Hi,
There are lot of mirrorless options in 4/3, APS-C and 24x36. There are a lot of mirrorless options in MFD, too, they are called technical cameras. But technical cameras could do with a lot of more integration. So, I suggest some of the MFD manufacturers should go and build a real good mirrorless MFD camera.
Just to say, there is not a lot of wrong with flipping mirror cameras. But recent innovations from smaller sensors may migrate to larger sensors.
First we need to ask our selves, why do we need that flipping mirror? It was invented to implement "What You See is What You Get" viewing back around 1936. At that time there was a debate between rangefinder shooters and SLR shooters. Rangefinders had a lot of benefits, but SLR went trough a rapid development and came out on top. Still, classical Leica M cameras still had a lot of attraction.
First time I had an amateur video camera in my hand I realised something. It had an electronic viewfinder. That should also work for still cameras.
Now, as mentioned before, original SLR designs had a few issues, but those have been ironed out by nearly 80 years of development.
- Vibration. The mirror flips out of the optical path before exposure, It needs to move fast to reduce shutter delay. That movement causes vibration on the camera. It cannot be dampened out entirely, due to conservation of momentum. (*)
- Long focal plane to flange distance that leads to the need of retrofocus lenses.
- Shutter delay. The mirror needs to flip between two positions, it needs to accelerate, retardate and stop before exposure.
- Many parts need to be in alignment. Bayonet, sensor, mirror, secondary mirror for AF, AF-sensor and focusing screen. Both mirrors are moving.
On smaller sensors we have seen rapid development of new technologies that go hand in hand with mirrorless:
- CMOS sensors with very fast readout this is really needed for contrast detecting auto focus (CDAF)
- On sensor phase detection AF (PDAF) - this eliminates the need of both AF-sensor and AF-adjustment
- Backside illumination - may solve some problems with beam angle sensitivity of sensors as well depth is reduced
- Some small size sensors offer thin photodiode layer and have optical insulation between adjacent cells.
- The new 100 MP sensor from Sony support electronic first shutter curtain.
So, it is quite feasible that next generation of MFD sensors will be better suited to non DSLR designs.
A mirrorless camera can still use an existing lens program with an adapter that is essentially an extension tube. Sony, who make the only 24x36 mirrorless aside Leica have two adapters for the A7-series. One is just an extension tube with electrical contacts and mechanical aperture control, the other contains an AF motor and a AF-module.
Hasselblad has something called the HTS adapter. It can give T&S with most Hasselblad lenses, but it has a built in 1.5X converter needed to make place of the flipping mirror. With a mirrorless design the 1.5X converter would not be needed. It would also be possible to use Canons T&S lenses, as demonstrated by both the Hartblei HCam B1 and the ALPA FPS.
The ALPA FPS and the HCam are forerunners in mirrorless MFD. Both have computer controlled focal plane shutters and aperture control for Canon lenses. The FPS can also control aperture on Contax 645 lenses with an adapter, I think.
The ALPA FPS has some advanced optical viewfinder for viewing and a very long calibrated focusing gear for focus. Best way to focus seems to be a laser rangefinder. The HCam B1 has a motorised sliding back.
Both the FPS and the HCam work well with Canon lenses but Rodenstocks and Schneiders are not fully integrated. A mirrorless camera would need to have full integration.
Live view essentially solves both "What You See is What You Get" viewing and gives optimal focusing capability with actual pixels magnification that ideally can be positioned anywhere.
With next generation sensor we may have on sensor PDAF coming to MFD. If it is designed for mirrorless operation it would work better with non retrofocus wide angles. But that will not happen as long as flipping mirror cameras dominate.
Best regards
Erik
(*) Conservation of momentum would make it possible to flip the mirror as retarding the mirror would consume the momentum created by accelerating it. But in a coupled system with tripod and lens the forces will cause complex oscillations.