A7RII vs a 20-40mPixel medium format

[vjbelle]

which one can actually mount on 35mm dSLM platforms, no ?

That's not the point.  If someone wants to make a mount then probably they all could mount on a DSLM..... all it takes is a mount.  My comment was about overall lens quality.  I could care less that they don't mount on a DSLM.  They will, of course, mount on an Actus Mini which would be my first choice. 

Victor

[AlterEgo]

That's not the point.

the point is that one can bring MF optics to 35mm dDSLM - so you can consider dSLMs as having that optics... so optics quality will be there (or will that optics be able to handle the higher sensel density of smaller sensors ?).

[ErikKaffehr]

Hi,

If you put two lenses of equal quality, say diffraction limited at f/5.6 on two cameras of different formats, the larger format will always win as long as we can use the optimum aperture, as magnification will be smaller.

DoF requirements may force smaller aperture on the larger format, so MTF would be lost to diffraction.

Lets see it this way, the image diameter on the Sony A7rII is 43 mm, while on the P45+ it is about 60 mm. So the A7rII needs to be magnified 60/43 = 1.40x compared the P45+. This essentially means that if we assume that the P45+ lens achieves 60% MTF at 40 lp/mm than the Sony lens would need to achieve 60% MTF at 56 lp/mm. Or we could guess that the lens on the A7rII would need to achieve 84 %MTF at 40 lp/mm. The Otus at optimum aperture would be somewhere there. So it would need an Otus to match a Planar 100/3.5 on a P45+ back, and that Otus should not be stopped down f/5.6.

Best regards
Erik

A brief comment regarding both platforms..... I can't stress enough that there is nothing in the 35mm world that can compare with either the Schneider Digitars Or Rody's.  The differences between the two platforms all boils down to lenses.  At 100% pixels there is a visual difference between the my Otus and my Schneider Digitar.  Even at just 1 f stop apart the differences remain.  There is a clarity that exists with the digitar that is very pleasing.  In print it would be almost impossible to see but it is there.  Call it the MF look or whatever you want but when manufacturers finally start making slower lenses with very large image circles the gap will start to close. 

Victor

[vjbelle]

Hi,

If you put two lenses of equal quality, say diffraction limited at f/5.6 on two cameras of different formats, the larger format will always win as long as we can use the optimum aperture, as magnification will be smaller.

DoF requirements may force smaller aperture on the larger format, so MTF would be lost to diffraction.

Lets see it this way, the image diameter on the Sony A7rII is 43 mm, while on the P45+ it is about 60 mm. So the A7rII needs to be magnified 60/43 = 1.40x compared the P45+. This essentially means that if we assume that the P45+ lens achieves 60% MTF at 40 lp/mm than the Sony lens would need to achieve 60% MTF at 56 lp/mm. Or we could guess that the lens on the A7rII would need to achieve 84 %MTF at 40 lp/mm. The Otus at optimum aperture would be somewhere there. So it would need an Otus to match a Planar 100/3.5 on a P45+ back, and that Otus should not be stopped down f/5.6.

Best regards
Erik

Erik...... this isn't a matter of sharpness.  I decided on an image where I could easily move either camera so that both matched each other in the horizontal plane.  They are both very sharp.  However the Otus image shows a slight ghosting (for lack of a better explanation).  This ghosting has appeared with any 35mm lens I have tried from Leica to Zeiss to Nikon.  None of this is present with my Digitar's.  And...... I'm only comparing at close to the same MP range.  Even if I were to use my IQ180 back and then inspect at 100% pixels the same level of 'clearness' is there.  None of the ghosting and I'm now asking the lens to work a lot harder.  Just my observations......

Victor

[Torbjörn Tapani]

And you can be sure that there are no internal reflections caused by an adapter or different camera than the Otus was designed for?

[eronald]

Hi,

If you put two lenses of equal quality, say diffraction limited at f/5.6 on two cameras of different formats, the larger format will always win as long as we can use the optimum aperture, as magnification will be smaller.

DoF requirements may force smaller aperture on the larger format, so MTF would be lost to diffraction.

Lets see it this way, the image diameter on the Sony A7rII is 43 mm, while on the P45+ it is about 60 mm. So the A7rII needs to be magnified 60/43 = 1.40x compared the P45+. This essentially means that if we assume that the P45+ lens achieves 60% MTF at 40 lp/mm than the Sony lens would need to achieve 60% MTF at 56 lp/mm. Or we could guess that the lens on the A7rII would need to achieve 84 %MTF at 40 lp/mm. The Otus at optimum aperture would be somewhere there. So it would need an Otus to match a Planar 100/3.5 on a P45+ back, and that Otus should not be stopped down f/5.6.

Best regards
Erik

Erik, I'm not so sure you're right. MTF 50 is an accepted single-number way of measuring, but you cannot scale the MTFs around so easily as the curve is not the same on all lenses. By analogy with circuits, some lenses will drop at I guess 6db per octave or whatever at the frequency you are at, some are still flat etc. Also, MTF gives you the native (hopefully) (Imax-Imin)/(Imax+Imin) at that spatial frequency, but afterwards you can postprocess ...

Mixing reasoning in linear spatial and Fourier db spaces is not always a good idea, at least it seems counterintutive to me with my very rudimentary understanding of mathematics

Edmund

[ErikKaffehr]

Hi Edmund,

I am aware of MTF comparisons being tricky, but please note that I explicitly said "diffraction limited lenses". MTF drops almost linearly with increasing frequency on diffraction.

Post processing is of course a dominant factor, but the more of the MTF is coming natively from the lens the less we need to sharpen.

So very clearly, assuming a linear drop of MTF over frequency is an oversimplification.

Below are MTF plots measured on a Planar 80/2.8 at f/8 and a Sony 16-80/3.5-4.5 ZA at 80 mm and f/8 on 6.8 micron P45+ resp. 3.8 micron Sony Alpha 77 SLT. Near optical center and with no sharpening in Lightroom. Here the curves drop rather nicely. In this case the DSLR is an APS-C camera, so image circle is around 28 mm while on the P45 it is around 30 mm. So and image shot on the P45+ needs half of the magnification of the APS-C camera. So here we see that  MTF at 30 lp/mm is around 0.64 on the APS-C, but for the P45+ we would check the curve at around 15 lp/mm and get an MTF of 0.82, clearly better.



The next figure is with a bit aggressive sharpening. Very clearly, sharpening can twist things around. If we still look at MTF at 30 lp/mm on the APS-C, we would end up at 0.92, clearly better than the unsharpened P45+. Looking at the P45+ at 15 lp/mm it would have an MTF (after sharpening) of 1.01. This would give a visible difference to the advantage of the P45+.  Very clearly, once we apply sharpening things get tricky. But, everything kept constant a larger format should offer an advantage, until we run into the diffraction limit due to DoF requirements.

Erik, I'm not so sure you're right. MTF 50 is an accepted single-number way of measuring, but you cannot scale the MTFs around so easily as the curve is not the same on all lenses. By analogy with circuits, some lenses will drop at I guess 6db per octave or whatever at the frequency you are at, some are still flat etc. Also, MTF gives you the native (hopefully) (Imax-Imin)/(Imax+Imin) at that spatial frequency, but afterwards you can postprocess ...

Mixing reasoning in linear spatial and Fourier db spaces is not always a good idea, at least it seems counterintutive to me with my very rudimentary understanding of mathematics

Edmund

[vjbelle]

And you can be sure that there are no internal reflections caused by an adapter or different camera than the Otus was designed for?

Yes....... the same type/identical image is produced when using an 800E. 

[BernardLanguillier]

Erik...... this isn't a matter of sharpness.  I decided on an image where I could easily move either camera so that both matched each other in the horizontal plane.  They are both very sharp.  However the Otus image shows a slight ghosting (for lack of a better explanation).  This ghosting has appeared with any 35mm lens I have tried from Leica to Zeiss to Nikon.

Victor,

This is interesting. I have been using both Otus extensively on a D810 and have never noticed any ghosting, but I am unsure what you are exactly refering to.

Would you are some examples of that phenomenon? That would help me understand whethet I have been overlooking something.

Thanks.

Cheers,
Bernard

[JohnBrew]

I believe Victor explained his reasoning clear enough. I've done a similar test and find the problem to be one of clarity. I was able to bump the clarity slider to 100% (for the Otus) and then move other sliders (sharpening, luminous noise, etc.) around and was able to realize about 98% of the MF (Alpa/Rodenstock/50 mp back) combo. Then you have to deal with the color issues of DSLR versus MF  .

[BernardLanguillier]

I believe Victor explained his reasoning clear enough. I've done a similar test and find the problem to be one of clarity. I was able to bump the clarity slider to 100% (for the Otus) and then move other sliders (sharpening, luminous noise, etc.) around and was able to realize about 98% of the MF (Alpa/Rodenstock/50 mp back) combo. Then you have to deal with the color issues of DSLR versus MF  .

That I can understand.

Trying Iridient Developper 3 may be interesting to extract the last hits of detail from those files.

Cheers,
Bernard

[vjbelle]

Victor,

This is interesting. I have been using both Otus extensively on a D810 and have never noticed any ghosting, but I am unsure what you are exactly refering to.

Would you are some examples of that phenomenon? That would help me understand whethet I have been overlooking something. 

Thanks.

Cheers,
Bernard

Bernard.... I am not sure if 'Ghosting' is the correct term to use but its as close as I can come to describing the differences between my two files.  As I said in my post I placed both cameras in position so that the horizontal field of view was matched.  To me there is a fairly large difference between the two files.  You can download the files at (https://www.dropbox.com/sh/dtvtz4py065aipl/AAB3YyBtKXhIq0yMGuz_gr5na?dl=0) - if it all works.  I don't do this much.   Note that focus on both images was very careful and was focused on the painted Cardinal and then focus was checked on both sides.  Images were processed in C1 with standard sharpening. 


Victor

[Bart_van_der_Wolf]

I believe Victor explained his reasoning clear enough. I've done a similar test and find the problem to be one of clarity. I was able to bump the clarity slider to 100% (for the Otus) and then move other sliders (sharpening, luminous noise, etc.) around and was able to realize about 98% of the MF (Alpa/Rodenstock/50 mp back) combo.

Hi John,

While you can get a long way with Clarity, it's actually more likely a difference in MTF due to the larger magnification factor on a physically larger sensor.

Clarity in most editors is mostly a mid tone contrast adjustment, possibly with some masking. A much better job can be done with Topaz Clarity, which creates absolutetly no halos, and no color shifts, and additionally allows to separately control several contrast ratios, and adjust the blackpoint, midpoint, and whitepoint of the tone curve.

However, the difference in MTF response is better compensated, or more elaborately enhanced beyond that, with a tool like Topaz Detail. That allows to adjust the modulation of several sizes of detail, exactly what potentially caused the difference in the look of the different sensor sizes. Also completely without risk of any halos from being created, like e.g. many High-pass sharpening techniques.

Then you have to deal with the color issues of DSLR versus MF  .

Which is mostly profiling related.

Cheers,
Bart

[Jack Hogan]

Erik, I'm not so sure you're right. MTF 50 is an accepted single-number way of measuring, but you cannot scale the MTFs around so easily as the curve is not the same on all lenses. By analogy with circuits, some lenses will drop at I guess 6db per octave or whatever at the frequency you are at, some are still flat etc.

Lets see it this way, the image diameter on the Sony A7rII is 43 mm, while on the P45+ it is about 60 mm. So the A7rII needs to be magnified 60/43 = 1.40x compared the P45+. This essentially means that if we assume that the P45+ lens achieves 60% MTF at 40 lp/mm than the Sony lens would need to achieve 60% MTF at 56 lp/mm.

Edmund, Erik is in my opinion formally correct: the linear spatial resolution recorded in the raw data at a given MTF needs to scale inversely with linear sensor size in order for two different formats to show the same 'sharpness' in lp/ph when the photographs are displayed at the same size. It's all about magnification: http://www.strollswithmydog.com/equivalence-sharpness-spatial-resolution/.

60%MTF is 60%MTF independently of format.  At that MTF the P45+ in the example above would show 40lp/mm*60mm/sensor diag = 2400 lp/picture diagonal (equivalent to the lp/ph units that we are more used to seeing).  At 56lp/mm*43mm/sensor diag the a7RII would also show 2400 lp/picture diagonal.  So MTF60 would appear equally 'sharp' to an observer viewing both images at the same size.

Sharpening is a different can of worms.

Jack

[vjbelle]

Victor,

This is interesting. I have been using both Otus extensively on a D810 and have never noticed any ghosting, but I am unsure what you are exactly refering to.


Cheers,
Bernard

Bernard..... a few brief comments on the legendary Zeiss QC.  This is my third copy of the Otus 55mm.... the first two were out of alignment.  Even this one is slightly skewed but probably the best of the bunch.  I also purchased the Otus 85mm.  The first thing I do with any lens is shoot a chart.....  it doesn't tell everything but its a good starting point.  I couldn't believe what I was seeing with the 85mm.  An entire top edge out of focus, corners that were out of wack, one sharp another blured.  I have found that all of these anomalies extend to everyday shooting.  All of this for an outlay of $4500.00.  I owned that lens for 2 hours, packed it up and sent it back.  I was so pissed/disappointed that I didn't even want a replacement..... just a refund.  This experience goes hand in hand with my quest for a 135mm.  I had two copies, one right after another with both showing the same severe alignment issues.  Needless to say I don't own a Zeiss 135mm.  If they want the big bucks then they should provide perfect lenses - or at least a lot better than I was presented with. 

Victor 

[eronald]

Bernard..... a few brief comments on the legendary Zeiss QC.  This is my third copy of the Otus 55mm.... the first two were out of alignment.  Even this one is slightly skewed but probably the best of the bunch.  I also purchased the Otus 85mm.  The first thing I do with any lens is shoot a chart.....  it doesn't tell everything but its a good starting point.  I couldn't believe what I was seeing with the 85mm.  An entire top edge out of focus, corners that were out of wack, one sharp another blured.  I have found that all of these anomalies extend to everyday shooting.  All of this for an outlay of $4500.00.  I owned that lens for 2 hours, packed it up and sent it back.  I was so pissed/disappointed that I didn't even want a replacement..... just a refund.  This experience goes hand in hand with my quest for a 135mm.  I had two copies, one right after another with both showing the same severe alignment issues.  Needless to say I don't own a Zeiss 135mm.  If they want the big bucks then they should provide perfect lenses - or at least a lot better than I was presented with. 

Victor

Zeiss is not Zeiss, SK is more of a label than a brand, I think the only still lenses with real QC any more are -maybe- Leica, and the high-end Canon teles. Lenses made for cinematographers get all the love these days.

 I expect Canon and Nikon have made special "instrument" sensor chips which allow for on-camera lens measurement, but they probably keep those for their own use.

Things may change with mirrorless AF lenses, as the phase detectors on the sensor can probably help do the QC on the lens, at the point of manufacture.

Edmund

[JohnBrew]

Bart, thank you for your response. I'm sure the third party plug-in you mention will work wonders. I pushed the comparison file around every which way, just to experiment, and what I relate is what worked. But trying to show this in a jpeg on the internet is fruitless. The fact I am now able to get so close to a MF file with my DSLR is fairly impressive in my view. Too bad my other lenses are not up to the resolving power of the Otus. And in case you are wondering, a loupe at 1" did confirm the Rodenstock to be sharper, but from any other distance the difference was not discernible.

But to keep this thread on track, I think cramming more and more pixels into a FF sensor, is going to present it's own different problems. I have always found color fidelity to be better when larger pixels are at play. But the boffins at Sony are truly working miracles these days and the parameters of sensor design are being pushed in new ways. If I ever advance beyond what my D810 is capable of, I want what the MF people call a FF for MF digital back which hasn't been produced by anyone just yet. OTOH, an Aptus with Sony A7RII certainly has it's appeals  .