I remember it well. The niggling feeling at the back of my mind that there’s a whole fleet of cameras with bigger, badder sensors out there, and if I had just one of those beasties in my hands, I’d magically be able to make better work. Sound familiar? The itch was so severe that I’ve moved from 1/1.63” (LX5), to Micro 4/3 (Olympus Pen), APS-C (X100, X-Pro1), Full Frame (Nikon Df, Pentax K1) and reaching the final destination of Medium Format (GFX50R)… albeit mini-MF when compared to film sizes. This week let's consider this desire and examine the value attached to going bigger… or, for that matter, smaller. Does it actually, truly, make much difference?
Babbacombe Downs 4
I’m going to level with you: I’m not the most technically minded person, so if you want a maths lecture, you’ll need to go elsewhere. Furthermore, this is far from the first time this question has been looked at… there are so many comparisons, often with examples, which I won’t be doing here. It’s a reasonable argument that they will do a much better job than me of reaching a valuable conclusion for you. So why am I even bothering to add to the din? My problem with the big photography review websites, that ‘leave no stone unturned’, is that they often try to consider every aspect from every photographic perspective… sports, wildlife, landscape shooting, video, the lot. All have very different needs but none of them seem to truly align with mine. I’m guessing as you’re here, then your work might have similar requirements to my output. Most importantly, the (sometimes) rigorous testing might scrutinise test-charts under controlled lighting or a portrait that takes into account the differences in focal length equivalency. What they rarely do to my satisfaction is consider the topic, any topic for that matter, in real-world applications. We learn to love, loathe, or something in-between, a lens or camera by day-to-day usage, over a period of time. So that’s what you’ll get from me here, a more subjective analysis based on my long-term experience with each format.
As with many things in life, when we think of a singular topic in photography they cannot be considered in isolation. There are a number of variables related to sensor size that can all impact on the final image output - resolution, equivalence, sensor generation, mirrorless vs DSLR, SNR, pixel density… the list goes on. To level some of this playing field a little, I’ve noted in recent years that the rate of sensor advances has slowed significantly. When I first set off on my photographic journey over a decade ago, each generation of chip would leap-frog the previous cohort with a notably sizeable increase in ISO performance, resolution, SNR and the rest. In more recent times, camera manufacturers have even taken to releasing an upgraded model with the exact same sensor on board. The net effect of all of the above is that it’s nigh on impossible to get an absolute answer to the original question… but let's make a stab at it. Everyone loves a stabber trier.
Saunton Sands 1
With all these variables at play, it is important for me to define my subjectivity further. I almost always shoot at base ISO, 99 times out of 100. If you regularly shoot in the upper limits of your ISO, this discussion probably won’t help you much :( In truth, I have no idea what my GFX50R is like at ISO 3200 or even 1600… genuinely, I’ve never gone there. Until recently, I would always shoot on a tripod, but I’m still working at base ISO when handheld, even at night. This means I have a good understanding what each sensor is like with ideal settings, but have no idea how they perform if stressed a little, at least at the time the image is captured. This last disclaimer is because I regularly push my images by 2 or 3 stops in post-production. In that regard I’m very demanding of a clean RAW file to tinker with. If these are the kind of parameters that you work in, then this deliberation is more likely to be of value to you.
By now I can hear you out there saying, ‘Get to the point Mr Feltham, no more blah blah…’ So, first key question: Have I noticed improvements with every sensor size increase? Yes. Yes, I have. If thats what you wanted to read to encourage you to fork out for the next step up, then great. Enjoy your new camera :) Of course more needs to be said here. Resolution has limited the size I could print my work in the past, so that is one reason why I moved on from the Nikon Df (16MP) to the Pentax K1 (36MP). I reviewed some Df images recently, and there was some noise visible in the shadows at base ISO. This is something I haven’t noticed in K1 or GFX files. Manufactured by Sony, they performed miracles with the 36MP chip… it actually seems to bend the laws of physics. That the higher resolution sensor produces cleaner files than a 16MP chip is magic indeed. The high-res sensor was originally released in 2012 (!!) in the D800. Game. Changer. On reflection, it was for me as well, as the Pentax K1 was the point at which the resulting RAW files cleared the ‘good enough’ bar, and by some margin. However it must be stated that it was the increased resolution and performance, as well as the size, that had a part to play here.
I suspect there will be a number of you lusting after the Medium Format GFX sensor, so want some kind of analysis as to the value that this increase provided. There was a clear improvement over the K1 - and a big improvement when compared to the Df. I would estimate I get about a stop extra of dynamic range when comparing the GFX with the K1. I get similar to the GFX output when employing pixel-shift (4 exposures combined) in the Pentax, but there are limitations to using this function. Lots of folk have mentioned the ‘creaminess’ that I get from the GFX output and I see it too… I think what’s being observed is an increase in the total amount of light (and therefore colour) recorded by the larger sensor, and probably an extra boost from the larger individual pixel wells. I must admit that this is the point at which I’m crashing into the ceiling of my physics knowledge, and with some force. Nonetheless I can confidently estimate I was also getting about a stop advantage every time I moved up through the other formats, M4/3 > APS-C > FF. If you’ve moved up through the ranks, has this been your experience too? More importantly, do you think it’s been worthwhile?
Abbey Park 4
Drawing the discussion to the close, the GFX system is a steep price to pay if you consider you can get a refurbished Pentax K1 or D810 for well under £1000… and take home 95% of the image quality. If you’re on a budget (who isn’t?), this is absolutely where I’d recommend you put your hard-earned cash. You’ll get a cleaner final image when compared to the smaller formats but for a similar price-point. However, I don’t regret buying the medium format rig as I push my files so much in post-production. Because of this, to put it simply, I can create work that I couldn’t with any other gear. If you don’t push your files as aggressively as me, I’d recommend the FF high resolution options if you want large prints. Even taking equivalence into account, FF will also offer faster glass, so if you desire shallow DOF portraits this is again probably the most fruitful path to take. Your pressure points will likely be different to mine, so as always in photography, you have to distill your needs and balance that with what you want / can afford / get approval from significant others without divorce. Although as I’m sure we all recognise, sometimes it's not just about need, is it?