Digital back out-resolving the lens

[tmh]

Hi there

I use a Rodenstock 45mm F4.5 Apo-Sironar Digital lens on a tech cam coupled with a 22mp Leaf back. I selected this lens for it's large image circle, perfect for the stitching I do about 80% of the time.

With the big 9 micron pixels of the 22mp Leaf back I'm getting brilliant results and am very happy. I frequently also borrow a DM33 and see no obvious drop in quality with the reduction to approx 7 micron pixel size.

I'm looking into trading up to a P45+ (almost purely for long-exposure capabilities), where the pixels drop to 6.8microns in size.

I can't seem to find the exact MTF charts or lines p/mm details for this lens, the Rodenstock website shows detailed specs for the 35 but not the 45.

This not being the HR lineup of Rodenstock lenses, my question is; is this lens likely to be out-resolved if I move to a back with a higher resolution? And what does the result actually look like - just a bit soft?

Does anyone have any example images of this lens on 39mp (or even higher) back, or a similar resolution non HR Rodenstock lens on a 39mp back?


Thanks in advance

TMH

 

[dougpeterson]

7.0 and 6.8 are round off error from each other.

If you find the results excellent at 7.0 you will find them excellent at 6.8.

 

[goesbang]

Hi there

I use a Rodenstock 45mm F4.5 Apo-Sironar Digital lens on a tech cam coupled with a 22mp Leaf back. I selected this lens for it's large image circle, perfect for the stitching I do about 80% of the time.

With the big 9 micron pixels of the 22mp Leaf back I'm getting brilliant results and am very happy. I frequently also borrow a DM33 and see no obvious drop in quality with the reduction to approx 7 micron pixel size.

I'm looking into trading up to a P45+ (almost purely for long-exposure capabilities), where the pixels drop to 6.8microns in size.

I can't seem to find the exact MTF charts or lines p/mm details for this lens, the Rodenstock website shows detailed specs for the 35 but not the 45.

This not being the HR lineup of Rodenstock lenses, my question is; is this lens likely to be out-resolved if I move to a back with a higher resolution? And what does the result actually look like - just a bit soft?

Does anyone have any example images of this lens on 39mp (or even higher) back, or a similar resolution non HR Rodenstock lens on a 39mp back?


Thanks in advance

TMH

I had the 35mm Apo Sironar Digital through my transition from P25 -P45+ - P65+ -Aptus12 -IQ180.

Like you, I was delighted with the performance of this lens on the P25. Moving to the P45+ showed no appreciable loss of resolution, even though in theory, the back is outresolving the lens. It was not until I moved to the 60MP sensor that the lack of res became noticeable. Even so, I was still able to shoot and deliver work that exceeded my clients' expectations until I was able to afford to upgrade the lens. On the 80MP class sensors, the images are just not up to par, not only because of the distinct lack of resolving power, but also the significant colour cast when trying to use that lovely large image circle.

The 45 Apo Sironar Digital, like the 35, is at its best with sensors in the 9micron class, but still does a very good job on the 6.8's. Given that the IQ improvement going from 22Mp to 39MP is so big, I would suggest you go ahead with the lens you already have. Even the pettiest pixel-peeper will struggle to criticise your images for lack of resolution. Real pictures are not resolution charts anyway.

Keep the lens, upgrade the back and enjoy making beautiful images. That's what it's all about, isn't it? Just keep in mind that when you next upgrade (to 60, 80 or more MP's), your lens will likely need to be upgraded too.

Best of luck!

 

[torger]

It's very personal. I own a 35mm Apo Sironar Digital, and to me the results is not pleasing on my Aptus 75 (7.2 micron). Corners get a bit blurry, and I'm only pleased if corners are (almost) indistinguishable from the center at 100% pixel peep. Oh, it's for sale by the way

 

[tmh]

Thanks for the replies everyone... it seems only with real-world tests can one fully evaluate this sort of thing...


TMH

 

[m21apsh]

Anyone willing to upload some sample?
to show how digital back out-soulation the lens
Many thanks

Jeff

 

[dchew]

Jeff,
It is pretty easy to see for yourself. Just grab any lens and shoot it wide open. Then stop down one stop and shoot it again. If the image looks sharper stopped down, then you just saw a case when the sensor out resolved the lens when it was wide open.

I don't mean to sound glib; I'm trying to make a point. All this stuff is a matter of degrees. Where was the plane of focus? What is acceptable depth of field? What aperture? Corners or center? Which copy of the lens?

I think TMH came to the right conclusion. It is important to do both kinds of testing: flat chart/wall tests and tests on real images. Only then can you get a feeling for how the sharpness of some combination will affect your subjects and shooting style. In this case, the real question is will the sharpness limitations of this lens be enough to make a difference in his images when he moves to a new sensor? Or, will the limitations of the lens negate the benefits of switching to a different sensor?

I think the sharpness discussions are rampant because today's sensors and new lenses are taking us to a whole new level in regards to sharpness compared to scanned film of the equivalent format. We are now nit-picking the differences. Maybe in another 10 years we won't even discuss resolution of a new sensor/lens because we will be used to it. It will be like light meter accuracy. Very little discussion about that when new cameras are released. Why? I think because it has become ubiquitous.

To TMH:
Personally, I don't think you should spend another second worrying about it. Maybe you won't realize 100% of the benefit of the P45+, but it certainly won't be worse than your results today. Plus you get the long exposure benefits!

Ciao,
Dave

 

[jotloob]

I had the 35mm Apo Sironar Digital through my transition from P25 -P45+ - P65+ -Aptus12 -IQ180.

Like you, I was delighted with the performance of this lens on the P25. Moving to the P45+ showed no appreciable loss of resolution, even though in theory, the back is outresolving the lens. It was not until I moved to the 60MP sensor that the lack of res became noticeable. Even so, I was still able to shoot and deliver work that exceeded my clients' expectations until I was able to afford to upgrade the lens. On the 80MP class sensors, the images are just not up to par, not only because of the distinct lack of resolving power, but also the significant colour cast when trying to use that lovely large image circle.

The 45 Apo Sironar Digital, like the 35, is at its best with sensors in the 9micron class, but still does a very good job on the 6.8's. Given that the IQ improvement going from 22Mp to 39MP is so big, I would suggest you go ahead with the lens you already have. Even the pettiest pixel-peeper will struggle to criticise your images for lack of resolution. Real pictures are not resolution charts anyway.

Keep the lens, upgrade the back and enjoy making beautiful images. That's what it's all about, isn't it? Just keep in mind that when you next upgrade (to 60, 80 or more MP's), your lens will likely need to be upgraded too.

Best of luck!

I have now shot my first images with the APO-SIRONAR DIGITAL 4,5/35mm using the CFV-39 .
I am surprised about the image quality , especially after reading such a lot of negative comments about that lens .

Focusing with HPF ring and f=8 , HH .
But what I found is a strong magenta cast which can not be fully removed by PHOCUS in scene calibration . So it requires some time consuming PP in PSCS6 .

Anyone else working with that lens ? And what DB ?

 

[torger]

I got mine sold, and replaced it with the SK Digitar 35mm whose performance I'm much happier with. What I did not like with the Rodenstock was sharpness
when pixelpeeping shifted corners. The SK35 is not perfect either but considerably better. I did not have more magenta cast with the Rodenstock, I think actually a bit less so I'm bit surprised that it did not work out on your CFV-39, it should work alright. I use an Aptus 75 which I think should perform similarly to your CFV-39 in terms of color cast.

 

[dougpeterson]

I have now shot my first images with the APO-SIRONAR DIGITAL 4,5/35mm using the CFV-39 .

But what I found is a strong magenta cast which can not be fully removed by PHOCUS in scene calibration . So it requires some time consuming PP in PSCS6.

Anyone else working with that lens ? And what DB ?

In my experience this lens should not produce any color cast (after scene calibration, aka lens cast correction) when using a Phase One or Leaf back and Capture One.

I would not accept this as "normal" and would suggest you contact your Hasselblad dealer to explore the issue with them.

 

[ondebanks]

7.0 and 6.8 are round off error from each other.

If you find the results excellent at 7.0 you will find them excellent at 6.8.

Small tweak to this - there are no 7.0 micron MFD sensors. Doug was referring to the 7.2 micron Leafs. Doesn't alter the point he was making.

Ray

 

[dougpeterson]

Small tweak to this - there are no 7.0 micron MFD sensors. Doug was referring to the 7.2 micron Leafs. Doesn't alter the point he was making.

Ray

Good catch.

 

[Shashin]

Putting your lens on a smaller pixel back will not make softer images. It only effects the 100% monitor view. But if you make a particular print size, you are not going to change your viewing distance proportional to the pixel pitch. What you are getting is higher frequency detail. How much that can be seen depends on many factors.

I use a 6um pixel sensor and I would say you are not going to have any problems.

 

[yaya]

you might notice some sharpness difference between the 7.2nm and the 6.8nm sensors (in favour of the 7.2nm) that are due to the smaller "gaps" between the pixels on the Dalsa sensor. This also helps with sharpness and light falloff when utilising shifts

BR

Yair

 

[jotloob]

In my experience this lens should not produce any color cast (after scene calibration, aka lens cast correction) when using a Phase One or Leaf back and Capture One.

I would not accept this as "normal" and would suggest you contact your Hasselblad dealer to explore the issue with them.

The APO-SIRONAR-DIGITAL 35mm produces a stronger color cast than the HR-DIGARON 40mm . An LCC shot corrects both successful in PHOCUS Scene Calibration . Therefore no need to contact the dealer .
I have to correct my earlier statement about strong magenta cast a bit .
I have rechecked what I have applied during PP .
When I apply polarization in NIK COLOR EFEX , natural "gray" materials like crushed rocks , gravel , asphalt etc. catch a magenta cast of different intensity . This is sometimes hard to remove .

 

[ondebanks]

When I apply polarization in NIK COLOR EFEX , natural "gray" materials like crushed rocks , gravel , asphalt etc. catch a magenta cast of different intensity . This is sometimes hard to remove .

I'm not surprised. Some things you cannot really simulate in post-processing software - the effects of a polarizing filter are one such thing. Polarization info is lost once the photon is recorded. After that, it's anyone's guess.

Can you smell the whiskey breath of John Martyn in one of his recorded songs? Same thing. The recording medium doesn't capture the full range of impinging information.

Ray

 

[Guy Mancuso]

Basic rule I figured out after having 5 different backs and 3 different micron backs is anything bigger than the 6 micron sensors your pretty much safe with almost any good quality lens. Its when you hit the 6 micron especially the 60 and of course the 5.4 80mpx backs that you need to start thinking better glass as a general rule at least in my mind. The P40+ is somewhat a back that can still handle some of the older lenses.

This is generally a pixel peeping comment and not scientific fact although I would not be surprised if i was correct in my assumptions.

 

[torger]

There's some psychology involved. With my medium format system I'm much more picky about lenses than for my Canon system. MF is more expensive and I want it to show. If I upgrade from my 33 to a 60 megapixel back which would cost me a substantial amount of money I surely would like pixelpeep quality be as good. When I eventually do it I hope I'm able to keep cool though and be satisfied with my SK35 rather than long for a Rodie 32 .