RF 85/1.2 L vs RF 85/1.2 L DS (new video w/ comparisons)

[navastronia]

I don't know what you are seeing in the files but I'm not seeing the same as you. Many of his comparison images are not sharp anyway but here is the best comparison I could see and the earrings don't look different to me.


View attachment 189145
View attachment 189146

I stared at them for a while, really went over them with a fine-toothed comb, and it does look to me like the DS has more DoF at the same apertures.

Look at the way the fence poles are rendered in the below 2 images. The focus point appears to be the same, and yet there's more bokeh on the 85 L than the 85 L DS.

You can also see that the leaves on the right-hand side are more OoF on the 85 L.

85 L



85 L DS

 

[jd7]

No it doesn't, how does T stop impact depth of field? It doesn't.

I agree about T stop. I cannot see how T stop could effect depth of field.

However it seems Canon thinks the DS lens does give greater DOF one way or another.

Canon RF 85mm F1.2L USM DS first shoot - Canon Europe

Wedding photographer Félicia Sisco was the first pro to shoot with Canon's EF 85mm f/1.2L USM DS lens. How does the Defocus Smoothing coating work, and how does it compare to the non-DS lens?

www.canon-europe.com

I haven't tried to investigate why that might be though.

I note the Canon article says depth of field "appears" deeper with the DS lens, but since depth of field is inherently about perception anyway I cannot see that "appears" is particularly significant in Canon's statement.

 

[stevelee]

Why would you want greater depth of field in a lens that you got to make the background appear blurrier?

 

[stevelee]

Equivalence is a relatively simple concept that ends up with answers people don't really like because it goes against their initial intuition and they normally miss one or two of the basic criteria,

But as a concept in photography it is well understood and defined, it has been written about at length and explained in very great detail and is very easy to test for yourself even if you only have one camera and a zoom lens.

It may be well defined, but people very often mean different things by it. I won't bother to document from various threads here. I recall a web site that expends page after page dealing with what it means.

 

[privatebydesign]

It may be well defined, but people very often mean different things by it. I won't bother to document from various threads here. I recall a web site that expends page after page dealing with what it means.

Then they are wrong, if it has a definition people can't mean different things by it, that is the very definition of definition! Some people might use the word incorrectly in which case we need to try to educate them.

This is probably the link you mean. http://www.josephjamesphotography.com/equivalence/

 

[stevelee]

Then they are wrong, if it has a definition people can't mean different things by it, that is the very definition of definition! Some people might use the word incorrectly in which case we need to try to educate them.

This is probably the link you mean. http://www.josephjamesphotography.com/equivalence/

Yes, I think that is it. Do you agree with his ten-bullet-point definition? If not, why is he wrong?

In his and other discussions much seems to be made of total light hitting the sensor. I’ve never understood the practical ramifications of that when taking pictures.

 

[jd7]

Why would you want greater depth of field in a lens that you got to make the background appear blurrier?

I think because how blurry the background appears isn't just a function of depth of field, and in any event bokeh is not inherently about amount of blur.

Depth of field describes which parts of the subject are considered to appear "in focus" when viewed under the relevant conditions (regarding display size and distance from view to image and assuming average eyesight).

As a general proposition, shallower depth of field is associated with a blurrier background - at least assuming the same focal lengh is used. I believe that as you move to longer focal lengths you can have situations where for a given framing it is easier to get greater depth of field while getting at least as much background blur or more, because the amount of blur is substantially related to the actual aperture (rather than f-stop, which is relative aperture).

However, I also understand that different lens designs can create different amounts of blur at a particualr f-stop / depth of field. Further, better bokeh would mean a better quality of blur, but that does not necesasrily have to mean more blur.

Perhaps greater DOF could be an advantage with an 85 f/1.2, so as to help get more of the subject in focus while still getting great bokeh? (Of course, the amount of your subject which is in focus depends substantially on distance to subject, but for example in a portrait setting it is easy with 85mm to be close enough to the subject that at f/1.2 you are only getting, say, eyes in focus with nose and ears already out of focus. That may be a look you want, but equally it may not be.

 

[privatebydesign]

Yes, I think that is it. Do you agree with his ten-bullet-point definition? If not, why is he wrong?

In his and other discussions much seems to be made of total light hitting the sensor. I’ve never understood the practical ramifications of that when taking pictures.

Yes I do.

He lays out a clear definition of equivalence, essentially taking images with identical photographic metrics, DoF, noise, angle of view, subject movement, etc, within the limits of available sensors, from systems with different sized sensors when reproduced at the same size.

There is no practical ramification of the same amount of light hitting the sensor when taking pictures, but when viewing them at the same size output the entire point is the different sensors images need to have the same noise levels (again within available sensor efficiency), to do that you have to even up the amount of light each sensor receives regardless of size.

This is incredibly easy to prove especially if you have a FF camera and a 2x zoom like a 70-200 or a 24-70. For instance take a picture at 150mm, f5.6, and 400 iso, compare that entire image to a crop the size of an APS-C, so a 1.5 crop shot at 100mm, f3.5 and 200 iso. Take those two images from the same place and they will be identical by every meaningful photographic metric. To make it easier to see the similarities do the same at higher iso's.

 

[jd7]

Yes, I think that is it. Do you agree with his ten-bullet-point definition? If not, why is he wrong?

In his and other discussions much seems to be made of total light hitting the sensor. I’ve never understood the practical ramifications of that when taking pictures.

I may be wrong about this, but I think of it this way. If you use a larger sensor, the image has to be enlarged less to get to a given output size. So, for a given output size, the light captured by the sensor doesn't have to be spread out as much as from a smaller sensor to get to the given output size, ie the larger sensor has, in a sense, "more light" to squash into the output image than than a smaller sensor. I understand this makes a difference because it means for a given output size, the signal to noise ratio is higher for a larger sensor than a smaller sensor, and signal to noise ratio is important in determining things like how much noise will appear in the image.

(Note that in what I've said above, I am assuming the larger and smaller sensrors have similar quantum efficiency.)

EDIT: I see PBD just beat me with a post which I think is saying something similar to what I've said above(?)

 

[Ozarker]

I stared at them for a while, really went over them with a fine-toothed comb, and it does look to me like the DS has more DoF at the same apertures.

Look at the way the fence poles are rendered in the below 2 images. The focus point appears to be the same, and yet there's more bokeh on the 85 L than the 85 L DS.

You can also see that the leaves on the right-hand side are more OoF on the 85 L.

85 L

View attachment 189148

85 L DS

View attachment 189149

For me, I just can't see enough difference between the two that favors $300 more for the DS. I had wondered whether or not I would regret not waiting for the DS, but now I am glad I didn't. Happy for those that own, and find it better, for themselves. I just don't see it. I prefer the harder edges on the bokeh balls myself. Once one starts stopping down the effect is lost anyway.