Will MF resolution win over 35mm stabilisation?

[BJL]

... for landscape users without the most stable tripods it's not at all obvious that an MF back will be able to hold more rez than a stabilised 35mm camera of this new generation.

Clearly, the stabilization solution for a tripod-using DMF user whose current tripod is not good enough is to get a better tripod, not to change to a completely different system and format size.  Meanwhile, Pentax is already rolling out lens-based IS for its "645" system, and for SLRs, lens-based does have the advantage of giving A stabilized image in the optical viewfinder.

Maybe the place for further progress in stabilization for the ">35mm" formats will come when someone offer a mirrorless system in such a format.  If that happens, my guess is Pentax with its particular long-standing interest in providing bigger-than-35mm format cameras of maximum portability and hand-holdability, and its existing know-how in stabilization.  More so since it is probably only the smaller medium formats like 44x33mm that will see much handheld use in the future, which is where in-camera IS is relevant.  Leica should jump on this, given its tradition of small, agile mirrorless cameras and recent adoption of CMOS sensors, but I doubt it will.

[EricWHiss]

I figure if they can make MF backs with multishot, then they can also make IBIS work with those sensors too.   It's interesting to me that the new MFT cameras are adding multishot after IBIS.

Phase one hopefully will incorporate this in the future as they already have the gyros in their XF body now. Of course their viewfinder may still be lacking but IBIS would be a nice thing to have.

[bjanes]

The first real A7RII images are trickling in on dpreview. They are pretty detailed. I think there is going to be a good market in high-rez lenses. Also, for landscape users without the most stable tripods it's not at all obvious that an MF back will be able to hold more rez than a stabilised 35mm camera of this new generation.

I think MF needs new tech. The gap is narrowing more and more.

Edmund

To obtain maximum image detail with high resolution 35 mm sSLR and MFDB  cameras, one must use a good tripod and mirror lock up (excepting mirrorless cameras). In this situation, image stabilization is not relevant. When focusing via the reflex mirror, alignment issues between sensor and focusing screen/autofocus sensors may result in defocus of the captured image. For MFDB, this is discussed by Joseph Holmes here and here.

Since live view focusing is done directly from the image projected on the sensor, many of the alignment issues are avoided and live view focusing is generally more accurate than autofocus or manual focus achieved via the autofocus sensor located off the image sensor somewhere in the camera body or via the focusing screen. However, workable live view currently requires a CMOS sensor and the highest resolution MFDBs are CCD and live view is not available.

Shutter vibrations also cause loss of image quality. Generally, leaf shutters have less vibration than focal plane shutters, so MFDB cameras using leaf shutters have an advantage over 35 mm interchangeable lens dSLRs which almost always use focal plane shutters. For MFDBs using a focal plane shutter, the larger size and mass of the focal plane shutter would be a disadvantage. Electronic first curtain shutters available in some 35 mm eliminates shutter vibration.

With these considerations in mind, it is not clear that an 80 MP MFDB in the field would be sharper than Canon 50 MP of Nikon 36 MP. The 50 MP MFDBs using the Sony CMOS should have a sharpness advantage over 35 mm under the conditions discussed above.

Regards,

Bill

[BJL]

I figure if they can make MF backs with multishot, then they can also make IBIS work with those sensors too.   It's interesting to me that the new MFT cameras are adding multishot after IBIS.

Maybe: but a big difference between multi-shot and IBIS is that multi-shot only requires a couple of slow movements of the sensor, all of them along simple vertical and horizontal steps planned in advance, whereas IBIS requires numerous rapid movements in all directions with very low lag from detection of the camera's movements to the compensating sensor movement.  That rapid response is where the extra acceleration forces required for a larger, heavier sensor come in.  I am sure it it technologically feasible; the question is just how bulky the required servo-motor system would be, and if the demand for hand-holdability is enough to make it economically worthwhile.

(EDIT: dumb idea about passive maglev stabilization deleted!)

[Aphoto]

Here is a 42MP Sample from the A7rII + Zeiss 55mm (900$ – as you can see, you don't always need an Otus):

http://we.tl/HqvHylmbu4

Pretty close to a 40MP/50MP back, I think (had myself a 40MP back for 2 years, and don't really see a difference).

And at f/1,8 (!):

http://we.tl/vHXgqLDVer

Is MF really better? (certainly not at f1,8  )

[Aphoto]

Here is a 42MP Sample from the A7rII + Zeiss 55mm (900$ – as you can see, you don't always need an Otus):

http://we.tl/HqvHylmbu4

Pretty close to a 40MP/50MP back, I think (had myself a 40MP back for 2 years, and don't really see a difference).

And at f/1,8 (!):

http://we.tl/vHXgqLDVer

Is MF really better? (certainly not at f1,8  )

to be correct:
of course, MF is or can be better with the right lenses – especially in terms of resolution, but what I really wanted to say: the gap is pretty thin, and even sometimes, the much cheaper 35mm wins.
That is, why I don't regret the change to 35mm (except for the feeling when I take a picture. It is not photographing any more, it is more taking pictures with a tiny electronic device).

[hjulenissen]

Is MF really better? (certainly not at f1,8  )

MFDB: 44 33   (0.77 crop)
FF:      36 24   (1 crop)
APS-C: 22.2 14.8 (1.6 crop)

36x24mm 55mm f/1.8 ISO 100, when comparing with 44x33mm should (in my view) should use a 71mm lens at f/2.33 and ISO 60.

http://www.josephjamesphotography.com/equivalence/

-h

[Bo Dez]

Here is a 42MP Sample from the A7rII + Zeiss 55mm (900$ – as you can see, you don't always need an Otus):

http://we.tl/HqvHylmbu4

Pretty close to a 40MP/50MP back, I think (had myself a 40MP back for 2 years, and don't really see a difference).

And at f/1,8 (!):

http://we.tl/vHXgqLDVer

Is MF really better? (certainly not at f1,8  )

Since you asked, relatively, these may be decent by 35mm standards but by medium format standards I think they are pretty ropey.

Couldn't find the sharp point in the first one. The colour and tone is flat, colour depth seems wanting in some areas, the fringing and moire seem on the excessive side. Some of the aliasing is horrendous.

The 1.8 shot has enough spherical aberration to drive a truck through. That lens looks terrible, IMO, sorry.

Yes MF really is MUCH better. Even your wide open argument, at f2, the 100-S Summicron will blow you away.

[landscapephoto]

Here is a 42MP Sample from the A7rII + Zeiss 55mm (900$ – as you can see, you don't always need an Otus):

http://we.tl/HqvHylmbu4

Pretty close to a 40MP/50MP back, I think (had myself a 40MP back for 2 years, and don't really see a difference).

It is indeed pretty good. There may be a bit more flare around reflection than what I get on MF and we don't know how the image would hold it if were processed to increase saturation a bit, but it is pretty good.

And at f/1,8 (!):

http://we.tl/vHXgqLDVer

For f/1.8, this is great. That lens is a success.

[BJL]

36x24mm 55mm f/1.8 ISO 100, when comparing with 44x33mm should (in my view) should use a 71mm lens at f/2.33 and ISO 60.

I agree about comparing using the larger format at about f=71mm, f/2.33, so as to be covering the same angular field of view and same effective aperture diameter, and thus having diffraction and OOF effects scale in proportion to image size.  But unless the subject is moving, I would be happy for each camera to be used at whatever shutter speed and ISO sensitivity works best for that camera.

One hazard of comparing images taken at equal aperture ratio with different formats and different focal lengths is that the resulting DOF change could artificially make the smaller format image look "flatter" and the larger format image more "three dimensional", to quote some words that often arise in such comparisons.

On the other hand, the ability of a larger format to get the same DOF, OOF and diffraction effects (when viewing the resulting images at equal apparent size) at a higher f-stop could give an advantage in optical performance through lower aberrations.

[NancyP]

Small sensor user, geekish amateur, here. My photos don't rate a MF sensor yet.

Do any of the MF sensors have the sensor shift technology? For instance, Olympus http://www.techradar.com/us/news/photography-video-capture/cameras/crazy-but-true-olympus-sensor-captures-details-between-the-pixels-1283486

The thought is, one of the serious problems about chip manufacture is scale-up - fewer large sensors per wafer, more risk of defect per large sensor, fewer large sensors sold = considerably higher price for extra-large sensor. Plus, design costs for extra-large sensor. Probably the design costs for implementing a sensor shift technology in a MF camera would be lower than getting more pixels. I suppose that for tripod-based architectural and product photography, a sensor shift technology would be fine. Turn it off if you want to shoot weddings or fashion.

[landscapephoto]

Do any of the MF sensors have the sensor shift technology?

That technology appeared in MF first, actually. At present, you can get it from Hasselblad and Sinar.

[eronald]

That technology appeared in MF first, actually. At present, you can get it from Hasselblad and Sinar.

AFAIK this "officially" originated with Jenoptik.

Edmund

[ErikKaffehr]

Hi,

Sensor shift has been around for long in MFD. It is not used as a cost saver, the main advantage is that it avoids colour interpolation. The sensor is moved a pixel width between exposures. So for ecah point, four exposures are made using each of the RGBG filters.

On some cameras it is possible to make exposures with half pixel steps. In this case resolution is increased, at least if the pixels are small fill factor.

The technology does make the sensor cheaper, as sensor size is not affected.

Best regards
Erik

Small sensor user, geekish amateur, here. My photos don't rate a MF sensor yet.

Do any of the MF sensors have the sensor shift technology? For instance, Olympus http://www.techradar.com/us/news/photography-video-capture/cameras/crazy-but-true-olympus-sensor-captures-details-between-the-pixels-1283486

The thought is, one of the serious problems about chip manufacture is scale-up - fewer large sensors per wafer, more risk of defect per large sensor, fewer large sensors sold = considerably higher price for extra-large sensor. Plus, design costs for extra-large sensor. Probably the design costs for implementing a sensor shift technology in a MF camera would be lower than getting more pixels. I suppose that for tripod-based architectural and product photography, a sensor shift technology would be fine. Turn it off if you want to shoot weddings or fashion.