sometimes, but mostly not.
The possible issue with a shallow depth of field is that it is, obviously, not as shallow at true f/4 as it is at true f/2 – and those who are searching for maximum bokeh effect are going to see this as a limiting factor. To provide some exact numbers, the depth of field at minimum focusing distance with the true f/4 aperture is 3.63 mm (0.14 in) – that's a one third of 1 centimeter. With the true f/2 in a similar situation the depth of field would be 1.82 mm (0,07 in) – which means about 2 mm (0.08 in) difference. If you take your ruler out and see the difference yourself, it's pretty much close to nothing (even at these magnification ratios). Personally I almost always close down more for better focal plane extension at close distances. But of course one can argument that the real issue with the depth of field is not the focal plane, but the speed in which the background is blurred: with f/4 the background doesn't 'blur our' as fast as with the f/2, which affects general rendition. This is true and if you are, for example, photographing flowers at very close distance and want to melt the background as much as possible for aesthetic reasons, then the Batis 2/40 CF is limiting (not only because of aperture mechanism, but also for its focal length). For this kind of aesthetics you preferably would want a lens that has a longer focal length for better perspective flattening and stays fully open at widest possible aperture (cue in the Milvus 2/100M Makro-Planar).
All in all, there's no denying that the aperture mechanism of the Batis 2/40 CF brings some limitations at close focus distances. This is a design choice, probably to ensure high optical performance and to keep lens from growing any bigger. One has to ask if these limitations have effect on the context of his own photography. Personally for me, there is no issue, as I'm willing to exchange two stops of light for better image quality and I don't even see a very big differences in aesthetics for that matter – at close focus distances the background gets melted enough to provide subject separation and more. Nonagones, when they exist, are rare sightings – and to be honest, there are far more important things pursue in photography than the shape of bokeh balls (like 'emotion', 'decisive moment', 'documentary values', 'storytelling' etc.). In short, I think it's a good design choice as getting close to all kinds of subjects with an outstanding optical performance has definitely more value than what the limitations of this design take away from it. For some, it can of course be an issue, but it depends on what you are trying to achieve.