Canon EF 400mm f/2.8L IS III one of two “Big White Lenses” coming ahead of Photokina [CR3]

sportskjutaren

Pro sports photographer
#61
They could reduce the weight from the current one whilst adding in an extender, but a new model with an extender will obviously weigh more than a new model without an extender! Some people want to cut as much weight and size as possible, though I repeat I don't know how many there are of each opinion.
You got a valid point there.
We will have to wait and see what the official announcement will say.
 
Sep 6, 2017
73
3
#62
You have been arguing for an f/8 600mm, and it has been pointed out to you that it would not AF on some Canon bodies and cannot be used with TCs. In addition, f/8 is above the diffraction limit for APS-C sensors and high resolution ones like the 5DSR and others down the line. f/8 does not AF that well on most bodies. In short, if you think that TCs leave a lot to be desired, which I dispute, a maximum aperturte f/8 leaves far more to be desired.
Aperture limited auto focus applies only to phase detection auto focus. I admit this would be a problem if you exclusively use optical viewfinders for focusing. But Canon has fielded a solution to this problem for many years now. It's called Dual Pixel Auto Focus. DPAF is not limited by small apertures, it's more accurate in subject tracking than conventional PDAF, and it doesn't front or back focus as PDAF systems routinely do. Focusing with an LCD screen and a loupe, or an EVF is just as effective as an optical viewfinder.

And diffraction limit? You're implying that images shot at or above f/8 on 24MP APS-C or a 5DSR all of the sudden turn to mush? I haven't noticed that yet.
 
Sep 6, 2017
73
3
#63
As for contrast, can't that be easily dialled back in during processing?
I've only found it possible to obtain a usable image in post-processing using added contrast if I overexpose by about 2/3 of a stop or more, and have a very well balanced histogram on all of my RGB channels. I'll almost always prefer using a bare lens and cropping to my desired focal length in post than using a TC and boosting contrast. That's just me though.
 

neuroanatomist

Spends too much time on this forum
Jul 21, 2010
23,317
365
#64
Aperture limited auto focus applies only to phase detection auto focus. I admit this would be a problem if you exclusively use optical viewfinders for focusing. But Canon has fielded a solution to this problem for many years now. It's called Dual Pixel Auto Focus. DPAF is not limited by small apertures, it's more accurate in subject tracking than conventional PDAF, and it doesn't front or back focus as PDAF systems routinely do. Focusing with an LCD screen and a loupe, or an EVF is just as effective as an optical viewfinder.
The point is that as long as Canon sells DSLRs with an f/5.6 PDAF limit, they’re not going to release lenses that aren’t compatible with that limit. An EF 600mm f/8 lens is not going to happen.
 
Feb 28, 2013
1,355
8
#65
Please don't forget that with the announcement of the Sony FE 2.8/400 mm the limits of weight have been shiftet quite a lot.
The Sony lens is amost 1 kg (!) lighter than the Canon.
Togehter with the A9 you gain about 1.8 kg over the Canon combo.
(1DX II + EF2.8/400 II; disclaimer: this is a pure weight comparison, not about features or IQ)
Maybe Canon sees the need here to put that lens on a diet.

Of course, production costs are always an important factor, as well.
Sony wants to replace Canon as "the professionals camera of choice" and is mimicking all the lenses Canon do for professionals. What I don't see is true innovation and ground-breaking lens designs so whilst the "soft targets" are being replicated that's not really moving the game on such as the Canon EF 11-24mm f4L did with its rectilinear design.
 
Mar 2, 2014
26
4
#66
I find it highly unlikely we will know anything about its durability until long after the release date. We might have a few people disassemble one to make qualitative assessments in September or October, but unless Sony publishes its reliability engineering (which I’ll note is a discipline based in statistics), anything conclusive will take time and a sizeable sample.
Time will tell, there will be a lot of assessment going through the lens. Roger Cicala at LensRental may do the lens dismantling assessment as well.
 

AlanF

EOS 5DS R
Aug 16, 2012
3,956
193
#67
Aperture limited auto focus applies only to phase detection auto focus. I admit this would be a problem if you exclusively use optical viewfinders for focusing. But Canon has fielded a solution to this problem for many years now. It's called Dual Pixel Auto Focus. DPAF is not limited by small apertures, it's more accurate in subject tracking than conventional PDAF, and it doesn't front or back focus as PDAF systems routinely do. Focusing with an LCD screen and a loupe, or an EVF is just as effective as an optical viewfinder.

And diffraction limit? You're implying that images shot at or above f/8 on 24MP APS-C or a 5DSR all of the sudden turn to mush? I haven't noticed that yet.
I am well aware of DPAF - it's the major means of focus on my M5 (as well as an alternative for my other bodies).Have you tried DPAF for moving wild-life and birds? DPAF is very fine for video and static subjects but not for action.

Of course I am not implying that an image turns to mush suddenly at the DLA. You lose resolution linearly with f-number as you go through the DLA for diffraction limited lenses. Look at the MTFs of high quality lenses on say photozone.de: the Canon 35mm f/1.4 II resolves on the 5DSR 5562 lines/height at f/2.8, 5425 at f/4, 5134 at f/5.6 and 4685 at f/8. http://www.opticallimits.com/canon_eos_ff/964-canon35f14mk2?start=1. There is a similar loss of resolution for the Sigma 85mm f/1.4 http://www.opticallimits.com/canon_eos_ff/1000-sigma85f14art?start=1. The 35mm f/1.4 drops 16% in resolution from f/2.8 to f/8.
 

Mikehit

EOS 5D Mark IV
Jul 28, 2015
3,016
222
#68
As I understand it, diffraction is dependent on the size of the pupil diameter, and the f number is a calculation of pupil diameter and focal length so the pupil diameter of f2.8 on an 85mmlens is different to the pupil diameter of f2.8 on a 35mm lens.
Then the two lenses are tested on different bodies which adds another variable on the limit of lpm resolution - the only real way to even attempt this is on a test bench which very few people do.
Add the MTF 50 is about micro contrast, so to say that it "...can be taken as a measure for sharpness..." in the way the article uses it is a gross oversimplification.

I find MTF charts useful for looking at the characteristics of a given lens but is very elastic when comparing lenses.
 

AlanF

EOS 5DS R
Aug 16, 2012
3,956
193
#69
As I understand it, diffraction is dependent on the size of the pupil diameter, and the f number is a calculation of pupil diameter and focal length so the pupil diameter of f2.8 on an 85mmlens is different to the pupil diameter of f2.8 on a 35mm lens.
Then the two lenses are tested on different bodies which adds another variable on the limit of lpm resolution - the only real way to even attempt this is on a test bench which very few people do.
Add the MTF 50 is about micro contrast, so to say that it "...can be taken as a measure for sharpness..." in the way the article uses it is a gross oversimplification.

I find MTF charts useful for looking at the characteristics of a given lens but is very elastic when comparing lenses.
No Mike. The size of the Airy Disk (diffraction spot) depends on on the focal length divided by the size of the pupil diameter, that is the f-number. It is the same size for any lens of the same f-number, independent of its focal length. See for example the equations in https://en.wikipedia.org/wiki/Airy_disk. The diameter of the Airy disk is ~1.22*wavelength*f-number.
 

Mikehit

EOS 5D Mark IV
Jul 28, 2015
3,016
222
#70
Thanks Alan.
I stumbled across this which explains it:

https://www.cambridgeincolour.com/tutorials/diffraction-photography.htm

Technical Note: Independence of Focal Length
Since the physical size of an aperture is larger for telephoto lenses (f/4 has a 50 mm diameter at 200 mm, but only a 25 mm diameter at 100 mm), why doesn't the airy disk become smaller? This is because longer focal lengths also cause light to travel farther before hitting the camera sensor -- thus increasing the distance over which the airy disk can continue to diverge. The competing effects of larger aperture and longer focal length therefore cancel, leaving only the f-number as being important (which describes focal length relative to aperture size).
 
Likes: bfbn
Feb 19, 2016
79
2
#71
I think a new 400mm 2.8 is inevitable.

Canon won't stop trying to improve such lenses anyway. But Sony threw down the gauntlet with their 400 2.8 GM. Not only is it a lot lighter but the weight is distributed much more to the centre so by all accounts it's far nicer to hold and shoot with.

I think in general Canon tends to follow its own path but when it comes to lenses like 70-200 2.8 and 300 and 400 2.8 lenses they will be keen to always fight back and stay ahead of the competition. To be known among professionals as having the best line up of telephoto lenses is great for the brand loyalty and prestige of Canon.

Sony really have got some great lens designers, that is beyond dispute. What I am curious about is whether they have improved quality control and manufacturing tolerances. Their 16-35 2.8 GM for example is probably the best wide angle 2.8 zoom on the market and on paper better than the Canon 16-35 L III. However lots of people buy them finding them to be badly de-centered at the longer end. The Sony Zeiss 35 1.4 is a lovely design, similar to the old Rollei/Contax 35 with a beautiful rendering but lots of credible people on fredmiranda report finding that after trying 4 or 5 copies they couldn't get a good one.

Surely Sony can't hope to sell a lens like the 400 GM with sloppy build quality but regardless it's an area I expect Canon (and Nikon) to stay ahead in.
 

Antono Refa

EOS Rebel SL2
Mar 26, 2014
640
18
#73
I did not mean that we are going to wake up on a certain date within the next 12 months and find that EF is no longer produced, sold or serviced, but that we are in a time of transition--with EF already planned for a gentle phase out. Certainly there have been tweaks in the pipeline, some small, some large, and they are being introduced now. But if the best possible mirrorless Canon can build to compete and excite involves a new mount, EF will no longer be the sole focus of lens development, obviously. EOS will remain for sale for some years, and adapters will extend the use of EF beyond that.
And Canon will replace EF with some other mount because... ?
 

docsmith

EOS Rebel T7i
Sep 17, 2010
735
58
#74
I was wondering a bit on this rumor. It could be the 400 III looks a heckuva lot like the 400 II. But it is a bit odd that no one has seen one of these floating around in the wild.
 

TAW

EOS M6
Oct 5, 2012
106
0
Oregon, USA
#77
I would love to see a 400 with a built in extender (don't expect to but can hope). My real hope is the next version of the extenders could remain independent of the lens (like they are today) but have an engage / disengage feature similar to the 200-400. I am not sure if that is optically possible? Regardless, if it is possible, I think the ability to have the extender on the lens and be able to engage and disengage it (without having to physically remove it) would be really valuable (and something I would be willing to pay nicely for...)
 

neuroanatomist

Spends too much time on this forum
Jul 21, 2010
23,317
365
#78
My real hope is the next version of the extenders could remain independent of the lens (like they are today) but have an engage / disengage feature similar to the 200-400. I am not sure if that is optically possible?
Technically possible, but practically not going to happen because of a very undesirable result. Such an extender, with the optics disengaged, would essentially be an extension tube – providing a shorter minimum focus distance with a concomitant increase in max magnification, but at the cost of preventing distant/infinity focus. Since you generally use a supetele lens with distant subjects, a switchable TC that prevents you from focusing on those distanct subjects when disengaged is basically a non-starter.

With the 200-400mm lens, the TC is in front of the rear element groups, not behind them as an external TC is placed.