Review: Canon RF 14-35mm f/4L IS USM

Canon Rumors Guy

Canon EOS 40D
CR Pro
Jul 20, 2010
10,765
3,149
Canada
www.canonrumors.com
Bryan at The-Digital-Picture has completed his full review of the Canon RF 14-35mm f/4L IS USM which was announced this past June and will begin shipping in limited quantities this month.
From The-Digital-Picture
The first question on my comparison list was: How does the Canon RF 14-35mm F4 L IS USM Lens compare to the RF 15-35mm F2.8 L IS USM Lens?
… In the image quality comparison equalized at f/4, the two lenses are similar overall, with each having some advantages at specific focal lengths. The f/4 lens has dramatically stronger geometric distortion and has stronger lateral CA. The f/2.8 has stronger peripheral shading in most equalized comparisons.
… As I said at the beginning of the review, the...

Continue reading...


 
Aug 7, 2018
598
549
And yet the output is very impressive.
Maybe, but the difference between a cheap and an expensive lens always was that you need less correction at the expensive lens. Of course today Lightroom and other software handles cromatic aberration quite well for example, but with a good lens it should be lower in the first place. Same with flares and distortion. If I need artificial intelligence to correct it, it should not be that expensive. I own the Tamron 15-30 f/2.8 and it has much less distortion, but only costs half the price. I hope some reviewer will make a comparison between those lenses.

Of course the IS is quite nice. 7 stops of compensation helps a lot in Dubai, where you need a written permission to use a tripod in many areas, as even the streets are owned by developers there. The same is true in more and more cities. More than one second of exposure and still sharp image means I could take handheld night shots at f/4 with ISO 400 or even lower.
 
Upvote 0

fox40phil

People, Events, Sports & Wildlife
Apr 12, 2013
333
214
Germany
www.phileas-schoenberg.de
@Bryan

Why are the comparison shots like here (14-35 vs. 15-35) or vs. Sigma 14 1.8 etc.. not like the example shots from the review?
It seems, that those comparison shots of the new 14-35 are complete perfect (with DPP correction)? And then the stretched shot from the review is anything else then perfect?
So are the comparison shots corrected (only the 14-35?) and the one from the review is not - like the text mean.
1629217864229.png
 
  • Like
Reactions: 1 users
Upvote 0

Joules

doom
CR Pro
Jul 16, 2017
1,801
2,247
Hamburg, Germany
@Bryan

Why are the comparison shots like here (14-35 vs. 15-35) or vs. Sigma 14 1.8 etc.. not like the example shots from the review?
It seems, that those comparison shots of the new 14-35 are complete perfect (with DPP correction)? And then the stretched shot from the review is anything else then perfect?
So are the comparison shots corrected (only the 14-35?) and the one from the review is not - like the text mean.
View attachment 199640
Looks like in the comparison tool, the corrections are not forced off. Note that for example the RF 24-240 mm lens also has its corrections applied.
 
  • Like
Reactions: 1 user
Upvote 0
Jul 21, 2010
31,023
12,776
Yes, that is quite strange. Why does Bryan use a corrected image for the distortion comparison? Of course if the corrections are turned on, there is no distortion at all.
His standard workflow involves shooting RAW and converting with defined settings using DPP. So, if DPP has forced corrections, those corrections will be applied for the comparison tool images.
 
  • Like
Reactions: 2 users
Upvote 0
Jul 21, 2010
31,023
12,776
That is cheating by Canon. I wonder how that can be legal.

What will happen if you do not use DPP, but Lightroom instead? Will the distortion be visible or will Canon produce a fake RAW to hide the distortion?
We already know what will happen, since Bryan posted RAW files opened in Capture One that show the uncorrected distortion. If you have the lens profile installed in C1, LR, etc., then the distortion will be automatically corrected.

The difference is that in DPP, you can not turn the correction off.
 
  • Like
Reactions: 2 users
Upvote 0
I wonder how that can be legal.

What will happen if you do not use DPP, but Lightroom instead? Will the distortion be visible or will Canon produce a fake RAW to hide the distortion?
1) Because they don't claim it is corrected optically. If Canon wrote "this lens is corrected for distortion optically and does not rely on software correction" then yes, this would be illegal; that's the very definition of false advertising. But they don't say that. All they say in their advertising is that it has "excellent optical performance". Since they don't specify distortion, just optics in general, they're technically above-board.
If someone wanted to build a case with trading standards against Canon's advertising, they'd be better off going after the 85mm f/2 IS (which Canon still claims is both "compact" and has "quiet" AF, which anyone who's so much as glanced at it knows are two completely erroneous claims). Even then, that's more indiciative of a wider industry problem than anything specific to Canon. Every manufacturer needs to shape up. (If you think Canon's advertising is misleading, go check out Fujifilm, oh boy.)

2) Lightroom's adherence to Canon's opcodes is pretty random. With some lens and body combinations Lightroom will follow Canon's adjustments, and with other combos Lightroom does entirely its own thing. Capture One more consistently follows Canon's opcodes, though still not every time as Bryan has demonstrated with his particular lens. If you want to have all the corrections, using a manufacturer's own software is always the most accurate solution, though it's very rare for Lightroom to not recieve a 99.9% accurate profile within a couple of weeks of a lens being on store shelves. DxO's Pure Raw is also very good for correction distortion and much better than DPP for cleaning up noise at the same time, though it does result in files twice as large and it's not really worth buying unless you're dead-set on having the most absolutely 100% perfectly-corrected files possible. Raw Therapee is the best software for totally bypassing any opcodes and seeing the raw file as 'raw' as possible.

By the way, every raw is a "fake raw". The R5 has opcodes for noise reduction and exposure boosting, hiding the fact its real sensitivity is much lower than it states. Sony raws have opcodes for sharpening and saturation boosts. Fuji raws have opcodes for damn well everything and anything you can name, but most notably extremely strong chromatic noise reduction. Nikon raws have opcodes for distortion, noise and sharpening, though generally (varies by camera model) to a lesser extent than Sony's sharpening or Fuji's and Canon's noise reduction. I've not pulled apart raw files form Panasonic or Olympus in a good while but I'm willing to bet their files are full of opcodes for all kinds of hidden adjustments, too. There's not actually any such thing as a truely "raw" file. Any raw file you look at has been through some form of processing, which at the bare minimum includes a subjective evaluation of colour and contrast just in order to turn all the 1s and 0s into a picture you can actually see. Raw Therapee can show you a file without demosiacing and opcodes, but then you just get a pixelated mess which you really can't use.



Anyway.


This does still take the piss quite a bit considering how much more expensive this is than the EF equivalent. They're charging £1750 for this, while the EF f/4L IS is £999 (and of course much less used) and third-party equivalents are cheaper again. Granted, this 14-35 is the (fractionally) smaller than the EF and gives you an extra 2mm, but I've never been in a situation where 16mm wasn't already too wide and it's not like the EF lens was ever prohibitively large or heavy. (And there's the Tamron f/2.8-4 which is even smaller and lighter than the RF lens, granted with 3mm less on the widest end.)
For this lens to justify costing 80% more it had to really deliver, and this doesn't seem to, regardless of whether it's relying on software or not. I'm not seeing any extra resolving power here and Bryan's testing demonstrates more fringing than the EF lens even with all the corrections, slightly more vignetting, more flare wash-out, and more distortion... nah, that's not a £1750 lens. I don't care that it's 100g lighter and 1cm shorter, nor do I care that it brings with it an extra 2mm. (Though looking at these results I'm betting it's not a full 14mm, at least not after such heavy correction.) Those elements I'd pay, at most, £150-200 more for, given the end results are about the same quality. £750 extra is a bad, bad joke.

Buy the EF (or the Tamron), and spend the saved money on a trip somewhere you'll actually get use out of the lens, or get some really nice ND filters, or just pocket it and save it for the next lens. Nobody should be paying £1750 for this and I feel very sorry for the poor saps who blindly pre-ordered. Hopefully people will learn, stop pre-ordering lenses, and force manufacturers to actually deliver some real value. These are £999 results in a £1750 shell... just don't do it. Don't. This is silly.


So much for mirrorless providing better optimisation for wide-angle lenses. It's bizarre and somewhat ironic that its been telephoto which has benefitted most since the move to mirrorless, even though telephoto formulas don't actually benefit from the shorter flange distance.
 
  • Like
Reactions: 8 users
Upvote 0
Canon is making very compact lenses, very good. But I think he should pay more attention to ergonomics, specially for professional customers, the three rings seem to be one, especially AF and MF, it's really easy to get confused. Same goes for the 24-105 f4 and 70-200 f4. I would like to have a large zoom ring well spaced from the others.
 
  • Like
Reactions: 2 users
Upvote 0

Jethro

EOS R
CR Pro
Jul 14, 2018
996
1,035
So much for mirrorless providing better optimisation for wide-angle lenses. It's bizarre and somewhat ironic that its been telephoto which has benefitted most since the move to mirrorless, even though telephoto formulas don't actually benefit from the shorter flange distance.
And yet it should be remembered that the actual review ends up concluding "the RF 14-35’s image quality is excellent (I’ll get over the strong barrel distortion aspect), the Nano USM AF system is silent, high-speed, and accurate, and the L-series build quality promises to hold up to the rigors of constant use."

It's at a price level beyond what I need, but that price is pre- the eventual discounting, and provides 'L' build quality in a native RF mount (and full access to IS) for those who do need it. For everyone else, the EF equivalent, or third-parties, are almost certainly better options.
 
Upvote 0

InchMetric

Switched from Nikon. Still zooming the wrong way.
CR Pro
Jun 22, 2021
267
287
Maybe, but the difference between a cheap and an expensive lens always was that you need less correction at the expensive lens. Of course today Lightroom and other software handles cromatic aberration quite well for example, but with a good lens it should be lower in the first place. Same with flares and distortion. If I need artificial intelligence to correct it, it should not be that expensive. I own the Tamron 15-30 f/2.8 and it has much less distortion, but only costs half the price. I hope some reviewer will make a comparison between those lenses.

Of course the IS is quite nice. 7 stops of compensation helps a lot in Dubai, where you need a written permission to use a tripod in many areas, as even the streets are owned by developers there. The same is true in more and more cities. More than one second of exposure and still sharp image means I could take handheld night shots at f/4 with ISO 400 or even lower.
“Always”. Not a good angle.
I judge by the image that I get not the one I have to struggle to obtain. Welcome to the present.
 
Upvote 0
I would like to know the actual focal length at the wide end after taking into account the loss of view caused by the distortion correction. It seems if you are forcing a correction onto the file that the focal length should be the equivalent of 14mm AFTER the correction. I have a feeling this is not the case with this lens.

Dissapointing to see this approach in such an expensive lens...
 
  • Like
Reactions: 3 users
Upvote 0
Dec 25, 2012
750
376
Maybe, but the difference between a cheap and an expensive lens always was that you need less correction at the expensive lens. Of course today Lightroom and other software handles cromatic aberration quite well for example, but with a good lens it should be lower in the first place. Same with flares and distortion. If I need artificial intelligence to correct it, it should not be that expensive. I own the Tamron 15-30 f/2.8 and it has much less distortion, but only costs half the price. I hope some reviewer will make a comparison between those lenses.

Of course the IS is quite nice. 7 stops of compensation helps a lot in Dubai, where you need a written permission to use a tripod in many areas, as even the streets are owned by developers there. The same is true in more and more cities. More than one second of exposure and still sharp image means I could take handheld night shots at f/4 with ISO 400 or even lower.
I think that last mm on the wide end plus the other compromises result in that distortion.
Also, I feel that the Tamron is very nice but maybe not quite as sharp (I have no idea) but the real point being that the last increment of performance is very hard to see in so many cases.
The most frequent source of our disappointment comes from user error.
 
Upvote 0

LSXPhotog

Automotive, Commercial, & Motorsports
CR Pro
Apr 2, 2015
786
976
Tampa, FL
www.diossiphotography.com
I've read the review and I have a lot of thoughts about this lens...and a lot of skin in the game, as I have TWO of them pre-ordered. I don't plan to keep both of them, but I'm desperate to lose the RF adaptor on my EF 16-35 f/2.8III as soon as humanly possible because I hate using it that way...it's obnoxiously long and I am not someone that holds too much value in f/2.8 for an ultra-wide.

Several months ago someone commented on the announcement of this lens and they joked that it would have loads of mandatory digital lens correction. I didn't think that was possible and I said that it was something I could see with their "silver ring" lenses, but never something I could see them doing with an L-series lens...boy was I wrong. But guess what? I looked through the sample images FIRST and then began to read the details on the lens and I had absolutely no idea significant correction was occurring because the lens was sharp as a tack. I'm not too thrilled with the handling of chromatic aberrations either...to be honest, I'm not overly thrilled with a lot of elements in the lenses I own but they best serve the job I need them to.

This is a very small lens. So small, in fact, that I am willing to forgive some of these perceived flaws because it will bring down the size/weight of Canon's ultra-wide lens options tremendously. If you don't value size and weight, then why are you even considering an f/4 lens? Go buy the f/2.8 and stop acting like anything slower than f/2.8 has no place in your bag at any price. LOL

I'm desperate for this lens. I am DONE adapting my EF ultra-wide and I'm eager to get to work with this thing. Is it perfect? No...and the RF 15-35 has some major flaws of its own - just like every other manufacturer's ultra-wide lenses. This, however, is the smallest and lightest I have ever seen, takes a 77mm filter and will be joining my camera bag by the end fo the month.
 
  • Like
Reactions: 6 users
Upvote 0