Canon RF 24mm f/1.2L & RF 85mm f/1.2L in the works [CR1]

Jan 12, 2011
760
103
“We find the RF 24mm f/1.2L an interesting choice for the lineup, especially with the speed of the lens.”

24mm is an interesting choice but makes perfect sense given the 1.8x crop in 4K on the EOS-R. With that crop it will end up somewhere between a 40 and a 50mm, something a 35mm 1.2 could not do in video mode.

Plus, they need a fast wide for landscape photographers.
 
Upvote 0
"3D effect" is just fine on my EF 35mm f/1.4L II. I'm very impressed with it for a wide angle lens. Sharp as heck too.

I'll try not to bring up the term "3D" in this exchange. It's a word that I feel is over-used. To be frank I think that I nearly hate myself from expressing my feelings in this thread. Sometimes I feel like I see things that just aren't there. But then I can't also deny, as Lenstip's focusing scale tends to show, and other examples here and there, such as Marianne Oelund's thread on DPreview, that lenses do indeed vary quite considerably in the way they render transitions from sharp to blur. And I can't deny that I find shots like these, well, let's say, aesthetically challenged in the blur department :
https://3.img-dpreview.com/files/p/sample_galleries/4037492029/4652016161.jpg
https://www.magezinepublishing.com/...hres/a-50mm-lens-bike-4P2A0426_1536166554.jpg
I mean, those double images as soon as one leaves the centre, or the nervous rendering of the bicycle's front wheel and handlebar are plain evil to my eyes. I believe that most lenses would struggle here, but that exactly where I'm expecting Canon engineers to work their magic for a 50mm f1.2 €2500 lens.

If I may do so, this is what I see in this picture :
Screenshot 2018-10-23 at 18.46.35.jpg
What I feel is that, to some extent, the subject is in front of a painted canvas. I'm lacking here a certain transitional quality, which some lenses are able to better convey (although at this aperture it's normal that the background becomes unreadable at some point).
It's as if this photo has three very defined, different stages : sharp as f!ck, then a very short, slightly nervous transition zone, then blur with sharply defined edges to the blurred areas.

It's difficult to criticise the 35mm II, because, in a way, it's a "perfect" lens. I guess that this is exactly how a lens is supposed to behave. And in many ways it's a lot better looking than a lens with over-corrected spherical aberration and onion rings, or double lines bokeh.

Years ago I was after "perfect" lenses like this (or at least as good as I could get them then), because relatively speaking, they behaved better wide open than other lenses that were just sub-par, with onion ring bokeh, etc. I used to use a 50mm Makro Planar instead of Nikon's cheap AF-D 50mm lenses because it had a more neutral and controlled bokeh at f2 than the Nikons, throughout the frame. But these days, I feel that lens designers seem better equipped to fine tune the aberrations of a lens to purposefully bias the bokeh rendering towards the rear without excessively affecting other areas of performance, and I think that I'm after those lenses right now :D.

Forgive me for picking nits, but I do not think the picture supports your opinion in regards to the blurring effects visible between the two people. In the near neck strap, I can see there are white words or graphics repeated in a pattern which are completely lost in the far strap. Additionally, the far strap is blurred to the point of being semi-transparent as is clearly visible with the red shirt logo and more subtly visible as part of the strap almost disappears into the black shirt.
View attachment 181171


In regards to their faces, the near face includes just enough detail to see some skin folds around the mouth, eyes and interior shaping of the pinna (ear). Note that I shrunk the near face to better match the size of the far face.
View attachment 181172

Certainly I am zooming into small details here, but these were things that were evident to me from your thumbnail image. OTOH, we are discussing opinions around subjective impressions, so all I can really say is that my opinion on this matter does not match yours.

I think you're right to nitpick :D. Your observations are valid and I agree with you. It's difficult for me to exactly nail what I see in these pictures that I find objectionable. Perhaps it's less the amount of blur than the sharp edges around the blur that troubles my mind. In the first crop you're right to highlight that the strap on the guy in the background is blurrier and more transparent indeed. But the edge of the strap fades into the black shirt just as rapidly as the strap in front. Perhaps it's that.
 
Last edited:
  • Like
Reactions: 2 users
Upvote 0

stevelee

FT-QL
CR Pro
Jul 6, 2017
2,383
1,064
Davidson, NC
Just as with color balance and dynamic range, what I am usually striving to do with background blur is to suggest, and maybe approximate, how the scene looked to me in person. When my eyes focus on something, there is a certain amount of blur in the appearance of things different distances away. In real life I can refocus my eyes between various things, so the photograph is not ever going to be a simulation.

I don't drink heavily, and usually will not have had anything alcoholic to drink at all when I am making pictures. So in bright sunlight, my pupils are never fully dilated. Fast lenses wide open don't give a natural look on sunny days.

I do like the effect you can get with a telephoto lens. There is no pretense of a normal viewing angle, so a fairly shallow depth of field can look right, I guess because we are used to seeing pictures made that way.
 
Upvote 0
Canon's strength has always been lenses and they know this going forward.

Jack

What do you think so? Nope, I know Canon very well for over 50 years. We all of our concern about Canon's behavior from 1989 incident by abandoned world best lenses, FD and FDn without warning. Canon forced everyone to use the EOS system with EF lenses. We, professional photographers, do not support R camera and any RF lenses at all.
 
Upvote 0

Ozarker

Love, joy, and peace to all of good will.
CR Pro
Jan 28, 2015
5,933
4,336
The Ozarks
I'll try not to bring up the term "3D" in this exchange. It's a word that I feel is over-used. To be frank I think that I nearly hate myself from expressing my feelings in this thread. Sometimes I feel like I see things that just aren't there. But then I can't also deny, as Lenstip's focusing scale tends to show, and other examples here and there, such as Marianne Oelund's thread on DPreview, that lenses do indeed vary quite considerably in the way they render transitions from sharp to blur. And I can't deny that I find shots like these, well, let's say, aesthetically challenged in the blur department :
https://3.img-dpreview.com/files/p/sample_galleries/4037492029/4652016161.jpg
https://www.magezinepublishing.com/...hres/a-50mm-lens-bike-4P2A0426_1536166554.jpg
I mean, those double images as soon as one leaves the centre, or the nervous rendering of the bicycle's front wheel and handlebar are plain evil to my eyes. I believe that most lenses would struggle here, but that exactly where I'm expecting Canon engineers to work their magic for a 50mm f1.2 €2500 lens.

If I may do so, this is what I see in this picture :
View attachment 181169
What I feel is that, to some extent, the subject is in front of a painted canvas. I'm lacking here a certain transitional quality, which some lenses are able to better convey (although at this aperture it's normal that the background becomes unreadable at some point).
It's as if this photo has three very defined, different stages : sharp as f!ck, then a very short, slightly nervous transition zone, then blur with sharply defined edges to the blurred areas.

It's difficult to criticise the 35mm II, because, in a way, it's a "perfect" lens. I guess that this is exactly how a lens is supposed to behave. And in many ways it's a lot better looking than a lens with over-corrected spherical aberration and onion rings, or double lines bokeh.

Years ago I was after "perfect" lenses like this (or at least as good as I could get them then), because relatively speaking, they behaved better wide open than other lenses that were just sub-par, with onion ring bokeh, etc. I used to use a 50mm Makro Planar instead of Nikon's cheap AF-D 50mm lenses because it had a more neutral and controlled bokeh at f2 than the Nikons, throughout the frame. But these days, I feel that lens designers seem better equipped to fine tune the aberrations of a lens to purposefully bias the bokeh rendering towards the rear without excessively affecting other areas of performance, and I think that I'm after those lenses right now :D.



I think you're right to nitpick :D. Your observations are valid and I agree with you. It's difficult for me to exactly nail what I see in these pictures that I find objectionable. Perhaps it's less the amount of blur than the sharp edges around the blur that troubles my mind. In the first crop you're right to highlight that the strap on the guy in the background is blurrier and more transparent indeed. But the edge of the strap fades into the black shirt just as rapidly as the strap in front. Perhaps it's that.

I don't think I dispute your observations much, but the fact is that it is a wide angle lens. I believe that has some to do with not getting the bokeh you might be looking for. In my experience, the shorter the focal length the more harsh the bokeh. So for me, the 35L II is just fine. I haven't owned any other Canon 35mm lenses. Maybe I should compare it with my 24-70 with both lenses @ f/2.8.
 
Upvote 0

cayenne

CR Pro
Mar 28, 2012
2,868
796
Thats nice and all but both are sure to be near $3k each, no thanks. Pricing on these RF lenses so far is absurd.

Well, not everyone can afford Porsches either.....for some, they will have to settle for Hondas, Chevies or Kia's......

Like with Apple products, not everyone is in their target market, but there seems to be PLENTY that are.

Me? I just wait and save my nickels, save and sacrifice to get what I perceive to be the best in most all things I buy. I'd rather cook at home (better food anyway), and not buy little piddly sh*t all the time, and save my money for really nice things a couple times a year.

I'll buy these lenses and bodies...it just won't be all at once, and I will give up buying other things less of worth to me to get them.

Just my $0.02.....

cayenne
 
  • Like
Reactions: 2 users
Upvote 0
Mar 25, 2011
16,848
1,835
Well, not everyone can afford Porsches either.....for some, they will have to settle for Hondas, Chevies or Kia's......

Like with Apple products, not everyone is in their target market, but there seems to be PLENTY that are.

Me? I just wait and save my nickels, save and sacrifice to get what I perceive to be the best in most all things I buy. I'd rather cook at home (better food anyway), and not buy little piddly sh*t all the time, and save my money for really nice things a couple times a year.

I'll buy these lenses and bodies...it just won't be all at once, and I will give up buying other things less of worth to me to get them.

Just my $0.02.....

cayenne
Prices will drop, and drop a lot. Those who wait will get much lower prices.

It is puzzling as to why Canon released a middle of the road camera along with some pretty high end lenses. Except for the 35mm and the 24-105, there is a mismatch. It sounds to me like there was initially a plan to bring both middle and high end cameras to the market, and lenses were developed for both. Something may have delayed the high end camera, but since the lenses were ready, and lenses were needed for the "R", they pushed them out. If we are seeing even more very wide aperture expensive lenses in the near future, that pro level camera can't be far away. The EOS R is a camera destined to end up in big box stores, and those expensive lenses will not be there with it.
 
  • Like
Reactions: 5 users
Upvote 0
While I am confident that lenses have character in my experience the light has a dramatic ence to the 3Dish effects of flat images. In macro I move the camera some degree around a subject and only ONE angle gives the depth I want.

To compare lenses we would need (1) the same subject / surroundings, (2) same light, (3) same sensor and (4) same processing chain and then a lot of different types of images like portrait, landscape with different distance settings ...

About character of lenses: I have often thought about selling my FD lenses but I am glad that I still have them. With the M50 they are usable - in terms of handling - because of the cameras EVF and the good focus highlighting.
 
Upvote 0
About the new EF-M 32 1.4 which is definitely a new lens design: While having 14 lens elements it has only 8 lens groups. My first thought was that Canon kept the group count on a reasonably low number to reduce ghosts / flares and contrast degeneration.
But maybe they tried to make lens groups for correction purposes of CA but these groups act as "single lenses" for the rest of aberrations?

In my view this lens creates a lot of depth with its sharpness transitions while being very sharp in the focal plane without being too sharp / unnaturally sharp. The bokeh is nearly free of artifacts, just the cat eye shape is visible if you have highlights in the background (you can find some images I made with this lens including 3 100% "stripes" to take a look at the lens' quality at http://michaelbockhorst.de/2018/10/canon-ef-m-32-1-4-stm-sample-photos/ ).
But - I said it in my previous post - it would be great to see a side by side comparison e.g. between this lens and the RF 50 1.2, their lens compositions are close at first glance but the EF-M sports a concave front lens. The RF lens has two plane surfaces beneath the diaphragm while the EF-M has two concave surfaces more like classical double gauss lenses.

@MayaTlab: Interesting standpoints about these subtle image properties beyond these "measurable" parameters!

Maybe a lot of RF lenses go this route?
 
Upvote 0

Jack Douglas

CR for the Humour
Apr 10, 2013
6,980
2,602
Alberta, Canada
What do you think so? Nope, I know Canon very well for over 50 years. We all of our concern about Canon's behavior from 1989 incident by abandoned world best lenses, FD and FDn without warning. Canon forced everyone to use the EOS system with EF lenses. We, professional photographers, do not support R camera and any RF lenses at all.

And in my state of ignorance I just kept on using my FD lenses with my F1; a lot was going on in my life at that time so I felt no photographic pain, but I can sympathize. Getting into digital with Nikon, it took very little nudging to move back to Canon and I don't feel shortchanged. The equipment is far better than the photographer in my case (IOW I'm not a pro)!;)

Jack
 
  • Like
Reactions: 1 user
Upvote 0
@MayaTlab: Interesting standpoints about these subtle image properties beyond these "measurable" parameters!

I hope that what I'm trying to express in my uneducated, unsophisticated way is completely measurable by optical engineers. I don't want to be the apostle of "3D pop" fluff or whatever and write articles like the one on Petapixel. The good news is that I'm pretty sure they are able to do so, now more than ever (I don't subscribe to the notion that "old" lenses are systematically better than "new" ones, far from it).
I found it interesting that, in the DC Watch interview, Nikon engineers said that while the theory behind leaving residual aberrations to enhance / bias bokeh was well understood decades ago, they struggled with the quantitative aspect. Years ago Nikon announced that they had developed what I believe is a measurement and software system to improve their understanding of a lens' aberration. Perhaps this was to them just another step forward in their capacity to quantify some aspects of bokeh behaviour. I have very little doubt that Canon engineers are trying to do so as well.

But in my opinion, looking at what the 50mm L RF is producing in OOF areas, I don't think that bokeh was that high on Canon's list of priorities with this lens (let's say a lot lower than MTF figures). Beyond what could be a personal preference for lenses biasing the bokeh towards rear smoothness (something the RF may actually do to some limited extent in the dead centre at f1.2 ?), I don't think that anyone with a pair of reasonably functioning eyes can miss the very strong and defined double images that start to appear half-way through the frame or how the bokeh suddenly changes character. For a €2500 lens which main claim to fame is its f1.2 aperture, I'm expecting better.
 
Last edited:
Upvote 0
It is puzzling as to why Canon released a middle of the road camera along with some pretty high end lenses. Except for the 35mm and the 24-105, there is a mismatch. It sounds to me like there was initially a plan to bring both middle and high end cameras to the market, and lenses were developed for both. Something may have delayed the high end camera, but since the lenses were ready, and lenses were needed for the "R", they pushed them out. If we are seeing even more very wide aperture expensive lenses in the near future, that pro level camera can't be far away. The EOS R is a camera destined to end up in big box stores, and those expensive lenses will not be there with it.

Pretty sure they indeed wanted to release 2 R camera's like Nikon did, but delayed the 'Pro' one for some unknown reason.
Maybe it has to do with the (firmware) performance not being adequate. Just look like how they released the R with a promised feature firmware update.
 
Upvote 0
I hope that what I'm trying to express in my uneducated, unsophisticated way is completely measurable by optical engineers. I don't want to be the apostle of "3D pop" fluff or whatever and write articles like the one on Petapixel. The good news is that I'm pretty sure they are able to do so, now more than ever (I don't subscribe to the notion that "old" lenses are systematically better than "new" ones, far from it).
I found it interesting that, in the DC Watch interview, Nikon engineers said that while the theory behind leaving residual aberrations to enhance / bias bokeh was well understood decades ago, they struggled with the quantitative aspect. Years ago Nikon announced that they had developed what I believe is a measurement and software system to improve their understanding of a lens' aberration. Perhaps this was to them just another step forward in their capacity to quantify some aspects of bokeh behaviour. I have very little doubt that Canon engineers are trying to do so as well.

But in my opinion, looking at what the 50mm L RF is producing in OOF areas, I don't think that bokeh was that high on Canon's list of priorities with this lens (let's say a lot lower than MTF figures). Beyond what could be a personal preference for lenses biasing the bokeh towards rear smoothness (something the RF may actually do to some limited extent in the dead centre at f1.2 ?), I don't think that anyone with a pair of reasonably functioning eyes can miss the very strong and defined double images that start to appear half-way through the frame or how the bokeh suddenly changes character. For a €2500 lens which main claim to fame is its f1.2 aperture, I'm expecting better.

The RF 50 1.2 is in my opinion a very capable lens to make news photos or maybe landscape it needs to have great contrast and resolution over the whole frame. Very good technical IQ at f/1.2 is the reason to buy this one. Transitions in out of focus zones maybe isn't first concern in that case.
The EF 50 1.2 is the more artistic lens - if I speak with artists they usually have very simple equipment but do much more work "in front of the camera" by chasing light, subjects and different views. Here the overall look of the image is more important than having pixel related sharpness.
Maybe Nikon combines both in one lens (that is what I read in your comments)?!

About quantifying bokeh, just my ideas:
Usually you track points for the subject and try to find out how precise the images are on the focal plane of a sensor.
(1) With newer computing capabilities maybe Nikon (and the rest of the gang) look how non-focused areas are reproduced by a lens. Nothing less or more than tracking thousands of points forming the shape of the subject of interest. Now you have the image of an unsharp subject and you can do some evaluations of its shape on the focal plane. Computational methods can easily decide between a simple gaussian blurr or another blurr pattern.
(2) You simulate different designs and look which is perceived best by a group of photographers (customers, etc.) OR you use the "unsharpness patterns" of existing lenses after putting their output into a mathematical model.
(3) You decide which perceived "unsharpness patterns" are the most welcomed and use the corresponding lens design for production.
 
Upvote 0
Maybe Nikon combines both in one lens (that is what I read in your comments)?!

I'm not sure that it's actually possible to have both one's cake and eating it, but I think that they're trying to find a pleasing compromise between two extremes that could be more or less adequately represented on one end, by this new 50mm RF, or perhaps the Zeiss Otus, and on the other, by the Nikon 58mm f1.4. Perhaps this is this search for a well-received compromise that made it difficult for Nikon engineers to find the right quantification for what they were trying to achieve. They knew very well the qualities that they were looking for, but less the quantities that would create a "just right" compromise for users to enjoy.

The important word here is "trying". I don't think that we'll all agree on what an ideal compromise would be, and IMO it can change depending on focal length, max aperture, focusing distance, type of lens, etc. That said, I'm pretty certain that the lukewarm reception the 28mm f1.8G and 58mm f1.8G received, and the very positive reception the 105mm f1.4 and 24mm f1.8G received, are guiding their hands. Which is why I think that it's very important that reviews provide a much more comprehensive assessment of lenses, particularly in the OOF areas. The interest of the Nikon 58mm was lost on me for years until I gradually built a newfound appreciation for what it was trying to achieve.

I'm sure that Canon understands what Nikon is trying to achieve, and even that they're doing the same thing to some degree, but maybe this is where each lens engineer's personal sensibilities come into play, and they collectively don't feel quite as strongly as Nikon that this is a worthy endeavour to pursue ?
 
  • Like
Reactions: 1 user
Upvote 0

Bob Howland

CR Pro
Mar 25, 2012
917
588
Prices will drop, and drop a lot. Those who wait will get much lower prices.

It is puzzling as to why Canon released a middle of the road camera along with some pretty high end lenses. Except for the 35mm and the 24-105, there is a mismatch. It sounds to me like there was initially a plan to bring both middle and high end cameras to the market, and lenses were developed for both. Something may have delayed the high end camera, but since the lenses were ready, and lenses were needed for the "R", they pushed them out. If we are seeing even more very wide aperture expensive lenses in the near future, that pro level camera can't be far away. The EOS R is a camera destined to end up in big box stores, and those expensive lenses will not be there with it.
I don't think that puzzling is the correct word. The two high end lenses almost certainly are just to get press space. However, replacing them with a 16-35 f/4 L, 70-200 f/4 L and 28-300 f/4-5.6 would have been much more consistent.
 
Upvote 0
What do you think so? Nope, I know Canon very well for over 50 years. We all of our concern about Canon's behavior from 1989 incident by abandoned world best lenses, FD and FDn without warning. Canon forced everyone to use the EOS system with EF lenses. We, professional photographers, do not support R camera and any RF lenses at all.

o_O
 
  • Like
Reactions: 1 users
Upvote 0
I'm not sure that it's actually possible to have both one's cake and eating it, but I think that they're trying to find a pleasing compromise between two extremes that could be more or less adequately represented on one end, by this new 50mm RF, or perhaps the Zeiss Otus, and on the other, by the Nikon 58mm f1.4. Perhaps this is this search for a well-received compromise that made it difficult for Nikon engineers to find the right quantification for what they were trying to achieve. They knew very well the qualities that they were looking for, but less the quantities that would create a "just right" compromise for users to enjoy.

The important word here is "trying". I don't think that we'll all agree on what an ideal compromise would be, and IMO it can change depending on focal length, max aperture, focusing distance, type of lens, etc. That said, I'm pretty certain that the lukewarm reception the 28mm f1.8G and 58mm f1.8G received, and the very positive reception the 105mm f1.4 and 24mm f1.8G received, are guiding their hands. Which is why I think that it's very important that reviews provide a much more comprehensive assessment of lenses, particularly in the OOF areas. The interest of the Nikon 58mm was lost on me for years until I gradually built a newfound appreciation for what it was trying to achieve.

I'm sure that Canon understands what Nikon is trying to achieve, and even that they're doing the same thing to some degree, but maybe this is where each lens engineer's personal sensibilities come into play, and they collectively don't feel quite as strongly as Nikon that this is a worthy endeavour to pursue ?

This has been an interesting discussion. You're spot on talking about compromise, and design aims. Wasn't the old story about the 85L f/1.2 that it was designed in close collaboration with (a?) fashion/portrait photographer(s?), and so it produced the type of image they wanted, giving all sorts of other compromises? Nowadays the prime concerns (of reviewers especially) seem to be sharpness (esp wide open and off centre) and AF performance, so those tend to be prioritised? Just a thought.
 
  • Like
Reactions: 1 user
Upvote 0