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

Nikon perhaps could build it

I'm not particularly attracted to very fast lenses anyway, but I certainly hope that they won't try to copy the 50mm f1.2 L RF. I'll be the only here but I don't think that it's a great lens. It's amazingly sharp, incredible, even, but I think that it's got very poor rendition in depth, and that bokeh quality degrades excessively with image height.

I like to use that image from Opticallimits to show what I mean :

Screenshot 2018-10-22 at 08.41.37-2.jpg

What I find very seductive with this photo is that there is a clear hierarchy of different planes according to distance, even within the blur, and that it remains readable. The tree behind the road sign is actually blurred, but not as much as the distant background. Besides, there is no trace whatsoever of double edges.

The 50mm RF, on the other hand, seems to flatten anything in the background and doesn't hierarchies well the different planes. In the following photo from Dpreview (shot at f4), I found it particularly hard to know if the objects on the beach were trunks on a distant beach or branches on a closer beach, the rendering is quite nervous and edgy, and there seems to be no gradual progression of the blur between the twigs in the centre near the path and the beach behind :

447A2268-4.jpg

I'm not exactly sure but I have read that the cause for that could be that some aberrations, particularly spherical aberration, are controlled in an excessively ideal way. It's great for resolution in the focal plane and for reducing focus shift, but apparently leaving a carefully tuned degree of aberrations here and there may actually benefit rendering in depth and bokeh quality in the background.

The lens also exhibits a lot of mechanical vignetting and bokeh quality degrades rapidly with image height.

It's probably an engineering tour de force, a superbly built lens (I'm expecting Lensrentals' measurements to show low sample variation, just like with most of Canon's newest designs, and their teardown clever mechanical design inside, as always), but I don't think it's a superbly well rounded photographic tool and I think that good rendering has been excessively compromised in the pursuit of other goals.

I have the same criticism against the 35 L II, and to a lesser degree the 85mm f1.4 IS USM. The trend that Canon follows with their recent lenses may actually be enough for me to abandon the system. I don't think that I've been particulalry attracted to a Canon lens since the 40mm STM (a very average, but IMO pretty well balanced lens, and very well priced).

I have the opposite opinion of Nikon's 58mm f1.4G prime : here I think that some aspects of rendering have been excessively prioritised at the cost of some other basic performance figures.

It's probably a very fine line to carefully thread between these two extremes but I think that this is were my ideal lens would be right now, and I hope that this is where Nikon's S primes, including the 50mm f1.2, and Canon's future primes, will be.
 
  • Like
Reactions: 2 users
Upvote 0

tron

CR Pro
Nov 8, 2011
5,222
1,616
If Coma is well controlled, this could be an amazing lens for Astrophotography - for example, with that f1.2 max aperture.
For astro there is a super lens: The Sigma 14mm 1.8 Art. The difference between 1.2 and 1.8 (~1 stop) is covered by the fact that the wider Sigma can be used with almost twice the shutter time (It's a 14mm vs a 24mm) without startrails so you can use the same ISO. Assuming you need Ultra wide of course. I always do need it in landscape astrophotography photos.
 
  • Like
Reactions: 1 user
Upvote 0
I'm in the minority. I want them to make slower, read "smaller and lighter," primes. These f/1.2 primes will all be 1Kg or heavier bricks on a small light body. Frustrating. It means that a lightweight FF mirrorless system is still many years away, until Canon gets around to making f/1.4 or f/1.8 lenses for this system.

You really NEED the patience of Job to stay with Canon.

The slower lenses won't be L lenses. They will come but what Canon needs first is to get enough exciting lenses into the ecosystem to sell the system.

If the rumor here is true that there is a mirrorless that is slated to be placed below the existing R, then it will make sense that buyers of that body will not be looking into $2000+ primes and zooms. It would make sense to launch consumer zooms/small primes when that body launches.
 
  • Like
Reactions: 3 users
Upvote 0

YuengLinger

Print the ones you love.
CR Pro
Dec 20, 2012
3,751
2,268
USA
I have the same criticism against the 35 L II, and to a lesser degree the 85mm f1.4 IS USM. The trend that Canon follows with their recent lenses may actually be enough for me to abandon the system. I don't think that I've been particulalry attracted to a Canon lens since the 40mm STM (a very average, but IMO pretty well balanced lens, and very well priced).

I have the opposite opinion of Nikon's 58mm f1.4G prime : here I think that some aspects of rendering have been excessively prioritised at the cost of some other basic performance figures.

It's probably a very fine line to carefully thread between these two extremes but I think that this is were my ideal lens would be right now, and I hope that this is where Nikon's S primes, including the 50mm f1.2, and Canon's future primes, will be.

You seem to be jumping to conclusions based on a few sample shots. Solid reviews and images from a variety of photographers are what we need.

And you lost me completely lumping in the ef 35mm f/1.2L II. It does gradations of distance very well for a 35mm, better than its predecessor, and the 85mm 1.4L IS does it as well as the 85mm 1.2L II, in my opinion.

Are you shooting with prints in mind?
 
  • Like
Reactions: 1 user
Upvote 0
Fine, I guess, but I will always want a fast, pro-quality 28mm and 100/105mm more. Canon bleedin' well filed the patents for a 28mm f/1.4L years ago and never pulled the trigger on it, and they had a 100mm f/1.2 design patented in the 90s which never got beyond the paperwork, either. Nikon and Sigma both have made or are making high-quality, fast lenses like this; come on, Canon, catch up already.
 
Upvote 0
Jun 12, 2015
852
298
I'm not particularly attracted to very fast lenses anyway, but I certainly hope that they won't try to copy the 50mm f1.2 L RF. I'll be the only here but I don't think that it's a great lens. It's amazingly sharp, incredible, even, but I think that it's got very poor rendition in depth, and that bokeh quality degrades excessively with image height.

I like to use that image from Opticallimits to show what I mean :

View attachment 181143

What I find very seductive with this photo is that there is a clear hierarchy of different planes according to distance, even within the blur, and that it remains readable. The tree behind the road sign is actually blurred, but not as much as the distant background. Besides, there is no trace whatsoever of double edges.

The 50mm RF, on the other hand, seems to flatten anything in the background and doesn't hierarchies well the different planes. In the following photo from Dpreview (shot at f4), I found it particularly hard to know if the objects on the beach were trunks on a distant beach or branches on a closer beach, the rendering is quite nervous and edgy, and there seems to be no gradual progression of the blur between the twigs in the centre near the path and the beach behind :

View attachment 181144

I'm not exactly sure but I have read that the cause for that could be that some aberrations, particularly spherical aberration, are controlled in an excessively ideal way. It's great for resolution in the focal plane and for reducing focus shift, but apparently leaving a carefully tuned degree of aberrations here and there may actually benefit rendering in depth and bokeh quality in the background.

The lens also exhibits a lot of mechanical vignetting and bokeh quality degrades rapidly with image height.

It's probably an engineering tour de force, a superbly built lens (I'm expecting Lensrentals' measurements to show low sample variation, just like with most of Canon's newest designs, and their teardown clever mechanical design inside, as always), but I don't think it's a superbly well rounded photographic tool and I think that good rendering has been excessively compromised in the pursuit of other goals.

I have the same criticism against the 35 L II, and to a lesser degree the 85mm f1.4 IS USM. The trend that Canon follows with their recent lenses may actually be enough for me to abandon the system. I don't think that I've been particulalry attracted to a Canon lens since the 40mm STM (a very average, but IMO pretty well balanced lens, and very well priced).

I have the opposite opinion of Nikon's 58mm f1.4G prime : here I think that some aspects of rendering have been excessively prioritised at the cost of some other basic performance figures.

It's probably a very fine line to carefully thread between these two extremes but I think that this is were my ideal lens would be right now, and I hope that this is where Nikon's S primes, including the 50mm f1.2, and Canon's future primes, will be.

I am an advocate for the qualities in a lens you are striving for. I favor the images I get from the 85LII over the 85L f1.4, and I think the 35L has a quality that makes it look better than the 35II, in some types if images. I really do like the 35LII, though.

I don’t think we have seen enough samples of RF50 f1.2 to judge it. (I havent) That fall off from sharp to defocused area/depth rendering is very hard to measure and compare. I see the difference in the lenses I use because I shot my family a lot, and that makes it easier to notice subtle differences in how images look.

Kudos to you, for trying to illustrate with pictures, but I believe they are too different to prove your point. Fingers crossed for you being wrong on the RF 50L.;)
 
  • Like
Reactions: 1 users
Upvote 0

Architect1776

Defining the poetics of space through Architecture
Aug 18, 2017
583
571
122
Williamsport, PA

Looks like Canon is making the R series seriously professional.
With a diameter opening of 54mm vs the Sony limited opening of only 46.1mm (Even the Canon M mount is bigger at 47mm for a comparison) it gives Canon (And Nikon) a real leg up going forward with large aperture professional lenses.
 
Upvote 0

Architect1776

Defining the poetics of space through Architecture
Aug 18, 2017
583
571
122
Williamsport, PA
I'm in the minority. I want them to make slower, read "smaller and lighter," primes. These f/1.2 primes will all be 1Kg or heavier bricks on a small light body. Frustrating. It means that a lightweight FF mirrorless system is still many years away, until Canon gets around to making f/1.4 or f/1.8 lenses for this system.

You really NEED the patience of Job to stay with Canon.

They likely will make RF lenses smaller and lighter as well. But I see them showing the pros "Here is what can be done" and stick with us on this one. Remember the first EOS were the 650/620 cameras. Nothing great and mind blowing about the bodies except they took the revolutionary EF lenses that blew the competition away for literally decades before the others caught up. Now you have the revolutionary RF mount with the extra control ring and ability to program the lens including which way to turn the focus ring. Leaving all others in the technology dust again.
 
  • Like
Reactions: 3 users
Upvote 0

eyeheartny

EOS R | 50 1.2 RF
Sep 3, 2018
56
32
Looks like Canon is making the R series seriously professional.
With a diameter opening of 54mm vs the Sony limited opening of only 46.1mm (Even the Canon M mount is bigger at 47mm for a comparison) it gives Canon (And Nikon) a real leg up going forward with large aperture professional lenses.

Yes, and the technical advantages of a bigger mount seem clear. The edge to edge performance on the RF lenses is excellent and better than other lenses. I think the wider mount diameter contributes to this, as the light rays at the edges have to bend less to reach the sensor.
 
  • Like
Reactions: 1 users
Upvote 0
You seem to be jumping to conclusions based on a few sample shots. Solid reviews and images from a variety of photographers are what we need.

And you lost me completely lumping in the ef 35mm f/1.2L II. It does gradations of distance very well for a 35mm, better than its predecessor, and the 85mm 1.4L IS does it as well as the 85mm 1.2L II, in my opinion.

Are you shooting with prints in mind?

You don't need to look at a ton of samples to already guesstimate to a certain degree what will come out of these tests. We already know, for example, that the 50 RF is unarguably incredibly sharp for an f1.2 lens, or that it will be quite significantly affected by cat's eyes. Some aspects of how a lens is biased or designed in terms of blur can be guessed from a few samples.

Solid reviews in general don't give a particularly comprehensive review of blur behaviour. They're still extremely useful and the 50mm RF should look like a brillant lens in most of them. Deservedly so. I'm sure it's taken a lot of hard work for Canon engineers.

The 35mm II is limited in its expression of distance within the blur because of how perfect a lens it is. Its blur characteristic between rear and front blur is as neutral as it gets, as Lenstip's focussing scale shots show.

This is the 35mm II :

35 L II.jpg

Here I've allowed myself to switch the 35mm II for the 55 Otus (a lens with similar blur characteristics as the 35mm II) as the way the focusing scale is shot makes it more comparable to another standard lens, specifically designed to under-correct some aberrations :

Screenshot 2018-10-22 at 22.44.44.jpg

As you can see the "LensAlign" word blurs in the exact same way whether it's in front or behind the focal plane with the Otus (or the 35mm II). On the contrary on the lens with poor spherical aberration correction, it blurs in a strong double edged fashion in front, but in a very smooth, readable way behind.

And here's the big deal : with the Otus (and the 35mm II), as soon as you're past a certain number, you can no longer read the word, and there's practically no difference in the look of the "LensAlign" word : it's as if the LensAlign word is repeated in the exact same fashion afterwards, with lots of well defined edges which start to intertwine with each others to the point that the word becomes undistinguishable and that everything becomes flat.

A practical illustration of the problem :
Screenshot 2018-10-22 at 23.11.41.jpg
https://www.imaging-resource.com/PRODS/canon-5ds-r/Y1C6A7739-35LMarkII.HTM
In that shot the two blurred men are at a distance from each others. What I'd like to see from my ideal lens is the traduction of that fact into how the blur behaves. Yet what do we see ? The blurred edges of the objects are similar regardless of the distance (see for example how the neck strap of the guy behind is just as sharply defined as the one from the guy in front). Maybe my brain is stupid, but to me this is difficult to process : I know that these guys are at a distance from each others (the size of their respective heads should be enough to know it, unless something quite bad was unraveling to their physical integrity), and yet they look like they're painted on a 2D canvas.

On the other hand, on the lens with strong spherical aberration, the "LensAlign" word remains readable far into the blurred area behind, and there's a strong difference in how that word looks between the numbers 5 and 10, for example. I think that this is a more elegant way to fade into the blur, but, much more importantly : it preserves distance information better since you can still see what the object is !

The same thing may apply at longer focusing distances, or even when closed down (depending on lens design).

The EF 50mm f1.2 behaves to some degree like the under-corrected lens, at least around f2-F4 in the centre. It's got a lot of other aberrations that make its overall bokeh performance questionable, particularly wide open and off-centre, though.

Another illustration of the difference between these two lenses, from a pdf from Zeiss :
Screenshot 2018-10-22 at 23.21.47.jpg
https://lenspire.zeiss.com/photo/app/uploads/2018/04/Article-Bokeh-2010-EN.pdf

That said, I've left the photos in color, because credit where credit is due, the 35mm L II controls longitudinal chromatic aberrations exceedingly well. Hat's off to Canon's engineers for that. It's also much, much sharper than the poorly corrected lens, which some may call "hazy".

IMO in 2018 a more balanced lens would try to straddle the fine line between these two extremes.
 
Last edited:
  • Like
Reactions: 1 users
Upvote 0

Jethro

EOS R
CR Pro
Jul 14, 2018
996
1,037
Looks like Canon is making the R series seriously professional.
With a diameter opening of 54mm vs the Sony limited opening of only 46.1mm (Even the Canon M mount is bigger at 47mm for a comparison) it gives Canon (And Nikon) a real leg up going forward with large aperture professional lenses.
Yes - there really has to be a 'pro' high MP body coming to use these properly. If these lenses are scheduled for late 2019, it might give a clue about that body's release date as well.
 
  • Like
Reactions: 1 users
Upvote 0

YuengLinger

Print the ones you love.
CR Pro
Dec 20, 2012
3,751
2,268
USA
A practical illustration of the problem :
View attachment 181150
https://www.imaging-resource.com/PRODS/canon-5ds-r/Y1C6A7739-35LMarkII.HTM
In that shot the two blurred men are at a distance from each others. What I'd like to see from my ideal lens is the traduction of that fact into how the blur behaves. Yet what do we see ? The blurred edges of the objects are similar regardless of the distance (see for example how the neck strap of the guy behind is just as sharply defined as the one from the guy in front). Maybe my brain is stupid, but to me this is difficult to process : I know that these guys are at a distance from each others (the size of their respective heads should be enough to know it, unless something quite bad was unraveling to their physical integrity), and yet they look like they're painted on a 2D canvas.

In the photo in question, we have a black shirt against a black shirt. Of course this reduces the illusion of depth in a 2D image. But the farther face is blurred quite a bit more than the closer, and pleasingly, I might add. Considering the focal length is only 35mm, Canon did a great job producing convincing and pleasing sharpness fall-off, or whatever the technical term is. They might not have achieved perfection, but if one is to speak of balance, they did very well with considering price, size, AF performance, weight, and, in the case of the 85mm 1.4L, the important addition of IS.

As for the 50mm f/1.2L, we agree, MayaTlab. One reason I chose not to buy it is because of how many images I've seen where the background blur is not pleasing at all. Handled correctly, the lens's bokeh is beautiful, but, in my opinion, overall it is simply too finicky of a lens to count on in a range of situations. I see some photographers, even wedding photographers, who build an entire portfolio with the same hazy, backlit, glowing style. Too formulaic, too specialized.

I have high hopes for the rf 50mm f/1.2L, and slim hopes for a new version to come to EF bodies.
 
Last edited:
Upvote 0

dominic_siu

R5, 1435, 2870, 100500, 28, 100 Macro , 135 (RF)
Aug 31, 2018
107
94
Would love high quality 1.8 lenses like what Nikon is doing with the Z. I have the EOS R and 85mm 1.4 IS. Love the output but it's heavy and unbalance.
Have you also use the battery grip? I also using 85 1.4L on my EOS R together with the grip, although much heavier just like 5D4 by this setup makes it more balance.
 
Upvote 0

hmatthes

EOS-R, RF and EF Lenses of all types.
Same here!
Somehow I am quite lucky that I haven't invested much in EF L pimes yet :cool: . For sure this will be expensive, but...
Now I can start saving ;)
I'm happy that I invested in the EF "Holy Trinity" f/2.8 Zooms 20 years ago in the film days -- 2 are still in service, only the 24~70 is newer replacing the 28~70 f2.8.
I shall replace them when the RF replacements come out. Long term investments.
 
Upvote 0