New kit - Collecting evidence Part 1 - Invertebrates

 In this post I mentioned that I was considering three new pieces of kit: a Sony A7rii, an Olympus OMD E M1 III and a Panasonic Leica 50-200. I came to two conclusions:

• For invertebrates, I need to do some practical comparisons in contexts that are more typical than I've tested before, switching between the A7ii and one of my close-up lens setups. If the success rate with the A7ii turns out to be extremely low for my typical subject matter compared to the close-up lens setups then it doesn't matter how much better the images are; potentially nice images that you can't actually capture are not much use!
• For flowers etc I need to do some practical comparisons between the 60mm macro and the 45-175. 

I'll talk about invertebrates in this post. I have done a flower exercise that I'll talk about separately.

I have now had my first session this year with invertebrates, spending two hours in the church grounds opposite our house. Rather than trying to do like for like comparisons between the A7ii setup and one of my close-up lens setups I just used the A7ii. 

I captured around 550 images and ended up keeping 92 of them, which are in this album at Flickr. That was a keeper rate of around 17%, which is quite high for my stuff. That is a bit misleading though because there were two very long sequences, one sequence of 14 images of a woodlouse wandering around and the other, much longer (52 images) of a fly grooming (or possibly flies - not sure it was all the same fly). The woodlouse shots were difficult, so I would expect a low keeper rate for them, but the fly grooming shots were easy, as the fly stayed in the same place for extended periods as I captured shot after shot, so I would expect a higher keeper rate for them. 

Taking the two long sequences out of the equation gives 26 keepers from 228 shots, a keeper rate of around 11%. Keeper rates can vary a lot from session to session, but 11% is comparable with what I would often get with my close-up lens setups. It provides no suggestion of the low keeper rates that I was fearing for my normal subject matter. 

I had also been concerned about whether the flash would be powerful enough for larger subjects. That is because of the need to throw extra light on the scene because of the very small apertures I'm using with the A7ii, coupled with the longer working distances for larger subjects captured at lower magnifications. There were two potential issues with longer working distances: because of the differences in working distance between the typically 1X or so and 8X magnifications that I would be using, it seemed highly likely that the flash heads would need adjustment to keep them pointing in a suitable direction; and there would be severe  (inverse square) drop-off in illumination hitting the subject as the working distance increased. 

As it turned out, that was not a problem. One reason for this was that I was more relaxed about letting the ISO go higher than I had previously been comfortable with. It struck me that ISO 3200 on the A7ii is very similar to base ISO of 100 with my bridge camera close-up lens setups or the ISO 800 that I typically used with my micro four thirds setups. So up to ISO 3200 with the A7ii I would be no worse off than previously. As it turned out, for the 92 keepers only 3 of them were higher than ISO 3200 (ISO 5000), and 70 of the them were lower than ISO 3200.

Another concern had been whether the flash diffusers would get in the way more than with the close-up lens setups because they were mounted a bit further forward so as to try to reduce the illumination drop-off. This too turned out to be unproblematic. 

A recurring issue turned up, again - very ugly reflections of the flashes on reflective surfaces. This has been troubling me for years; I have never found an arrangement that would rid me of this. So, I have learnt to live with it. However, I suspected it was a bit worse with the A7ii setup. I noticed that the outer diffusion layer was very battered and had developed folds which made it impossible to keep it at a nice distance from the flash heads across the whole width of the setup. I wondered whether that was causing a problem so I made a new front layer, which might or might not make a difference. 

With the new front layer in place I recalibrated the setup for colour cast. I captured an image of the grey panel on a ColorChecker Passport, loaded that into PhotoLab and used the eye-dropper to set the white balance. I then put those values for Temp and Tint into the PhotoLab preset I use for flash-based A7ii invertebrate captures.

Given that I seem incapable of providing illumination which removes the flash reflection issue I've resigned myself to making greater use of cloning to mitigate the problem. It often isn't practical to completely remove the reflections, but they can be much improved by using a mixture of full opacity cloning where practical (problem disappears if done well, needed particularly where an area is completely blown out, which unfortunately does happen sometimes) and low opacity cloning where full opacity cloning isn't practical (so as to tone down the highlights with a light wash of a suitable colour). The wasp at images 10 to 13 of the session's Flickr album are examples of the sort of subject for which this is a problem, and on which I used cloning, low opacity cloning in this case.

Given the "Great shot" culture on photography forums one needs to be careful about interpreting the response to one's photos, but I posted some of the images on three sites and the responses are consistent with my view of the images, that they seem very acceptable to my eye. Particularly telling for me are the references to good detail and large depth of field. There is of course no fine detail in the images because of the diffraction from the tiny apertures, but the level of detail looks no worse to me than with my close-up lens setups, and it might be a bit better in some cases, particularly for the higher magnification shots. As to depth of field, this looks significantly better to me at the higher magnifications. 

As an experiment I had decided to use a fixed f-number for the session, namely f/45. The effective f-number therefore varied as I changed magnification, getting larger as the magnification increased. This seemed to work well enough at the higher magnifications. At the lowest magnifications it was difficult to convince myself there was much if any significant increase in depth of field compared to my close-up lens setups. On my next outing I will try varying the f-number for some lower magnification shots so as to see if I can increase the depth of field while retaining adequate detail. 

One issue that was surprisingly problematic was dust spots on the sensor. I had put a lot of effort into cleaning the sensor and had got it down to just two spots. However, even though I had not since then opened the camera up there were half a dozen or more additional spots to deal with. I think there must be dust in the camera that gets moved around as I used it. I have even noticed extra dust spots appear on the sensor between successive test shots during the course of sensor cleaning. (I take three shots after each cleaning attempt, with the camera kept in the same position for each shot. It is between two of three such shots that I have seen an additional spot appear.)

I have done what I can to clean the sensor, again, but this can be frustrating, with spots increasing as well as reducing in number between cleaning attempts. I have resigned myself to just getting the number of spots down to only a handful and living with that. (I don't have much choice.) One thing that has made this much less aggravating and time-consuming is my discovery of "spot visualisation" in Lightroom. It makes dealing with spots far easier and faster, with a much lower probability of missing spots, which can be really annoying as, depending on when I notice them, dealing with them can mean backtracking not just through elements of the processing workflow but, more time-consuming, also updating the primary image set on my PC, two backup sets (one on the PC and one in the cloud) and a Flickr album. 

Where next with invertebrates?

More of the same I think with the A7ii, with some f-number variation. I'll try to go to one of the local nature reserves to see if I can find some more varied subjects and shooting situations.

So far it is going well enough with the A7ii that I'm not inclined towards the Olympus E-M1 III or the Sony A7rii at this stage. I'm going to put those possibilities on hold for now.

Here are a few of the images from the latest session.

I didn't achieve close-in shots like these three with my close-up lens setups. Partly that may be because I've only recently got more into this sort of shot, but mainly I think because I probably couldn't achieve this sort of result with my close-up lens setups because of not being able to get sufficient depth of field for my taste.

I'm also having more success with small subjects than with my close-up lens setups. The next one was the only small subject I found during the session, and I didn't actually realise it was a subject. I did a quick shot and then moved on because at a cursory glance I thought it was just a piece of detritus on the leaf. I suspect this was  actually based on my impression/assumption before taking the shot rather than really looking at the image I had captured. Especially given the single and rather inattentive attempt, I'm happy enough with the outcome.

On the other hand, these whole body shots of medium-sized subjects look rather similar to me to what I got with my close-up lens setups. These are the sort of shots for which I intend to try out smaller apertures.