• Log in with Facebook Log in with Twitter
Photo

Question ... Sony FE 100mm f/2.8 GM OSS STF


  • Please log in to reply
17 replies to this topic

#1 Klaus

Klaus

    Chief Editor

  • Moderators
  • 5,403 posts
  • LocationSydney, Australia

Posted 03 August 2017 - 11:32 PM

As already hinted it is next in line ... and the MTF graph is weird (vignetting as well).

 

You may be aware that it is a T/5.6 @ f/2.8. Now I thought that the aperture is still the guiding aspect as far as diffraction is concerned. That seems to be -sort of- true for f/2.8 (T/5.6), also for f/4 (T/8). However, at f/11 (T/22) the data is way, way in diffraction land. Subjectively the figures resemble almost f/22 here. f/8 is a bit like f/13 or f/16. It seem as if diffraction "over-accelerates" the more you stop down. 

The aperture ring does also range from T/5.6 to T/22 but not beyond.

 

Thoughts?


Chief Editor
photozone.de

#2 Brightcolours

Brightcolours

    Advanced Member

  • Members
  • PipPipPip
  • 4,785 posts
  • LocationThe Netherlands

Posted 04 August 2017 - 03:33 PM

The STF acts as a gradual aperture. Lets assume (for ease) the lens is an ideal T2.8 lens without the STF element. It loses a whole 2 stops over the entire "frame". To get to T5.6, Some of the "gradual" aperture must be smaller than f5.6. So you get a percentage of light that acts as if it is diffracted through an f8 or smaller aperture (maybe even f22?), all the way up to f2.8.

 

The STF or apodization element sits comfortably far away from the aperture:

Sony-FE-100mm-STF-GM-Construction.jpg

This will mean even stopped down this element will act as "gradual aperture" and add extra diffraction softening to the image.

 

Also, since this "gradual aperture" sits further away than an aperture normally would for a normal 100mm f2.8 lens (I am assuming that the actual aperture sits in the open gap between the 4th and 5th element, from the right), the diffracted light travels further and spreads out more at the image plane than you are used to. "Gradual aperture" distance-wise it is as if you test a longer lens (say 135mm, 2.8 / 100 * 135 = f3.78), and then you still need to add the gradual part. So lets say 5.6 / 100 * 135 = f7.56 on average for the light that passes through the STF element .



#3 Klaus

Klaus

    Chief Editor

  • Moderators
  • 5,403 posts
  • LocationSydney, Australia

Posted 04 August 2017 - 04:02 PM

So the effect is not noticeable on the Laowa and Fujinon because the aperture is closer to the APD elements on those lenses?


Chief Editor
photozone.de

#4 Brightcolours

Brightcolours

    Advanced Member

  • Members
  • PipPipPip
  • 4,785 posts
  • LocationThe Netherlands

Posted 04 August 2017 - 07:20 PM

Not sure where the Loawa 105mm STF's aperture is placed.

Optical-Structure.jpg

The STF element is not that extreme (T3.2 says your review, so 1 1/3rd stop difference compared to the 2 stop of the Sony). Possibly the Loawa has a dark edge from f2 to f2.8, going to no "ND" very fast, where the Sony's transition to "no ND" is much slower? I have not seen, so not looked into, either of the 3 lenses. 



#5 obican

obican

    Advanced Member

  • Members
  • PipPipPip
  • 378 posts

Posted 04 August 2017 - 08:00 PM

Not sure how this would help, but have you double checked with a 24MP body?



#6 youpii

youpii

    Advanced Member

  • Members
  • PipPipPip
  • 435 posts
  • LocationParis/France

Posted 04 August 2017 - 08:03 PM

Wikipedia has some nice animated GIF showing diffraction.

https://en.wikipedia...iki/Diffraction



#7 obican

obican

    Advanced Member

  • Members
  • PipPipPip
  • 378 posts

Posted 04 August 2017 - 08:29 PM

When I think about this a bit further, this is probably due to the heavy STF element. 

 

Thought about it like this:

 

There is a manual way to simulate a STF element by making multiple exposures, using different apertures every single time. You can also slowly close the aperture during a single exposure to end up with the same effect. The movement (and therefore the exposure times at different apertures) however, is not linear. You have to stay (and therefore expose) at the smaller apertures much longer. So the majority of the end exposure comes from that smaller apertures. Minolta 7 (film) had this feature built in but you can simulate it if you have an aperture ring or multiple-exposure capable camera. Or simply in Photoshop.

 

So, that's what the STF element is doing. So it's actually quite natural you hit the diffraction limit rather early and this is not due to the distance between the STF element and aperture, but due to the (rather) heavy STF element.

 

Also below is a quote from a guy who did that STF Simulation in the Lensrentals Photo Geek Contest 2013:

 

 

 

I don’t have a great Bokehlicious lens, but I can turn a normal lens into one! The Olympus OM 50mm f/3.5 Macro lens has a 6-bladed aperture with non-rounded blades, giving a relatively uniform bokeh shape. However, a Gaussian is often considered the ideal bokeh shape. To make a more Gaussian blur, I used an exposure time of 30 seconds and closed down the aperture slowly from f/3.5 to 22 during the exposure. This has the effect of exposing the inner area of the bokeh longer than the outer areas, making it brighter in the center. The lens itself has aperture click-stops at f/3.5, 5.6, 8, 11, 16, and 22. I used values from Pascal’s triangle to determine the exposure time at each click stop, which respectively were 0.03, 0.3, 1.6, 4.8, 9.7, and 13.5 seconds. The first two steps I just did as quickly as possible. The approximately Gaussian bokeh is clearly much smoother than when using the constant aperture, and in my opinion quite beautiful.

 

 

Sadly the photo is gone but Roger Cicala probably has it stored somewhere.

 

Also having a heavier transition at that 2.0EV effect would make it worse.



#8 Klaus

Klaus

    Chief Editor

  • Moderators
  • 5,403 posts
  • LocationSydney, Australia

Posted 04 August 2017 - 10:40 PM

Not sure how this would help, but have you double checked with a 24MP body?

 

Nah, that's not related. It's also not a defect - as mentioned the aperture ring just goes from f/2.8 (T/5.6) to f/11 (T/22) so there has to be a reason for the "limitation".


Chief Editor
photozone.de

#9 Klaus

Klaus

    Chief Editor

  • Moderators
  • 5,403 posts
  • LocationSydney, Australia

Posted 04 August 2017 - 10:45 PM

When I think about this a bit further, this is probably due to the heavy STF element. 

 

Thought about it like this:

 

There is a manual way to simulate a STF element by making multiple exposures, using different apertures every single time. You can also slowly close the aperture during a single exposure to end up with the same effect. The movement (and therefore the exposure times at different apertures) however, is not linear. You have to stay (and therefore expose) at the smaller apertures much longer. So the majority of the end exposure comes from that smaller apertures. Minolta 7 (film) had this feature built in but you can simulate it if you have an aperture ring or multiple-exposure capable camera. Or simply in Photoshop.

 

So, that's what the STF element is doing. So it's actually quite natural you hit the diffraction limit rather early and this is not due to the distance between the STF element and aperture, but due to the (rather) heavy STF element.

 

Also below is a quote from a guy who did that STF Simulation in the Lensrentals Photo Geek Contest 2013:

 

 

 

Sadly the photo is gone but Roger Cicala probably has it stored somewhere.

 

Also having a heavier transition at that 2.0EV effect would make it worse.

 

That doesn't explain why the effect is not present on the Laowa and Fujinon.


Chief Editor
photozone.de

#10 obican

obican

    Advanced Member

  • Members
  • PipPipPip
  • 378 posts

Posted 05 August 2017 - 01:31 AM

That doesn't explain why the effect is not present on the Laowa and Fujinon.

 

Maybe because they both have rather weak STF elements at almost around 1 stop for both while Sony is at 2 stops.

 

But you are saying Fuji and Laowa don't show that diffraction effect at all?



#11 Klaus

Klaus

    Chief Editor

  • Moderators
  • 5,403 posts
  • LocationSydney, Australia

Posted 05 August 2017 - 06:57 AM

Maybe because they both have rather weak STF elements at almost around 1 stop for both while Sony is at 2 stops.

 

But you are saying Fuji and Laowa don't show that diffraction effect at all?

 

Well, look at the MTF charts ... I don't see anything special there in terms of overall characteristic.


Chief Editor
photozone.de

#12 Brightcolours

Brightcolours

    Advanced Member

  • Members
  • PipPipPip
  • 4,785 posts
  • LocationThe Netherlands

Posted 05 August 2017 - 07:54 AM

I think it is a combined effect of the apodization element sitting further away from the image plane, and the Sony one being stronger and more gradual, the 2nd diffracting more light and the 1st spreading the diffracted light more.



#13 Klaus

Klaus

    Chief Editor

  • Moderators
  • 5,403 posts
  • LocationSydney, Australia

Posted 05 August 2017 - 08:36 AM

@Wim - Regarding the "graduated aperture" - is that even possible ?

Diffraction does only occur at the edge of materials that have a different refraction index, doesn't it ?

I have a hard time to believe that the APD element has a variable refraction index.

 

THAT BEING SAID - when looking at the bokeh highlights - the effect is only extreme (and I mean really extreme) at f/2.8. It is already massively reduced 1/3 stops down. That could indicate that the APD element does indeed consist of two different segments - with the outer ring having a very different refraction index. 


Chief Editor
photozone.de

#14 obican

obican

    Advanced Member

  • Members
  • PipPipPip
  • 378 posts

Posted 05 August 2017 - 10:16 AM

Well, look at the MTF charts ... I don't see anything special there in terms of overall characteristic.

 

But you've tested the Sony STF on a much higher density sensor, maybe that's why you see the diffraction effects more easily. That's why I suggested to test the Sony on a lower resolution sensor to see what happens. I'd still try to test the old Minolta STF version too.

 

Actually disregard, Fuji APS-C sensor isn't that far behind in density.



#15 Brightcolours

Brightcolours

    Advanced Member

  • Members
  • PipPipPip
  • 4,785 posts
  • LocationThe Netherlands

Posted 05 August 2017 - 11:06 AM

@Wim - Regarding the "graduated aperture" - is that even possible ?

Diffraction does only occur at the edge of materials that have a different refraction index, doesn't it ?

I have a hard time to believe that the APD element has a variable refraction index.

 

THAT BEING SAID - when looking at the bokeh highlights - the effect is only extreme (and I mean really extreme) at f/2.8. It is already massively reduced 1/3 stops down. That could indicate that the APD element does indeed consist of two different segments - with the outer ring having a very different refraction index. 

Wim does not know these things.

 

Anyway, here is a discussion about it:

https://www.dpreview...s/post/59063567



#16 Klaus

Klaus

    Chief Editor

  • Moderators
  • 5,403 posts
  • LocationSydney, Australia

Posted 05 August 2017 - 12:22 PM

Wim does not know these things.

 

Anyway, here is a discussion about it:

https://www.dpreview...s/post/59063567

 

Still doesn't explain why this only applies to this Sony lens. Even the old Sony 135mm f/2.8 T/4.5 didn't show that behavior (to this degree) - although to be fair - that was on 24mp and not on 42mp.


Chief Editor
photozone.de

#17 Brightcolours

Brightcolours

    Advanced Member

  • Members
  • PipPipPip
  • 4,785 posts
  • LocationThe Netherlands

Posted 05 August 2017 - 12:51 PM

Still doesn't explain why this only applies to this Sony lens. Even the old Sony 135mm f/2.8 T/4.5 didn't show that behavior (to this degree) - although to be fair - that was on 24mp and not on 42mp.

The Minolta 135mm f2.8 (T4.5) STF did not use the same apodization element, it did not have T5.6 but T4.5, and the element did not sit that far from the aperture?

Laowa-STF-105mm-f2-lens-design.jpg

And yes. sharpening skewing the MTF results a bit which has become more clear with the higher res. sensors also makes comparing results a bit difficult/pointless, in this case?



#18 wim

wim

    Advanced Member

  • Members
  • PipPipPip
  • 1,160 posts
  • LocationMaastricht, Netherlands

Posted 05 August 2017 - 09:10 PM

@Wim - Regarding the "graduated aperture" - is that even possible ?

Diffraction does only occur at the edge of materials that have a different refraction index, doesn't it ?

I have a hard time to believe that the APD element has a variable refraction index.

 

THAT BEING SAID - when looking at the bokeh highlights - the effect is only extreme (and I mean really extreme) at f/2.8. It is already massively reduced 1/3 stops down. That could indicate that the APD element does indeed consist of two different segments - with the outer ring having a very different refraction index. 

 

 

Who knows ....

 

I need to look at the formulas for a while, as it does sound in a way like a graduated aperture indeed.

 

In principle diffraction is based on slit-like aperture transitions, IOW, a sharp edge going straight from light to dark. I do not know whether there is an element involved that may affect diffraction when there is not a sharp transition, but a graduated one. It could well be; I've never looked into this type of behaviour in any detail.

 

In addition, I do not know how the APD element(s) are implemented, whether they still work at F/11, or whether it is a single fixed element of which more and more of the APD effect gets excluded by the actual diaphragm when stopping down.

 

Kind regards, Wim






0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users



© by photozone.de