The Kit List: Part Two - Why
In a previous blog I outlined the kit that I’m currently using. In this second part I go into some of my thinking (or lack thereof) about why I’ve made these decisions on kit. I’ll say at the outset that all kit choices involve some compromise and when considering the factors that matter, your mileage will vary.
For example, some may well argue that for ultimate image quality you should be using a large format 10x8 film camera. For landscape this may be true, but it’s not a camera that will get you a bird in flight shot nor will any difference be apparent if you only post on social media.
The trick, therefore, is to look at what you shoot today and what you anticipate shooting in the future and list the factors that matter. Even in the same genre, different folk will weigh these factors differently: it’s about minimising the compromises that you need to make and accepting that there are some that you will have to make.
The earlier post I looked at cameras, lenses and accessories. This post really only looks at camera choice - once you’ve selected the body the rest of the decisions follow. The choice of lens or tripod is rarely controversial - the stock advice is to buy the best you can afford - and if you’re prepared to die on a hill over your choice of camera bag, then more fool you… The choice of camera (Canon, Fujifilm, Nikon etc…), sensor (full frame, M43, APS-C) or form factor (DSLR or mirrorless) seem to stir passion and controversy beyond their actual merit. Imagine Apple vs Windows, on steroids and in multiple dimensions…
Ground Rules
Rather than try to cover the whole gamut of camera equipment, I think it’s worthwhile setting a few basic ground rules. Firstly, I’m only interested here in digital imagery. I know there are many folk who enjoy film photography, and I’ve been known to dabble occasionally myself, but for the vast majority of the photography that I do, digital is the obvious choice. (I can shoot quite freely, quickly review what I’ve shot and have full control in the digital darkroom).
I also shoot a wide range of subjects covering a wide range of focal lengths, from 15mm or shorter to over 600mm in full frame terms. While there are bridge camera than can cover a huge zoom range the quality is compromised, so I’m going to restrict myself to interchangeable lens cameras. The wide range of focal lengths also rules out rangefinder cameras (Leica) and medium format, as intriguing as both those options are.
This essentially reduces the field to DSLRs and mirrorless cameras. There’s a risk that some of the following may look like post-purchase rationalisation, but these are the factors that I considered when making the decision to divest myself of my Canon DSLR kit.
Sensors
The sensor is undoubtedly the single most important component in a digital camera and the biggest determinant of image quality in the camera body. Arguably, the lens plays as big a role in determining quality, if not larger, but in the camera itself the sensor is most important consideration. However, the impact that the sensor has on image quality can be, and often is overstated and evaluating which sensor may be best is not a matter of simply comparing one number with another.
The images below were taken with sensors ranging from 12.8MP to 30.4MP in terms of resolution and cover full-frame, APS-C and micro four thirds formats in terms of physical size. The release dates for the sensors also cover around 15 years. All the images were shot at, or close to, the sensor’s base ISO. Despite these material differences, when viewed in a web browser it’s far from obvious which image is taken with which sensor.
In short, I’ve listed the factors that I considered important when looking at the sensor options. If you want a more detailed (and better qualified) examination of the issues, DPReview’s sensor trilogy provides a more technical and detailed explanation.
Sensor Size
The physical size of the sensor matters for a couple of reasons. Firstly, for any given pixel count, smaller sensors will need to pack the pixels more tightly, which has an impact on noise (see later). Secondly, the physics of the optics means that for any given angle of view, larger sensors are better at isolating the subject with shallower depth of field. The converse also applies, smaller sensors are better able to maintain front to back sharpness, but it’s worth noting that the differences are often overstated. Super wide aperture stuff doesn’t feature highly in my work, so this isn’t a major consideration, but if you’re a bokeh monster then full frame is the way to go (more so than medium format, given lens choices available).
Resolution
The kiddie reviews almost always focus on the number of megapixels, but this is a gross over simplification. High megapixel counts are only really needed in two circumstances; if you want to print large or you want to crop tightly. For digital display high pixel counts are wasted and carry a cost in terms of storage and processing power, as well as a penalty in terms of noise. There is a reason why some of the most expensive, professional camera bodies have a relatively low pixel count.
If your display medium is a screen, then you probably don’t need much more than about 16MP (the same as my Apple 5K monitor). Facebook, Instagram and other social media sites typically downsize images to around a couple of megapixels, and club competitions use images of around the same size. So, a 20MP image gives plenty of scope for cropping and reframing.
In terms of print sizes, a 20MP image will produce a good quality image at about 90cm x 60cm, and good interpolation software will probably push that a little further. The actual print size will depend to some extent on the viewing distance, paper, lighting and a whole host of other factors. So, unless you want to print substantially larger than this on a regular basis, an ultra-high megapixel count probably isn’t a prerequisite. After all, photographers seemed to be able to make good quality prints long before the advent of 50MP or 100MP sensors.
Dynamic Range
The dynamic range measures the range of tones that can be recorded between the brightest part of the image and the darkest, without losing detail. The human eye is a fantastic instrument and, in photographic terms, can record a phenomenal 20+ stops of dynamic range. In other words, the detail in a light object that is over one million times brighter than the darkest instruments. The best cameras, by contrast, can only record around 15 stops of dynamic range, roughly 32 times less. Landscape photographers often face scenes with high levels of contrast that challenge even the best of cameras, although other genres can also present problems.
Depending on the genres of photography you purse it might make sense to sacrifice some resolution for dynamic range. It’s also the case that, everything else being equal, larger sensors perform better than smaller sensors.
Noise / Low Light
The final sensor characteristic that needs to be considered is the noise performance and low light performance. As light levels fall and ISO settings rise, then digital noise starts to creep into images, particularly in the shadow areas. Noise performance is also important when shadows need to be selectively lightened. For any given megapixel count, sensors with less densely packed pixels perform better than those with more tightly spaced sensors and newer sensors tend to outperform older ones.
How important this is depends on the types of photography pursued - sports and wildlife photographers, who regularly need to shoot at high ISOs, will value good noise performance more than those who regularly shoot at or close to the camera’s base ISO.
Conclusions
The conclusion that I reached reviewing the options was that I could probably do without a full-frame sensor, although should other factors be equal this would still be the preference. I did come to the conclusion quite quickly that the micro four thirds format wasn’t going to work for me, however. The compromises in terms of subject isolation were not ones I wanted to make. The major concern, however, was around noise performance. When I made my decisions the 20MP M43 sensor was really starting to show it’s age and there wasn’t a clear commitment to development from either of the main players, Panasonic and Olympus (now OM Systems).
My hands-on experience with APS-C (both Canon and Fujifilm) suggested that this format would work for me. However, the sensor is not the only factor at play.
Format - DSLR or Mirrorless
Another topic that seems to spark passions in some beyond anything that’s warranted. For me this was an obvious choice - the reason for divesting myself of my Canon gear was the announcement of the EOS R5 and the fact that the EF mount would become increasingly obsolete. Richard Butler recently wrote with far more eloquence and knowledge than I ever could about the challenges that the manufacturers’ face here.
In short, however, I chose mirrorless not because I have any long-standing prejudice against mirrors, but because that’s where the main manufacturers are investing their R&D dollars. Anyone holding out for a Nikon D7 or a Canon EOS 1Dx Mark iv is likely to be sorely disappointed.
Lens Range
Lenses are as big a determinant of images quality as the image sensor. Partnering the best sensor with a poor quality lens will produce results just as underwhelming as those produced on a lower quality sensor and high quality lens. Arguably, the former will produce the worse result, depending on where you want to use your images.
In the full-frame mirrorless space there is no shortage of high quality lenses from the camera manufacturers as well as the quality third party manufacturers such as Sigma or Zeiss. However, Sony, Nikon and Canon all see APS-C as a gateway drug to full frame and, consequently, none of them have produced a dedicated, high quality lens for crop sensor cameras.
Fujifilm, however, have plotted a different course. Their range of X-Mount lenses are widely acknowledged as being some of the best native APS-C lenses available. For sure, you can mount a full frame lens on a crop sensor camera from Sony, Nikon or Canon but the focal lengths will be weird and, more importantly, they will be unnecessarily large. Crop sensor cameras don’t really have a significant weight or size advantage, but their lenses do and, if you want to carry a decent kit on your back for several miles, that makes a difference.
In my mind, when I started looking at comparing systems rather than simply cameras the Fujifilm X Series started to make an awful lot of sense.
Autofocus
The last factor I needed to consider was autofocus performance. I had extensive experience of Fujifilm’s kit before fully making the switch, so I was confident that it could cope with most of my photographic needs. The two areas where I didn’t have much experience were macro and sports / action. The former is absolutely fine - indeed, the mirrorless system has a number of advantages including some support for focus stacking.
Sports and action was a different matter. I’ve done more of these photographic genres in the last eighteen months and I am enjoying it immensely. However, as I mentioned in the last post, I found the tracking performance on the Fujifilm X-T3 increasingly frustrating. There’s no substitute for speed and power so, with the cash I’d raised from the sale of the Canon gear, I took the plunge and supplemented the Fujifilm with the Sony A9ii and a couple of lenses.
I’m not making a full switch, however. In every other area the Fujifilm either has the upper hand or the differences are relatively small. There’s no doubt in my mind that autofocus is one of the main battlegrounds between the manufacturers and, for anyone interested in sports and action, is as good a reason for upgrading as resolution or noise performance.
Conclusions / Outcome
This isn’t a big reveal given that I’ve already outlined the kit I use in the previous post. The summary, however, is that eighteen months ago I reached the conclusion that for the vast majority of what I shoot the Fujifilm X-Series system is not just a good choice, but has advantages in terms of size, weight and functionality. The swing factor was the availability of high-quality native lenses for the APS-C system.
The Sony A9ii covers the main weakness in the Fujifilm’s system, namely the autofocus system. The choice of the A9ii, however, is not just about the tracking performance (which, admittedly, is superb). The larger sensor, with it’s lower resolution 24MP sensor, produces very clean files at high the ISOs that wildlife and sports photography can require. For me, it’s a specialised piece of kit for a specific set of jobs.
Fujifilm X-H2S
The arrival of the Fujifilm X-H2S is a significant development for Fujifilm. The new processor and stacked sensor potentially puts the autofocus performance on a par with the Sony offering. The addition of subject detection, which works incredibly well, ups the ante a little further.
For the moment, however, I’ll be sticking with the Sony A9ii as my main sports and wildlife camera. It’s early days, but I think the A9ii (just) has the edge in terms of hit rate and the processed files are still cleaner. Noise reduction software reduces the difference, but not yet by enough for the X-H2S to displace the Sony.