Patent: Canon RF 15-28mm f/4-5.6 IS and other wide angle optical designs

Jul 21, 2010
31,095
12,857
@neuroanatomist Indeed TDP shows them similar good catch. The only thing is that 16-35 has visibly lower vignetting than the RF version. This may or may not be a problem depending on the ISO one shoots at. If someone wants to shoot at low light with no tripod or with tripod but high iso (say for astro landscape work) then this would be a problem. Unfortunately the same applies to RF15-35 too. It is worse that the EF 16-35 2.8L III which is worse than 14mm 2.8L II.
Looking at TDP, it’s really not much worse. The EF 16-35/4 has a bit over 2 stops at 16mm, the RF 16-35/4 has a bit under 2.5 stops at 14mm, so probably something like 1/3-stop difference. Not really meaningful, IMO.

Again from TDP, the EF 14/2.8 II has over 3 stops of vignetting, and the EF 16-35/2.8 III has over 4 stops; stopped down to f/4, it’s still >3 stops. The EF 24/1.4 II and 11-24/4 both have over 4 stops, as well.

Based on those numbers, I’m not seeing how the RF 14-35/4 has a ‘problem’ with vignetting.
 
  • Like
Reactions: 1 user
Upvote 0

tron

CR Pro
Nov 8, 2011
5,222
1,616
I am referring to situations where we have to shoot wide open at high iso like astro or handhelding at low light. Then the compensation is too much. I believe the difference seems a little more. In fact you can see in image quality comparisons that the edges (marked as periphery) in 14-35 are definitely more grey.


Unless the pictures are subject to variation.

EDIT: If you check at the point where the RF has -2.07 stops vignetting the EF has -1.5 stops. That's between 1/2 and 2/3rds of a stop.


I didn't use it initially because it compares 14mm to 16mm so maybe the difference is indeed as you say or between the 2 numbers at 1/2 stop.

Regarding the III yes unfortunately it has more vignetting and the RF version is even worse.

So maybe after all the price in comparison to all of these lenses is the worst thing and nothing more.
 
Last edited:
Upvote 0
Jul 21, 2010
31,095
12,857
I am referring to situations where we have to shoot wide open at high iso like astro or handhelding at low light. Then the compensation is too much.
This discussion started with @dilbert bashing the RF 14-35/4, and your agreement with his issues.

The RF15-35/2.8, EF 16-35/2.8, EF 14/2.8, EF 24/1.4, EF 11-24/4 all have worse vignetting wide open than the RF 14-35/4, and thus require even more gain to compensate.

The EF 16-35/4 has ~1/3-stop less vignetting than the RF 14-35/4. Are you honestly saying that applying an additional 1/3-stop of gain in the corners is ‘too much’? Sorry, that’s patently ridiculous. Like those who claim the Canon 800/5.6 is far superior to the Nikon 800/6.3 because of that same 1/3-stop difference, that just sounds asinine and in that case it’s about the entire frame, not just the extreme corners.

Also, if you shoot the 14-35/4 at 14mm then crop it to a 16mm FoV, it has less corner vignetting than the EF 16-35/4…probably by about that same 1/3-stop you believe is ‘too much’. Plus, cropping away the corners gets rid of the worst coma area, so for astro use I’d say the RF 14-35/4 is a better choice than the EF 16-35/4, if one needed to choose just between those two.

Regardless, I don’t think there’s really much merit in debating these two f/4 lenses for astro use. The f/2.8 UWA zooms have over a stop worse corner vignetting, but give an extra stop across most of the frame. The 24/1.4 gives 3 stops across most of the frame. Even something like the 28-70/2 loses a stop based on FoV (i.e., using the 500/FL exposure duration approximation), but still gains a net stop over the 14-35/4.
 
  • Like
Reactions: 1 user
Upvote 0
Jan 27, 2020
826
1,796
The problem is when Canon produces lenses that generate images with severe problems, you aren't getting lenses that are cheaper (maybe lighter.) Look at the cost of the RF 14-35/f4L vs the 16-35 f/4L. 2mm wider, massive increase in distortion and a massive price increase but only 100g weight savings (the weight of a chocolate bar.)

With this lens, the suggestion is that the image circle doesn't cover the full frame and that's not a problem because in-camera electronics will just correct for the resulting vignetting and distortion. If the camera needs to boost the peripheral parts of the image by 2+ stops, there's no way that can't impact the final image quality. The better the quality of the input, the better the quality of the output.

With what is effectively MkI lenses for RF from Canon, photographers at the wide end are getting fucked over, pure and simple.
You can't compare a 16 vs 14mm lens at the wide end and consider them equal in any way. Plus we are dealing with a changing market, inflation and global shortages, so comparing prices is not really meaningful either. If you don't like auto digital correction, then why not just say that you don't like auto digital correction. That's fine. You don't have to speak for other photographers - most (it appears) who don't mind, since the image quality appears to be as good as older lenses where all the corrections are made by using more and more glass elements. I've tried the new RF 300mm f/2.8. It has a huge amount of digital correction. It's also comparatively as sharp as the EF 16-35mm f/4 L. So I am not get screwed. And my guess is most other photographers do not feel like they are getting screwed. So it is not pure and simple, which makes your statement wrong...pure and simple.

And I assume you are aware that the "f" word is not one to be used in polite company. Your continued use an overly strong word only shows that your argument is weak.
 
Upvote 0
Jul 21, 2010
31,095
12,857
You don't have to speak for other photographers - most (it appears) who don't mind, since the image quality appears to be as good as older lenses where all the corrections are made by using more and more glass elements.
I find that the digitally corrected RF 14-35/4 at 14mm is on par with the EF 11-24/4 at 14mm, a lens that isn’t that old, costs twice as much, and isn’t at the end of its zoom range.

I've tried the new RF 300mm f/2.8. It has a huge amount of digital correction.
Wait, what? The new RF 300mm f/2.8?!? What new RF 300mm f/2.8???
 
Upvote 0

tron

CR Pro
Nov 8, 2011
5,222
1,616
This discussion started with @dilbert bashing the RF 14-35/4, and your agreement with his issues.

The RF15-35/2.8, EF 16-35/2.8, EF 14/2.8, EF 24/1.4, EF 11-24/4 all have worse vignetting wide open than the RF 14-35/4, and thus require even more gain to compensate.

The EF 16-35/4 has ~1/3-stop less vignetting than the RF 14-35/4. Are you honestly saying that applying an additional 1/3-stop of gain in the corners is ‘too much’? Sorry, that’s patently ridiculous. Like those who claim the Canon 800/5.6 is far superior to the Nikon 800/6.3 because of that same 1/3-stop difference, that just sounds asinine and in that case it’s about the entire frame, not just the extreme corners.

Also, if you shoot the 14-35/4 at 14mm then crop it to a 16mm FoV, it has less corner vignetting than the EF 16-35/4…probably by about that same 1/3-stop you believe is ‘too much’. Plus, cropping away the corners gets rid of the worst coma area, so for astro use I’d say the RF 14-35/4 is a better choice than the EF 16-35/4, if one needed to choose just between those two.

Regardless, I don’t think there’s really much merit in debating these two f/4 lenses for astro use. The f/2.8 UWA zooms have over a stop worse corner vignetting, but give an extra stop across most of the frame. The 24/1.4 gives 3 stops across most of the frame. Even something like the 28-70/2 loses a stop based on FoV (i.e., using the 500/FL exposure duration approximation), but still gains a net stop over the 14-35/4.

1. Agree for astro. The 2.8 have the advantage. 14mm 2.8 II can also be used with cropping. He has more coma than both 16-35 but it seems brighter at the edges and it should be even more if we crop at 16 using that logic. But cropping somehow changes the comparison.

2. I already admitted that TDP specific vignetting comparison was not 100% apples to apples but that was available at TDP so there was no way to compare them at 16mm

3. If you believe TDP lens quality comparisons the difference in brightness at the edges of the two lenses is very obvious (1/3rd or not). And these comparisons were made at 16mm.

I repeat my post copying from above so as to see the part in my agreement which of course does not compare with Dilbert's use of language:

"I would accept 14-35's design if it was much cheaper than it is. Not when it costs almost as a much better design 2.8L lens.
Anyway people who do not have EF lenses (or the RF15-35) and can afford it may as well buy it. I understand they will be happy with its IQ and they will soon forget how much they paid for it."

So Vignetting maybe 1/3 of a stop worse but as I mentioned it shows more in the comparison and that comes from a much more expensive lens.
 
Upvote 0
Jul 21, 2010
31,095
12,857
3. If you believe TDP lens quality comparisons the difference in brightness at the edges of the two lenses is very obvious (1/3rd or not). And these comparisons were made at 16mm.

So Vignetting maybe 1/3 of a stop worse but as I mentioned it shows more in the comparison and that comes from a much more expensive lens.
It's ~1/3-stop. Honestly, looking at the two images there's really not much difference at the corners. That's confirmed by bringing them into PS, where the 8-bit brightness values for the corners of the 14-35 are ~44 and for the 16-35/4 are ~42. The centers of the images are slightly different, with the 14-35/4 being a bit brighter at a mean of ~123 and the 16-35/4 at ~115 (those values are for a 20x20 pixel square). To properly use the tool, you need to look at the EV differential shown by the labeled circles that are superimposed by the Imatest analysis software – those are measuring the EV difference from the center of the image, so slight differences in illumination from shot to shot (and camera to camera) are normalized.

I think you're seeing something that isn't really there, perhaps an optical illusion of the different scales used for the two images (the circles for the 14-35 are in 1/3-stop increments, for the 16-35 they are at 1/2-stop increments so there are fewer of them).

"I would accept 14-35's design if it was much cheaper than it is. Not when it costs almost as a much better design 2.8L lens.
Anyway people who do not have EF lenses (or the RF15-35) and can afford it may as well buy it. I understand they will be happy with its IQ and they will soon forget how much they paid for it."
The RF 15-35/2.8 costs 40% more than the RF 14-35/4, I wouldn't call that 'almost as much'. Having said that, for me personally the difference is not financially significant. I chose the 14-35 over the 15-35 because I really don't need f/2.8 in a UWA zoom (I had the 16-35/2.8 II for several years, and under 1% of my shots with it were at f/2.8), because it's significantly smaller and lighter, and because it uses 77mm filters.

That last point is relevant to the discussion of travel lenses – the combination of the RF 14-35/4, RF 24-105/4 and RF 100-500 gives coverage over a huge range of focal lengths in a relatively compact and light kit that all take the same filters.
 
  • Like
Reactions: 1 user
Upvote 0
That last point is relevant to the discussion of travel lenses – the combination of the RF 14-35/4, RF 24-105/4 and RF 100-500 gives coverage over a huge range of focal lengths in a relatively compact and light kit that all take the same filters.
Let's not forget the RF70-200/2.8 also has 77mm filters and is small enough for travel now :)
That said, all the newer filter systems have step up rings and a large built in CPL. They don't handle 14mm without vignetting though. Most promise 16mm only so any added filters including screw-on ND filters will add more vignetting <16mm.
 
Upvote 0
Jul 21, 2010
31,095
12,857
Let's not forget the RF70-200/2.8 also has 77mm filters and is small enough for travel now :)
True, although for me I generally want the extra reach for landscape compositions or opportunistic wildlife. I use my RF 70-200/2.8 mainly for indoor events, and the only filter I’ve used is the clear one I keep on there.

That said, all the newer filter systems have step up rings and a large built in CPL. They don't handle 14mm without vignetting though. Most promise 16mm only so any added filters including screw-on ND filters will add more vignetting <16mm.
Indeed. I use B+W filters, and I have CPLs and several NDs in both 77mm and 82mm. And I have the step up rings. And the filter wrenches that are sometimes needed when I use them. ;)

But having all the lenses in a travel kit take the same filter size is very convenient. Also, I’d far rather take round filters than a square setup.

I used to travel with 145mm filters for my TS-E 17, but the drop-in RF adapter means I no longer need to.
 
Upvote 0
Aug 12, 2010
169
172
You can't compare a 16 vs 14mm lens at the wide end and consider them equal in any way. Plus we are dealing with a changing market, inflation and global shortages, so comparing prices is not really meaningful either. If you don't like auto digital correction, then why not just say that you don't like auto digital correction. That's fine. You don't have to speak for other photographers - most (it appears) who don't mind, since the image quality appears to be as good as older lenses where all the corrections are made by using more and more glass elements. I've tried the new RF 300mm f/2.8. It has a huge amount of digital correction. It's also comparatively as sharp as the EF 16-35mm f/4 L. So I am not get screwed. And my guess is most other photographers do not feel like they are getting screwed. So it is not pure and simple, which makes your statement wrong...pure and simple.

It doesn't matter if 14 & 16mm are different focal lengths.

What matters here is that Canon are now designing lens that are flawed and are "fixing" those flaws with software and hoping you the customer won't care or notice. Previously these flaws would have resulted in a lens not being released.

Canon are designing optically inferior lenses and charging consumers more for that privilege - all in the name of reduced weight (negligable) and size (also negligable).

Right now, photographers have no choice: they either buy what Canon offers in terms of RF lenses or they have nothing.

Canon has its photographers over a barrel.
 
  • Like
Reactions: 1 users
Upvote 0
Jul 21, 2010
31,095
12,857
It doesn't matter if 14 & 16mm are different focal lengths.

What matters here is that Canon are now designing lens that are flawed and are "fixing" those flaws with software and hoping you the customer won't care or notice. Previously these flaws would have resulted in a lens not being released.

Canon are designing optically inferior lenses and charging consumers more for that privilege - all in the name of reduced weight (negligable) and size (also negligable).
Right now, photographers have no choice: they either buy what Canon offers in terms of RF lenses or they have nothing.

Canon has its photographers over a barrel.
So in your personal reality, the multiple variations of the EF-RF mount adapter simply don't exist, and thus it's not possible for photographers to buy a new or used EF lens to use with an R-series body. I guess it's a good thing for the rest of us that you're the only one living in your personal, altered reality.

Canon is designing lenses for the bodies they'll be used on. Digital correction is less expensive, and relying on it can save weight and size in an optical design. It can also save on production costs, and since Canon is adding other features (e.g. 2mm on the wide end) they can justify charging more for the lens, and higher price with lower cost means more profit. If that offends you, don't buy the lens. Since the RF 14-35/4L and especially the RF 16/2.8 seem quite popular, your wallet vote is probably irrelevant.

Incidentally, 14 vs. 16mm does matter, because an optically-corrected 14-35 would like have a bulbous front element and either not take front filters or need massive ones like the Nikon Z 14-24. Personally, I'd prefer the digital corrections and a compact lens that takes convenient 77mm filters.
 
  • Like
Reactions: 1 user
Upvote 0

Sporgon

5% of gear used 95% of the time
CR Pro
Nov 11, 2012
4,719
1,537
Yorkshire, England
As a photographer, all I care about is the final image. If the lens can be made cheaper and lighter by having digital correction in camera or automatically done in software, it doesn't bother me a bit. My guess is that most folks would gladly choose lighter and cheaper as well. That's what automatic digital correction is for - to make a cheaper and lighter lens. If the end result is a sharp, quality image, what's the problem?
Are you a FF user ?
 
Upvote 0
A lot of people should reconsider their lens choices now that high resolution sensors are becoming more common.
You may be better served with prime lenses and cropping. My choice for lenses are 20mm and 85mm.
A 20mm f/1.4 (or 1.8) is far superior than the zooms and yet can be cropped to 40mm if necessary leaving reasonable resolution for most applications.
An 85mm f/1.4 (or 1.8) will give amazing portraits and can be cropped all the way to 170 maintaining an effective aperture of f/2.8.
Some people may want three lenses, like: 20mm (or 14mm), 35mm (or 50mm), and 135mm.
I submit that this strategy would be almost as flexible as the zoom trinity at much lower weight and in many (if not most) cases give better results (especially for those images which do not need to be cropped much and in lower light circumstances).
 
Upvote 0
At least I can frame the image with the 90D without having to drain the battery.

The problem isn't the lack of a perfect lens, the problem is that Canon is pursuing lRF lens designs that have significant image quality issues that they're just hand waving away with software - and asking customers to pay more for the privilege of such lenses.
True, wish my 28-70 f2 and 85 f1.2 ds that costs around $5000 each had better image quality :ROFLMAO:
 
Upvote 0

LogicExtremist

Lux pictor
Sep 26, 2021
501
352
True, wish my 28-70 f2 and 85 f1.2 ds that costs around $5000 each had better image quality :ROFLMAO:
You've just proven the point Dilbert was making, the lenses you've mentioned aren't under-designed optically to cut corners with a reliance on software correction for extreme optical distortion! Those lenses have some of the best optics in the Canon RF lens range and are properly optically corrected like the better F L-series lenses. Do you own any of the RF 16mm f/2.8, RF 24-240mm or RF 14-35mm f/4 L lenses in question? If not, why not? :)

PS - you don't really want your 28-70 f2 and 85 f1.2 ds, you can do better, send them to me. I'll happily drop my standards and take your inferior lenses off your hands for free! :ROFLMAO:
 
  • Like
Reactions: 1 user
Upvote 0
You've just proven the point Dilbert was making, the lenses you've mentioned aren't under-designed optically to cut corners with a reliance on software correction for extreme optical distortion! Those lenses have some of the best optics in the Canon RF lens range and are properly optically corrected like the better F L-series lenses. Do you own any of the RF 16mm f/2.8, RF 24-240mm or RF 14-35mm f/4 L lenses in question? If not, why not? :)

PS - you don't really want your 28-70 f2 and 85 f1.2 ds, you can do better, send them to me. I'll happily drop my standards and take your inferior lenses off your hands for free! :ROFLMAO:
I was being sarcastic if you didn't catch lmao

Yeah I will just give you my most prized lenses that I can't live without. What's your address? I will even deliver them to you.
 
Upvote 0
I too was being sarcastic, if you didn't catch that! lmao ;)
Of course I didn't. So you meant I didn't prove the point Dilbert was making, and the lenses i mentioned are under-designed optically while having the worst optics in the RF range. Then again, you lower/raise? your standards to take my inferior lens off my hands for free. Too confusing
 
Upvote 0