Why the Micro Four Thirds system will fail 6 November 2021

 

Sandstone bridge This is an easy shot to make on a full frame camera with a single Raw capture and a few adjustments in post processing.  The very high subject brightness range would be more difficult to manage with a crop sensor camera. 

I was an early adopter of the Micro Four Thirds system (MFT)  in 2009 for two reasons.

I had become frustrated by Canon’s inability to fit their enthusiast level DSLRs with reliable autofocus and I could see that mirrorless interchangeable lens system cameras (MILC) would supplant digital single lens reflex types in due course.

MILCs have fewer components of which fewer are mechanical in nature. It follows that MILCs will be less expensive to make than DSLRs and less prone to errors resulting from misalignment of mechanical components. The MILC type can also have more accurate autofocus because the AF sensors are right on and are part of the imaging sensor.

For example: The Canon EOS 5D.4 (DSLR) and the  Canon EOS R (MILC) both use the same sensor and  deliver essentially the same image quality. The R is reported by users to have a more accurate and consistent autofocus system.

I did a price check today and found the 5D.4 selling at AUD3454 new retail and the R at AUD2293.

The R is smaller, lighter, has a better AF system and is $1161 less expensive.

MILC wins convincingly on all parameters relevant to the user experience.

The number of DSLRs being offered has declined dramatically over the last 2 years and will no doubt approach zero soon.  Ricoh/Pentax have said they will not try to make the transition to MILC.

Olympus and Panasonic showed considerable  courage over ten years ago by betting on MFT as the way forward.

As it happens they got one thing right: Mirrorless was the way to go.

And one thing wrong: The quarter size Four thirds/ Micro four thirds sensor.  I am just an ordinary consumer with no inside knowledge of the decision making processes of any of the camera makers so I really have no idea why they chose the small sensor.

The choice seems especially puzzling as Olympus had failed in their bid to persuade buyers to embrace their Four Thirds system DSLR cameras over the established Canikon models of the time.

The Micro four thirds system has two main problems, both of which are intractable and cannot be overcome:

1. No matter which way you evaluate it larger sensors can deliver better image quality than smaller ones. New technology will not help smaller sensors because larger sensors will simply utilise the same new tech.

2. In the early days of MFT the main selling point of the system was the small size of bodies and lenses relative to the DSLRs of the time.

But now we have full frame MILCs and that selling point has almost evaporated.

MFT bodies are no longer significantly smaller than full frame MILC bodies across the model range.  The elements which determine the size of a camera body are the monitor screen,  EVF and housing, handle and controls, battery, IBIS unit and heat sink architecture.

The other part of the size equation is the lenses. Understanding this is a bit more difficult due to the equivalence issue. This has in the past been a hot topic on user forums as some people had difficulty understanding the concepts involved.

We need to consider focal length/angle of view equivalence and aperture/depth of field equivalence.

The traditional full frame 24x36mm sensor has a diagonal of  about 43mm and an area of 864 square mm. 

The MFT sensor measures 13x 17.3mm giving a diagonal of about 21.5mm (half that of full frame) and an area of  225 square mm (a quarter of full frame).

They have a different aspect ratio but that does not impact our considerations in any substantial way.

Focal length/angle of view equivalence is easy enough to understand and generates no controversy.

Full frame focal length is equivalent to MFT focal length x2.

Thus a 25mm lens on MFT has about the same angle of view as a 50mm lens on full frame.

But when we come to consider  aperture/depth of field equivalence things get a bit more complex.

Consider the case of any object. It might be an elephant or a car, a boat, whatever. If we double the linear dimension the surface area increases 4 times (2x2) and the volume 8 times (2x2x2).   So when we scale something up or down different attributes increase or decrease at different rates.

Thus a MFT lens 25mm f1.4 does not have the same light gathering capability or depth of field as a full frame lens 50mm f1.4.

It is actually equivalent to a full frame lens with an aperture value two stops smaller, 50mm f2.8.

And this is how the second problem for MFT arises. Compact  MFT lenses are only smaller because they are not actually equivalent to full frame ones with the same f stop.

Here is an example:

You can buy a Canon RF 50mm f1.8 lens for RF mount full frame cameras. It is compact, light,  cheap and very sharp stopped down a notch or two. Retail price in Australia is around AUD 339.

The equivalent MFT lens would be 25mm f0.9. But no such thing exists and if it did it would be huge and cost a mint.

The nearest we have is the Olympus M.Zuiko 25mm f1.2. This is larger and  heavier than the RF Canon  and at AUD 1649 costs five times as much.

We could trawl through the whole catalogue of lenses and make the same discovery over and over.

By the way I have read reviews of that M. Zuiko 25mm f1.2 which tell us what we knew already, which is that the design and construction of ultra wide aperture lenses is difficult and not many of them are sharp wide open. So you spend all that money and still have to close the aperture down to f2.8 for sharpness across the frame. That is equivalent to f5.6 on the Canon RF 50mm f1.8 at which aperture the Canon is super sharp across the frame and has the benefit of sitting in front of a better sensor with more pixels and greater dynamic range.

Two years ago I realised that the end was nigh for Micro Four Thirds and sold off all my MFT gear while I could.

Panasonic has directed almost all its R&D into full frame and Olympus sold off  its imaging division. The rats are leaving the ship.

I  went with Canon RF mount which is working out well with a decent amount  of R&D coming through even with all the present problems arising from Covid-19 and chip supply issues.

What about APSC ?

When I run the numbers it is clear that APSC systems have the same problems as MFT, just not to the same degree. So I guess APSC will linger longer. But I think the camera makers would prefer to rid themselves of APSC to simplify their product catalogues and reduce their fixed costs.

I think APSC cameras are to camera makers very much like compact sedans to car makers: A good thing as long as they can sell boatloads of them.

Once sale volumes decrease, the end is nigh for any kind of product with a thin margin per unit.

Crystal ball ?

I think that camera makers will do their best to entice their customers

a) to my brand, not that other one

b) to a MILC system

c) a full frame system.

They have already gone a considerable way down this path.  Each is spruiking the advantages of their own brand. Even Nikon which has not been looking too strong recently appears to have redeemed itself with the Z9 which is giving the Nikon faithful something to cheer about even though most of them will not actually buy the thing.

Soon we will have the trifecta of top level pro bodies in place. These are the Sony A1, Nikon Z9 and the not yet announced but much rumored Canon R1. The R3 has been announced though and that model is close to top tier.

When these halo models have been settled into place in their respective hierarchies, the makers can turn their energies to the other end of the spectrum in the form of very compact, low cost full frame mirrorless models. These will be attractive to buyers who in the past would have gone for APSC.

The camera makers have signalled their intentions in this market space with  the Sony A7C, Nikon Z5 and Canon RP.

I think it likely that  camera makers want to divest themselves of  all  DSLRs and all kinds of crop sensor models and lenses.

Whether buyers will let them do that as quickly as they might like remains to be seen.

Maybe a few high spec compacts and some superzoom bridge types will survive the cleanout. Some models such as the Panasonic FZ300 have great versatility with a huge zoom range in a light compact package with potentially decent but inconsistent image quality. If something like that model had a better lens and better image quality I would be tempted to use it for travel duty. That is, when we can travel again after the Covid lockdowns in Australia.

We shall see how this all plays out.