New Lens Features

[rpcrowe]

The Sony 100mm f/2.8 STF lens incorporates a special apodization element which increases the quality of the bokeh of this lens, although cutting down the transmission to a great degree. This post is not to discuss the qualities of the STF lens just to mention that it has a feature which would not have been dreamed of several years ago.

I don't know if this would or could be possible but I would like to see a feature in a macro lens that would allow the use of a smaller aperture without suffering from the diffraction that softens an image.

Wouldn't it be nice to be able to shoot at f/22 or even f/32 or f/45 without having to take diffraction into consideration. This way we could have a lens that is able to achieve decent imagery with a much wider DOF. DOF is often a problem when shooting macro images...

[LePetomane]

The Fujifilm 56mm f/1.2 has the same feature but it is not a macro lens. Minimum focus distance is 27 inches.

[Manfred M]

Hardly a new feature Richard.

It looks like Minolta introduced this feature in 1999 with the f/2.8 135mm STF lens. So far as I can tell, the way the filter works is by reducing the influence if the edges of the Airy disk.

There is no "free lunch" in physics. The price that is paid was steep as the f/2.8 lens is a T-4.5 lens (a loss of 1-1/2 stops). Your 100mm lens seems to be even worse from a light gathering characteristic as it is nominally an f/2.8 lens but only a T-5.6 lens (a loss of 2-stops). Light waves diffract as the hit a sharp edge (like the camera's diaphragm) and this is what causes the diffraction. There is simply no way around that and your lens clearly shows the price that is paid to mitigate the effects of diffraction.

I suspect signal processing algorithms in a post-processing tool is likely the only way to improve diffraction issues in real life. I suspect someone is bound to be working on that.

[dje]

I suspect signal processing algorithms in a post-processing tool is likely the only way to improve diffraction issues in real life. I suspect someone is bound to be working on that.

Capture One has Diffraction Correction as an option in the Lens Correction tab.

Dave

[Manfred M]

Capture One has Diffraction Correction as an option in the Lens Correction tab.

Dave

I've used that setting and did not see any difference in the final image. It could be as simple as that I was above the diffraction limit in my aperture setting and no adjustment was made. Regardless, I am suspicious of any setting that is a simple on / off for a fairly complex issue.

[dje]

I've used that setting and did not see any difference in the final image. It could be as simple as that I was above the diffraction limit in my aperture setting and no adjustment was made. Regardless, I am suspicious of any setting that is a simple on / off for a fairly complex issue.

Yes I've never played with it seriously. On the occasions i have tinkered with it, I have noticed a small difference in sharpness however I guess it depends on what other sharpening settings have been made. C1's help file claims that it uses a sophisticated deconvolution technique that is processor intensive (hence it is turned off by default). Diffraction can be modelled mathematically based on the aperture setting and wavelength.

Dave

[xpatUSA]

Yes I've never played with it seriously. On the occasions i have tinkered with it, I have noticed a small difference in sharpness however I guess it depends on what other sharpening settings have been made. Sony's help file claims that it uses a sophisticated deconvolution technique that is processor intensive (hence it is turned off by default). Diffraction can be modelled mathematically based on the aperture setting and wavelength.

Dave

Dave, LR deconvolution is my sharpening algorithm of choice - rarely USM. And, if you get the radius close enough, it can offset diffraction quite well. Like many things, the more you use it, the easier it gets.

[Manfred M]

Dave, LR deconvolution is my sharpening algorithm of choice - rarely USM. And, if you get the radius close enough, it can offset diffraction quite well. Like many things, the more you use it, the easier it gets.

The input sharpening defaults in Lightroom / Adobe Camera Raw effectively are deconvolution sharpening. Input sharpening is the appropriate place to apply that type of sharpening.. USM is generally used for creative sharpening and output sharpening (especially when printing).

I generally find that the defaults for input sharpening work fairly well, although on occasion I will tweak the values during raw conversion. Often I want no input sharpening in the sky or when there is a lot of water in the image, so I will tweak that by masking out the sharpening in those areas in Photoshop. I will use one sharpened layer and one unsharpened layer and I will apply a layer mask to do this.

[xpatUSA]

The input sharpening defaults in Lightroom / Adobe Camera Raw effectively are deconvolution sharpening.

Effectively? But I read long ago that, in ACR, deconvolution is only applied above a certain level of one of the sliders (detail?).

[DanK]

Effectively? But I read long ago that, in ACR, deconvolution is only applied above a certain level of one of the sliders (detail?).

I think that is right. Adobe has never released anything definitive, but from what I have read and seen--including from some seemingly good sources--the detail slider has to be at 100 (right-hand edge) to do deconvolution sharpening. I have no idea how they do this, but the common view is that it is a continuum from a USM-like sharpening on the left to a deconvolution sharpening on the right. Some speculate that this is the same math that underlies smart sharpening in Photoshop, but I have no idea whether that is correct.

The input sharpening defaults in Lightroom / Adobe Camera Raw effectively are deconvolution sharpening. Input sharpening is the appropriate place to apply that type of sharpening

This is not correct, as far as I know. Lightroom has only two types of sharpening: output sharpening (in the export dialog and print module) and other sharpening, which I will call "develop sharpening." There is no distinction between input and creative sharpening other than when it happens (by default). The input sharpening it does by default when importing (which you can turn off) simply imposes a modest amount of this develop sharpening toward the USM end: amount 25, radius 1, detail 25 (out of 100), masking 0. Any later sharpening you do later in editing simply modifies those initial values. You can set Lightroom not to sharpen on input--leaving everything at zero--and do all of the sharpening at a later point, and as long as the final values are the same, the final image will be the same.

At one point, I had the import values set to zero, but I have found it more convenient recently to leave them at their default values.

[Manfred M]

I think that is right. Adobe has never released anything definitive, but from what I have read and seen--including from some seemingly good sources--the detail slider has to be at 100 (right-hand edge) to do deconvolution sharpening. I have no idea how they do this, but the common view is that it is a continuum from a USM-like sharpening on the left to a deconvolution sharpening on the right. Some speculate that this is the same math that underlies smart sharpening in Photoshop, but I have no idea whether that is correct.



This is not correct, as far as I know. Lightroom has only two types of sharpening: output sharpening (in the export dialog and print module) and other sharpening, which I will call "develop sharpening." There is no distinction between input and creative sharpening other than when it happens (by default). The input sharpening it does by default when importing (which you can turn off) simply imposes a modest amount of this develop sharpening toward the USM end: amount 25, radius 1, detail 25 (out of 100), masking 0. Any later sharpening you do later in editing simply modifies those initial values. You can set Lightroom not to sharpen on input--leaving everything at zero--and do all of the sharpening at a later point, and as long as the final values are the same, the final image will be the same.

At one point, I had the import values set to zero, but I have found it more convenient recently to leave them at their default values.

Dan - it seems you and I go through similar experiments in our work flows.

I have found three things affect my sharpening habits:

1. Learning - trying to figure out what works and why things work. Unfortunately some of the seminal books and other works on sharpening are at least a decade old and the cameras and output technology has changed, so they cannot be relied upon as being valid today.

2. End results - as long as I am happy with the final image, I don't really care. For most work appearing on the web, I stick with the defaults as this often gets me more than acceptable results. As I downsize to 2MP to from a 36MP camera output sharpening becomes rather meaningless as I can't control the end device that someone is viewing on.

When it comes to printing, I do care and I still haven't found the perfect workflow, so my experimentation continues.

I do creative sharpening (and unsharpening) whenever I print. It does make a difference there and here Photoshop is the only tool I use.

3. Shear laziness - the default sharpening routines and settings in the raw convertors are pretty darn good most of the time. I might tweak these settings manually if I'm not liking what the defaults are doing for me.

[DanK]

Manfred,

yes, it sounds like ware basically handling this the same general way.

One complication I found in trying to sort this out is that a lot of what is written touts one way as the way.

I sometimes create two sharpening layers in Photoshop and toggle them on and off to compare. That led me to return to a very old-fashioned and often disparaged method for some images: a high-pass filter. I find it is a very good tool for images that have a small number of high-contrast lines and a lot of areas that shouldn't be sharpened.

[William W]

. . . I don't know if this would or could be possible but I would like to see a feature in a macro lens that would allow the use of a smaller aperture without suffering from the diffraction that softens an image. . . Wouldn't it be nice to be able to shoot at f/22 or even f/32 or f/45 without having to take diffraction into consideration. . .

I think that it will be a while before Camera Lens manufactures put TR&D into putting an apodization element into a Macro Lens.

However, although probably the main reason for employing employing an apochromat (apochromatic lens) in the design of a (macro) Lens is to (better) reduce chromatic and spherical aberrations I do believe that there's likelihood that these lenses do perform better apropos the effects of diffraction.

I base this thought on my general reading on Microscopy and Micro-Photography, where an apochromat was employed in the optic.

Having that thought, I then had a quick search for Macro Lenses which employed an apochromat - there are a few, probably not a comprehensive list and in no particular order:

> Apo-Macro-Elmarit-TL 60mm F/2.8 ASPH
> Sigma 150mm F/2.8 APO Macro EX DG HSM
> Voigtlander 65mm F/2 APO-Lanther Macro
> IBE RAPTOR APO Cine Macro (various - 4 Lenses)

I think it is not yet released but it is in the pipeline:
> Laowa 100mm F/2.8 2:1 Ultra Macro APO

***

. . . This post is not to discuss the qualities of the STF lens just to mention that it has a feature which would not have been dreamed of several years ago.

Yeah I know, but, CiC allows a lot of wandering . . .

Having mentioning "Laowa":

My Laowa 105mm F/2 Smooth Trans Focus is, for me, a dream lens.

I did use the Minolta f/2.8 135mm STF, and the Laowa is miles ahead: noted that we're now 20 years better researched and developed.

I have neither used the Fuji nor the Sony. Anyway, for my purposes the Sony is a tad too slow (aperture) and looses too much as its T/Function (as Manfred has already commented).

But I had researched the Sony STF, (sort of tirelessly) and I found that reasonably credible commentators seem to put the Laowa, generally, in front of the Sony.

***

BTW - Minolta had an alternative for those who couldn't or wouldn't buy their Minolta STF 135mm F/2.8 T4.5, is was to buy their Maxxum 7.

Hidden away in the Custom Functions of the Maxxum 7 (CF 25 - 2) is the "STF" Mode.

This CF executes seven continuous multiple exposures at various apertures.

I've never used a Maxxum 7. Researching those who did, it appears that the STF Custom Function did not out perform using the Minolta STF 135mm f/2.8 T4.5, but I expect that the outcomes of how well the STF Custom Function performed, would be rooted in:
what lens was used;
what film was used;
and the operator's proficiency.

WW

[Antonio Correia]

The lens I sent back !
No regrets.

[xpatUSA]

I think that is right. Adobe has never released anything definitive, but from what I have read and seen--including from some seemingly good sources--the detail slider has to be at 100 (right-hand edge) to do deconvolution sharpening. I have no idea how they do this, but the common view is that it is a continuum from a USM-like sharpening on the left to a deconvolution sharpening on the right. Some speculate that this is the same math that underlies smart sharpening in Photoshop, but I have no idea whether that is correct. <>

Dan, out of curiosity, I've just Googled <"Eric Chan" deconvolution> and got:

"Yes Photoshop's Smart Sharpen is based on deconvolution (but you will need to choose the "More Accurate" option and the Lens Blur kernel for best results). Same with Camera Raw 6 and Lightroom 3 if you ramp up the Detail slider."

https://forum.luminous-landscape.com...?topic=45038.0

As you probably know, Chan, et al, is/was responsible for ACR.

He also said:

"Hi Deja, yes, the sharpening in CR 6 / LR 3 is a continuous blend of methods (with Detail slider being the one used to "tween" between the methods, and the Amount, Radius, & Masking used to control the parameters fed into the methods). As you ramp up the Detail slider to higher values, the deconvolution-based method gets more weight. If you're interested in only the deconv method then just set Detail to 100 (which is what I do for low-ISO high-detail landscape images). Not recommended for portraits, though ... "

https://forum.luminous-landscape.com...8027#msg378027

So your memory is pretty good!

[DanK]

Ted,

Thanks. The sources I had were mostly second-hand, so this is very helpful.

Dan

[revi]

I think that it will be a while before Camera Lens manufactures put TR&D into putting an apodization element into a Macro Lens.

However, although probably the main reason for employing employing an apochromat (apochromatic lens) in the design of a (macro) Lens is to (better) reduce chromatic and spherical aberrations I do believe that there's likelihood that these lenses do perform better apropos the effects of diffraction.

I base this thought on my general reading on Microscopy and Micro-Photography, where an apochromat was employed in the optic.

I think you are confusing "apodisation" and "apochromatic".
The first is the application of a weighing(?) function, e.g in certain applications of the Fourier Transform to limit artifacts.
The type of functions is used to lower the intensity of the signal within a window smoothly to zero, in order to remove sharp edges which cause "ringing" artefacts.
In lens design, it would be used to lower the intensity of the sidebands in the Airy discs by "smoothing" the sharp edges of the diaphragm (poor explanation, I know).

An "apochromate" is a lens grouping to reduce chromatic and spheric abberations, it doesn't apply a window function, but plays on the differences in dispersion
between different types of glass, and on the difference in spheric abberation between different lens profiles.

[William W]

Hiya Remco,

I think you are confusing "apodisation" and "apochromatic". . .

No. I wasn't confusing the two terms and I thought that I had explained that I knew that they are different.

I was merely expressing a 'thought' that I had. And maybe my 'thought' has some worth so I'll put it out there for discussion . . .

REF my bold now for emphasis:

1. The words used-

your words - "apochromate" is a lens grouping to reduce chromatic and spheric abberations,"

my words - "However, although probably the main reason for employing employing an apochromat (apochromatic lens) in the design of a (macro) Lens is to (better) reduce chromatic and spherical aberrations

2. My 'thought' "I do believe that there's likelihood that these lenses do perform better apropos the effects of diffraction."

3. My reason for having a 'thought': "I base this thought on my general reading on Microscopy and Micro-Photography, where an apochromat was employed in the optic.

***

However, if the meaning of your reply is:

"I don't think that your thought has any validity. I think that it has no validity because there cannot be any way that an apochromate will have any effect whatsoever on limiting airy disc artifacts" - then I accept that as an opinion - but I ask why is that opinion so definitively conclusive?

WW

[William W]

The lens I sent back !

Why?

WW

[revi]

Hiya Remco,



No. I wasn't confusing the two terms and I thought that I had explained that I knew that they are different.

I was merely expressing a 'thought' that I had. And maybe my 'thought' has some worth so I'll put it out there for discussion . . .

REF my bold now for emphasis:

1. The words used-

your words - "apochromate" is a lens grouping to reduce chromatic and spheric abberations,"

my words - "However, although probably the main reason for employing employing an apochromat (apochromatic lens) in the design of a (macro) Lens is to (better) reduce chromatic and spherical aberrations

2. My 'thought' "I do believe that there's likelihood that these lenses do perform better apropos the effects of diffraction."

3. My reason for having a 'thought': "I base this thought on my general reading on Microscopy and Micro-Photography, where an apochromat was employed in the optic.

***

However, if the meaning of your reply is:

"I don't think that your thought has any validity. I think that it has no validity because there cannot be any way that an apochromate will have any effect whatsoever on limiting airy disc artifacts" - then I accept that as an opinion - but I ask why is that opinion so definitively conclusive?

WW

My reply purely concerned the use of the two terms.

But now that you express it that way, I indeed have difficulty seeing how an apochromate will have any influence
on the Airy discs (without claiming there isn't an influence):
As I understand the physics, the Airy discs are purely due to diffraction on the sharp edge of the diaphragm, and independent of any chromatic or spherical abberations.

Better focussing of the incoming rays (which is basically what an apochromate does) should not change that diffraction, so shouldn't change the Airy discs.

On the other hand, the artifacts in the Airy disc are very similar to the result of a Fourier Transform of a sharp edge. Such sharp edges can occur
in data obtained through several analytical methods, like nuclear magnetic resonance. In order to limit the artifacts, apodisation functions are used to
remove the sharp edge (I'm simplifying here a bit, picking an apodisation function has something of a black art). But, that implies that you throw away
some of the information, which results in broadening of the signal you want.

So using an optical apodisation function (as described in the OP) results in limiting the intensity of the sidebands in the Airy discs, at the cost of a loss of signal (light).
And that loss of light means either longer exposures (with all that that implies), higher ISO or larger apertures, which will negate the advantage in certain applications.

Note that I don't mean to imply that an apochromate won't improve sharpness! A reduction in chromatic and spherical aberrations will imrove sharpness, especially off-center,
which in itself is worthwhile.

Otoh, can you still speak of acromatic or apochromatic groups in modern optical designs?