The Canon RF 85mm f/1.2L USM: The developers answer 10 questions

koenkooi

EOS R
CR Pro
Feb 25, 2015
1,487
1,288
Well, the Sigma 85mm f/1.4 Art lens can currently be bought new in the UK on the grey market for around £770. That's $975 at today's exchange rate.

So, if Canon would like to offer me their new 85mm RF lens for, let's say, $1000 then I would be very happy to give it a look.

Otherwise, err, the Sigma is pretty awesome on my 5Div to be honest ...
Why are you expecting a 1st party f/1.2 lens to be only 2.6% more expensive than a 3rd party f/1.4 lens?!?!
 

hollybush

EOS M6 Mark II
Feb 1, 2012
50
28
How is his argument wrong? 85 mm is 85 mm. The length is from the sensor the center of the lens. If 85 mm = 85 mm, then the camera where the length from the mount to the sensor is shorter, to make it to 85 mm, the length to the lens has to be longer.
I've seen this misconception pop up in various places a few times in the last few months, so let me correct it now:

The distance from the front of a lens to the focal plane can be less than its focal length. That, rather than the focal length itself, is actually the definition of "telephoto".

Of course, to shorten a lens requires adding extra/more expensive/different elements, and in this case the designers have chosen not to shorten the lens any more than they might have done for an EF version, so it ends up physically longer than the EF lens would have been and also longer than the old EF lens.
 

navastronia

EOS RP + 5D Classic
Aug 31, 2018
647
741
Sony's run muuuuuuch quieter (the decent to good lenses). I've got no experience with Nikons.

The only RF I don't have much experience with is the 24-105, I hear that one is significantly quieter than the rest, but it seems to me that Canon doesn't really give a shit about making quiet lenses for the RF yet.

I'm a stills guy, luckily. If I were a video guy... I wouldn't touch any RF stuff yet.
I don’t know anyone serious about film who 1) uses autofocus 2) captures audio with a shotgun mic on the camera. As was said above, that’s why we use booms and lapel mics.
 
  • Like
Reactions: CanonFanBoy

FramerMCB

Canon 40D & 7D
CR Pro
Sep 9, 2014
469
136
53
Well, the Sigma 85mm f/1.4 Art lens can currently be bought new in the UK on the grey market for around £770. That's $975 at today's exchange rate.

So, if Canon would like to offer me their new 85mm RF lens for, let's say, $1000 then I would be very happy to give it a look.

Otherwise, err, the Sigma is pretty awesome on my 5Div to be honest ...
Fair point. But I'm rather positive when I say, that you won't need to be worried about being tempted by this new RF lens. I will never, ever work on your 5D Mk IV. So, I would stick with your Sigma 85mm f/1.4 Art...
 
  • Haha
Reactions: CanonFanBoy

IsaacImage

EOS 90D
Feb 23, 2013
163
13
39
Toronto
www.IsaacImage.com
Remains to be seen. Let me explain.
A lens may be razor sharp at an infinity. And that’s how sites like LensRentals tend to evaluate lenses.
However, portraiture work is done at a much closer distance to your subject. A few meters for a head and shoulders with a 85mm lens.
Now here is something to be aware off.
e.g. Sigma 105/1.4 Art was confirmed to be a very sharp lens at an infinity by LensRentals.
Well, yup.. it sure is at around 2100 points
But, the same lens evaluated at MFD (1meter) to subject truly sucks in sharpness department with nearly scratching 1600 points mark. A massive difference.
Same is true for Sigma 85/1.4 Art and 135/1.8 Art.

The takeaway from this exercise is: the lens in question is unlikely to be a ruthlessly Sharp for portraits. Unless shot at a around 4 meters and longer distance to subject.
very interesting especially regards to a 105
I found 105 and 40 Art the sharpest lenses I ever had
do you have a link to the information ?
 

MayaTlab

EOS 90D
Oct 6, 2015
193
81
very interesting especially regards to a 105
I found 105 and 40 Art the sharpest lenses I ever had
do you have a link to the information ?
I have no idea where secureGSM gets his info.

That said both the 40 and 105mm from Sigma intentionally introduce some degree of aberrations, such as spherical aberration, to enhance background blur smoothness / transitions (a perfectly corrected lens will produce evenly illuminated shapes with a hard edge when defocused, a lens with a purposefully finely tuned set of aberrations may be able to produce a blur that's more gaussian), and the aberrations mix may vary depending on focus distance.

Given the available samples it doesn't seem to be at the detriment of resolution IMO, at least not in a particularly significant way.
 

YuengLinger

Sufficiently Pixelated
CR Pro
Dec 20, 2012
3,375
1,732
USA
Really looking forward to seeing examples of photos taken by photographers who purchase the lens, not just have a copy for evaluation. I've posted before, I'm not thrilled with the overall look of images from the ef 85mm 1.4L IS, though, mechanically, it performs well, achieving sharp focus, reasonably quickly. Just kind of flat results compared to similar lighting scenarios with the ef 85mm 1.2L II sold to buy the IS version.

(For comparison, what does "thrill" me? The ef 35mm f/1.4L II, the ef 135mm f/2L, the ef 85mm f/1.2L II which I sold, the Sigma 180mm f/2.8 Macro, the ef 70-200mm f/2.8L IS II...at times, when things come together just right, the ef 24-70mm f/2.8L II.)

Hopefully this newest 85mm will enter the ranks of Canon's best lenses.

But as great as it might be, I'm holding out for a more robust RF mount body, one that hopefully has a joystick as an alternative to touch screen AF point selection, two cards, a less punishing 4k crop factor, and better weather sealing.

Canon knows customers such as myself will be joining the RF club at some point! Very happy that serious portraiture is still a high priority.
 
Last edited:

SecureGSM

2 x 5D IV
Feb 26, 2017
2,376
1,245
very interesting especially regards to a 105
I found 105 and 40 Art the sharpest lenses I ever had
do you have a link to the information ?
I don’t have a link. these are numbers (repeatable through a quite a number of tests and lenses) from my own quick tests. Happy to share my methodology over message.

In addition to the information above:
Sigma 105/1.4 Art is an extremely slow focusing at MFD lens. Up to one full second slow at 1meter to subject. Not so much at 2m, 4m or infinity.
 
Last edited:

Joules

EOS R
CR Pro
Jul 16, 2017
1,349
1,593
Hamburg, Germany
I don’t have a link. these are numbers (repeatable through a quite a number of tests and lenses) from my own quick tests. Happy to share my methodology over message.
So, did you only do your test at MFD and infinity?

I don't think anybody who's for example ever seen a review from Christopher Frost should be shocked that a lens performs poorly when used at mfd, especially wide open.

What I would be shocked about is if that situation wouldn't improve dramatically by movinv further than MFD. The way you phrased your initial Response it seemed to me like one could misinterpret in a way that suggests a linear relation between distance and resolution. Without having looked at that myself, I would not expect that to be the case.
 

SecureGSM

2 x 5D IV
Feb 26, 2017
2,376
1,245
So, did you only do your test at MFD and infinity?

I don't think anybody who's for example ever seen a review from Christopher Frost should be shocked that a lens performs poorly when used at mfd, especially wide open.

What I would be shocked about is if that situation wouldn't improve dramatically by movinv further than MFD. The way you phrased your initial Response it seemed to me like one could misinterpret in a way that suggests a linear relation between distance and resolution. Without having looked at that myself, I would not expect that to be the case.
where do you read that I suggested a linear relation between distance and resolution?
i suggest the following:
Sigma 105/1.4 is a week performer at MFD, it is OK at around 1.6m, good at 3m and excellent at infinity.
Canon L glass are a more consistent performers transitioning from MFD to infinity. there is a weakness at MFD but not as much as in Sigmas case.

and finally:

++++ Without having looked at that myself, I would not expect that to be the case.
SGSM: Yup, have a look.. come back with solid facts and then lets have a conversation. at this stage your allegations are a little bit unsubstantiated...

p.s.: tests for SIgma 105/1.4 Art are done at: 1m, 1.6m, 3.0m, infinity
 
Last edited:

Joules

EOS R
CR Pro
Jul 16, 2017
1,349
1,593
Hamburg, Germany
where do you read that I suggested a linear relation between distance and resolution?
i suggest the following:
Sigma 105/1.4 is a week performer at MFD, it is OK at around 1.6m, good at 3m and excellent at infinity.
Canon L glass are a more consistent performers transitioning from MFD to infinity. there is a weakness at MFD but not as much as in Sigmas case.

and finally:

++++ Without having looked at that myself, I would not expect that to be the case.
SGSM: Yup, have a look.. come back with solid facts and then lets have a conversation. at this stage your allegations are a little bit unsubstantiated...

p.s.: tests for SIgma 105/1.4 Art are done at: 1m, 1.6m, 3.0m, infinity
I didn't accuse you of suggesting that there was a linear relation between the two. I wrote that it could be misunderstood to mean that. I at least got that impression on the first read, it simply is what comes to my mind when talking about numerical data with only two datapoints. One could do a lot of things if the day is long though. Apologies if it came across as an attack.

Thanks for following up with the Evaluation of more distances. I'm indeed curious about your method for getting the corresponding numbers. I didn't come across any reviews which measure resolution over distance. My Sigma 150-600mm seems to exhibit just the opposite effect, getting poorer with larger distances - or maybe that's just because the amount of air between lens and subject increases too far after some point, or simply because I didn't use it enough yet to really get a good grip in its performance. A qay to put some numbers on it could be cool.
 

YuengLinger

Sufficiently Pixelated
CR Pro
Dec 20, 2012
3,375
1,732
USA
So for your rather exorbitant retail price...you get an 85mm f1.2 lens...with a Blue goo element, a function ring, a much bigger lens, a non fly by wire AF system and pretty much every thing else that the EF lens does exemplary. Optically...sure it'll be sharp...but so is the EF version. Oh...and that lens was a portrait lens...so it doesn't need a closer focus distance than it's already got.
Price? Consider inflation and the price of the 85mm 1.2 II when released.

Having a sharper portrait lens with the same or better bokeh is pretty important to this portrait photographer, especially if it does AF noticeably faster.

Sorry, but I can't agree with your assertion that a portrait lens doesn't benefit from closer focus distance. Some portrait photographers even use macro just to get a little closer.
 

SecureGSM

2 x 5D IV
Feb 26, 2017
2,376
1,245
I didn't accuse you of suggesting that there was a linear relation between the two. I wrote that it could be misunderstood to mean that. I at least got that impression on the first read, it simply is what comes to my mind when talking about numerical data with only two datapoints. One could do a lot of things if the day is long though. Apologies if it came across as an attack.

Thanks for following up with the Evaluation of more distances. I'm indeed curious about your method for getting the corresponding numbers. I didn't come across any reviews which measure resolution over distance. My Sigma 150-600mm seems to exhibit just the opposite effect, getting poorer with larger distances - or maybe that's just because the amount of air between lens and subject increases too far after some point, or simply because I didn't use it enough yet to really get a good grip in its performance. A qay to put some numbers on it could be cool.
The Sigma 150-600 is a tricky one to calibrate and by the sound of it you never proper calibrated your lens at infinity.

Infinity for you lens at 600mm end is approx. 100meters to subject
For 150mm end the infinity sits at around 25 meters to subject.

Adding further to a complexity of the getting your lens adjusted to your camera, calibration has to be done at 16 point in total. It usually takes around 2-3 hours to calibrate a long Sigma zoom lens. I hope it helps.
 
Mar 15, 2018
80
98
United States
I have a general question for the portrait photogs... How can you get an entire face in focus at f1.2? When I shoot my 50mm at 1.8 on an 80D, it's basically a crapshoot if I get a subject's face entirely in focus, and I can't really tell until I'm watching on a monitor. Any tips?
 

YuengLinger

Sufficiently Pixelated
CR Pro
Dec 20, 2012
3,375
1,732
USA
I have a general question for the portrait photogs... How can you get an entire face in focus at f1.2? When I shoot my 50mm at 1.8 on an 80D, it's basically a crapshoot if I get a subject's face entirely in focus, and I can't really tell until I'm watching on a monitor. Any tips?
Many portrait photographers use lenses capable of extremely shallow depth of field so they can choose how much of the face to have in focus and out of focus. We practice over and over with friends, families, and mannequin heads and other objects. The shallow depth of field gives creative options that simply aren't available in most kit lenses or other slow lenses.

The short answer to your question: Your subject has to be far enough away for the entire face to be in focus. But is that the distance you want for a portrait? If not, you must stop down to an aperture that does have the whole face in focus.

If your subject's face is at an angle to the axis of the lens, then you need even more depth of field than if your subject is looking straight into the lens. But even straight on, a shallow depth of field will have the nose blurred out, or the ear lobes, etc.

A shallow depth of field can produce dramatic, lovely results, or just look off. Depends on the skill and intent of the photographer (and sometimes luck).

On a technical note, here's a link to one of many available DoF calculators:

 

mjg79

EOS 90D
Feb 19, 2016
160
102
So for your rather exorbitant retail price...you get an 85mm f1.2 lens...with a Blue goo element, a function ring, a much bigger lens, a non fly by wire AF system and pretty much every thing else that the EF lens does exemplary. Optically...sure it'll be sharp...but so is the EF version. Oh...and that lens was a portrait lens...so it doesn't need a closer focus distance than it's already got.

The twaddle about the RF lens being longer than the EF version due to the lack of a mirror box just made me laugh...so are all RF lenses over 35mm going to be bigger and heavier than their EF counter parts? Kind of makes me worry about the rumored RF 500mm f4.
Well this is the reality on two fronts. One is that we are at the point of diminishing returns, especially for lenses like fast 50s and 85s where for most sharpness isn't the be all and end all. We already saw that with the 85/1.2 L II vs the 85/1.4 IS L - for portraits the older lens has arguably a nicer rendering. And with fast 50s the new RF 50/1.2 is a technological tour de force and yet for photographing most people, especially women over the age of about 23, I would always choose the older EF 50/1.2 L.

Regarding the RF lens being longer it's not twaddle it's physics. Sony has shown this over and over again. Apart from primes around 28mm and wider there is either no size advantage or there is a size penalty. The 24-70GM for example is longer than the 24-70L II. The Sony 50/1.4 is huge as is their 85GM. There will be advantages with wider lenses - the 24GM, 16-35GM and Nikon's new 14-30/4 are all considerably smaller than their equivalent SLR lenses.

It's a pity. I always assumed that a 300/2.8 would be big regardless but I thought mid-range lenses would see big improvements. But if for example you look at the 35/1.4s or mid range zooms available for the Sony mount they are essentially SLR sized even the ones designed from the ground up. If you want an f1.4 aperture, sharp corners and auto focus at 35 or 50 or 85mm then there will be compromise somewhere. Only Leica make a top performing 35/1.4 that is small but it costs a fortune and sacrifices autofocus.

I think however that many people here are missing a key part of Canon's strategy. I believe Canon is serious when they say they will keep the EF mount going. The two compliment each other well and given it's electronic there is no problem adapting. This is more sensible I think than what Sony did. So for example Canon gives us the EF 70-200 2.8L III. It will work perfectly on either EF or RF though will balance better on an EF camera. In the RF mount they will give an extending 70-200 2.8L. For many professionals they will want the non-extending one but there will be plenty who will like the smaller design. There's no right or wrong.

I've used an A9 with some telephoto lenses, native and adapted. It's an amazing bit of kit, there's no question, but the ergonomics are poor compared to a 1DX. And that's not just button placement etc I believe it's physics and balance - such lenses will be big and heavy and a 1DX sized camera, or even a 5dIV one matches up better with a centre of gravity more centralised. There is no way Canon will push their professional users to start using something ergonomically inferior to the previous model. So while I think EF-S won't last much longer as the M can replace it better in 95% of scenarios, I think EF and RF will play together well. You will be able to use a 400/2.8, 35/1.4 and 135/2 from the EF range, for example (none of which would actually get any better by virtue of the RF mount anyway) along with the new 15-35/2.8 - and use them all on the EOS-R and when required mount the EF lenses on a EF mount camera (which would likely be most of the time with the 400 and possibly even with the others). How this plays out in terms of business only time will tell but there is a logic to what Canon says.
 
  • Like
Reactions: navastronia

mjg79

EOS 90D
Feb 19, 2016
160
102
I have a general question for the portrait photogs... How can you get an entire face in focus at f1.2? When I shoot my 50mm at 1.8 on an 80D, it's basically a crapshoot if I get a subject's face entirely in focus, and I can't really tell until I'm watching on a monitor. Any tips?
It's a fair question really. One thing to remember is that when further from the subject it becomes less of an issue. If for example you take a full length portrait with an 85/1.2 you get a fast fall off in focus without having to worry about whether it's eyeball or eye lashes in focus.

But the real answer is that the beauty of many top quality lenses is not just the aperture. Both the EF 50/1.2 and 85/1.2 are lenses I've used extensively. When I first bought the 85 I shot it all the time at 1.2. Looking back now I realise that was a mistake. Both of those lenses have a lovely look that isn't dependent on being wide open. The bokeh is nice, the fall off in focus graceful, the colours and contrast rich and deep.

Someone on this forum actually did a comparison - I think it was the 50L vs the 50 1.4. Shooting both at 1.4 or 2.8, the L still produced softer bokeh.

These days if I am taking a portrait it's rare I will use a lens wide ope, there is usually little to be gained from it. The 135L in particular I find looks great at f/2.8.
 

ken

Engineer, snapper of photos, player of banjos
CR Pro
Aug 8, 2016
86
94
Huntsville, AL
...
I'm sure the cat's eye distortion will be a bit better....I'm sure the vignetting will be a bit better too. But the way the article reads...it's like they have cured it....and from the images shown...they certainly haven't.
When the Canon engineers said " I believe to a certain extent that users can see the improvements to the bokeh shape being cut out in the peripherals [compared to the shape on a DSLR]." they were not talking about cat's eye distortion. They were talking about bokeh clipping that occurs at f/1.2 along the edges of a photo with the EF lens.

This is a well known phenomenon due to the mirror box and that lens being wide open. See examples here: https://www.dpreview.com/forums/thread/2894543
(or just google "having your bokeh balls chopped off".)

So it does appear, from all of the samples on that page, the RF lens doesn't suffer from this issue.