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

AJ

EOS RP
Sep 11, 2010
760
177
It’s just hard for me to get over the ugly reality that before it was the impeccably sharp and distortion free image I can view and enjoy, it used to be a bunch of ugly electrons!
And prior to that the image consisted of a bunch of hideous photons!
 
  • Like
Reactions: InchMetric

LSXPhotog

Automotive, Motorsports, Commerical, & Real Estate
CR Pro
Apr 2, 2015
554
581
www.diossiphotography.com
Also, I would like to add that I thought the lens was a little overpriced before. But now I think it’s significantly overpriced. What a shame. But jokes on me as I will keep mine ordered and use it basically every day and nobody will know it has these flaws but me.
 

mb66energy

EOS 5D Mark IV
Dec 18, 2011
1,531
385
Germany
www.MichaelBockhorst.de
RF 4.0 14-35: No adaptors, compact, 2 extra-mm, AF well suited for video
EF 4.0 16-35: Works on all Canon cameras, much cheaper, less distortion (!), can be used with C70 (C50?) as ultra wide with .71x-Adapter ...

Not easy to make decisions for the long term - maybe I should give up, sell all my gear and start from scratch ...
 

dominic_siu

EOS R5, RF2870, RF70200 2.8, RF50 1.8, RF100 Macro
Aug 31, 2018
85
68
I may need to reconsider if I would buy this lens after reading through Bryan’s review.:cry:
 
  • Like
Reactions: mb66energy
Aug 7, 2018
356
306
I wonder if the lens could be corrected optically if it was larger and heavier. Maybe a large curved front element would help. Why do some 14mm or 15mm lenses have a such an etreme front element that does not allow regular filters, but others have not? Is one optical formula better than the other?

So is the RF 14-35 even wider than 14mm uncorrected?
 

mb66energy

EOS 5D Mark IV
Dec 18, 2011
1,531
385
Germany
www.MichaelBockhorst.de
I wonder if the lens could be corrected optically if it was larger and heavier. Maybe a large curved front element would help. Why do some 14mm or 15mm lenses have a such an etreme front element that does not allow regular filters, but others have not? Is one optical formula better than the other?

So is the RF 14-35 even wider than 14mm uncorrected?
I think it is always a compromise between all lens errors: sharpness/res/contrast (including CA, LOCA, coma) is IMO the main design criterion. The next two are distortion + vignetting, then comes bokeh + sunstars.
Just take a look at an ultra wides lens without back cap from the front, use a brighter background to see the aperture. If you look straight into the lens, you see a small aperture which increases if you look at greater angles into the lens - helps against vignetting. For that, a bulbous front element helps, it bends the light stronger and the effective aperture for the image corners is larger.
But it makes lenses bigger, heavier, more expensive and sometimes denys the use of standard filters.

I think there is a movement in technology to optimize hardware AND post processing in conjunction. With today's dynamic range (is it an allowed term anymore :) correcting vignetting isn't that problematic and with a high res cam it is easy to correct the distortion but you always loose image quality AND image field (or effective mm) if you use distortion correction.

That is one of the reasons I will stay with the EF-S 10-22 which is maybe not the sharpest lens in terms of resolution while maybe good enough for 4k but it has low distortion, very good contrast/color and reasonable flare resistance.
 
Aug 7, 2018
356
306
I think there is a movement in technology to optimize hardware AND post processing in conjunction.
That sounds that the manufacturers do not have to bother about errors so much anymore, because post processing can correct that later. That is a very bad developement, because if the lens was great in the first place, it could still be improved by post processing. So a good lens would still be better after post processing.
 

Antono Refa

EOS R
Mar 26, 2014
1,337
445
I think it is always a compromise between all lens errors: sharpness/res/contrast (including CA, LOCA, coma) is IMO the main design criterion. The next two are distortion + vignetting, then comes bokeh + sunstars.
Yes, its easy to post process nowadays, and lens design is a compromise.

For 50% extra price, I'd expect a different balance of weight, size, and IQ. I might change my mind when price drops, but the more time passes the less interested I am in upgrading. The Galaxy S21 Ultra has 12MP raw which is good enough for me to often prefer to let my back rest.

IMHO, cheap, well performing 10-24mm + 16-35mm makes more sense than expensive less well performing 14-35mm
 

canonmike

EOS R6
CR Pro
Jan 5, 2013
454
392
Yes, its easy to post process nowadays, and lens design is a compromise.

For 50% extra price, I'd expect a different balance of weight, size, and IQ. I might change my mind when price drops, but the more time passes the less interested I am in upgrading. The Galaxy S21 Ultra has 12MP raw which is good enough for me to often prefer to let my back rest.

IMHO, cheap, well performing 10-24mm + 16-35mm makes more sense than expensive less well performing 14-35mm
I decided to order a copy of the 14-35, while I wait for the preferred Rf 10-24.
 

InchMetric

Switched from Nikon. Still zooming the wrong way.
CR Pro
Jun 22, 2021
159
187
That sounds that the manufacturers do not have to bother about errors so much anymore, because post processing can correct that later. That is a very bad developement, because if the lens was great in the first place, it could still be improved by post processing. So a good lens would still be better after post processing.
This is a psychology problem. And Canon presumably has the data about the percentage of their potential customers who lose sleep (and sales) over how the lens performs when you turn off the intended features of the system the lens was designed to work with.
To call this an “error” is a misplaced insult to the designers. It’s an engineering design choice to make the lens perform better in other respects that most customers actually care about.
The complaint isn’t about design errors, it’s about design trade off decisions that one customer might not agree with.
Imagine the hypothetical that the same size weight sharpness cost lens could be made with half the uncorrected distortion if it was made without the ability to take filters. Is shooting to produce this design an “error”?
 
  • Like
Reactions: RaPhoto

InchMetric

Switched from Nikon. Still zooming the wrong way.
CR Pro
Jun 22, 2021
159
187
I decided to order a copy of the 14-35, while I wait for the preferred Rf 10-24.
Similarly, I’m holding the RF 15-35 f2.8 while awaiting the 10-24. Except that I’m finding that 35 awfully useful for normal shots while the ultra wide capability is on the lens.
 

neuroanatomist

I post too Much on Here!!
CR Pro
Jul 21, 2010
26,118
4,709
Yes, its easy to post process nowadays, and lens design is a compromise.
It’s not just post processing, it’s real-time processing.

If this lens was mounted on a DSLR, someone looking through the OVF at 14mm would see the view looking more like a fisheye than a rectilinear lens, mutter a few choice expletives, and return the lens.

Since this is for MILCs, you’ll never actually see through the lens, and the geometric corrections are applied to what you see in the EVF. Canon can make it better in terms of features (wider, smaller filter, smaller/lighter lens), and charge more for those features. Relying on software instead of glass to correct aberrations means the smaller/lighter lens is cheaper to produce.

Higher price + cheaper to produce = more profit. Compromise? Maybe for us. For Canon, it’s a solid win.
 
  • Like
Reactions: RaPhoto and AJ

canonmike

EOS R6
CR Pro
Jan 5, 2013
454
392
Similarly, I’m holding the RF 15-35 f2.8 while awaiting the 10-24. Except that I’m finding that 35 awfully useful for normal shots while the ultra wide capability is on the lens.
Yep. Lovin all the new big boy R and RF toys, as we wait for our more relevant ones to be released.
 
Aug 18, 2021
1
2
Sounds like Canon chose a very similar set of compromises to Nikon's 14-30 f4. In Nikon's case the corrections are built in and unable to be turned off even in lightroom, so I'd say at least having the option to turn them off is nice for scenes that might not require it. Below (via photographylife.com) is the raw image uncorrected from the 14-30 also at 14 and it looks very similar. Given the Canon goes out to 35, has built in IS and appears at least as sharp if not more so than the Nikon according to TDP, I'd say it looks like a great option. When the Nikon came out everyone said how great it was that there was a small and light 14mm capable zoom for mirrorless and now the Canon equivalent comes out and people are complaining it isn't as good as the Sigma 14-24 that has a shorter range, no IS and is literally double the weight.
 

Attachments

  • Distortion-and-Vignetting-at-14mm-f4.jpg
    Distortion-and-Vignetting-at-14mm-f4.jpg
    97.2 KB · Views: 67

sanj

EOS R5
Jan 22, 2012
3,963
863
I wanted to get this. But now won't. I think my 15-35 2.8 is a better option.
 
  • Like
Reactions: tron

lawny13

EOS 90D
Mar 6, 2019
102
73
It’s not just post processing, it’s real-time processing.

If this lens was mounted on a DSLR, someone looking through the OVF at 14mm would see the view looking more like a fisheye than a rectilinear lens, mutter a few choice expletives, and return the lens.

Since this is for MILCs, you’ll never actually see through the lens, and the geometric corrections are applied to what you see in the EVF. Canon can make it better in terms of features (wider, smaller filter, smaller/lighter lens), and charge more for those features. Relying on software instead of glass to correct aberrations means the smaller/lighter lens is cheaper to produce.

Higher price + cheaper to produce = more profit. Compromise? Maybe for us. For Canon, it’s a solid win.

Nailed it...

You are the first to point out the fact that MILC allows for the lens to be corrected in real-time. This is also the reason for always on IS rather than just when you take the shot.

As for other ppls comments about the loss of mm after correction, we already know from other corrected canon lenses that after correction it is 14mm while the images without correction is in fact wider to allow for the correction.

Canon and other manufacturers intend to their lenses AND cameras... so their whole system to deliver images of a certain quality. And generally they do. There are very few lenses that don't need any distortion correction right? Even good EF lenses need a bit of pin-cushion or barrel distortion correction. So since people are going to have those corrections turned on anyways, what is the big deal if the trade off for more distortion in an uncorrected lenses is the fact that I can use the same filters on the 14-35, as I do on my 24-105 f4, and 100-500??

And of course to reiterate. The image I see on my LCD or EVF is corrected, and what I see is what I get when I press the shutter....

The one thing I don't exactly agree with is the price hike. I can understand that canon will charge more. But IMO it is a bit too far. Here in Europe, the RF 14-35 costs 1850 vs 1050 (euros) compared to the EF 16-35.

Whereas the Rf 15-35 f2.8 is 2450 euros vs the EF 16-35 f2.8 at 2050.

The 800 euro difference on the f4s is ridiculous. In my book that extra 2mm isn't worth an 80% hike.
 
Last edited: