rscheffler wrote:
Gary, have you compared SOOC jpeg images with your 24-105Z against the same images, but raw versions, processed in Lightroom with corrections applied?
The 24-105Z has fairly strong distortion at 24mm and requires aggressive correction. With the few images I've processed from it so far, at 24mm the SOOC corrected image has a considerably narrower angle of view than the LRC distortion corrected version, which suggests to me that at least for this lens, Canon applies a fairly aggressive crop after distortion correction. At least at the wide end - I haven't yet compared other focal lengths. Perhaps this is also true for other lenses? ...Show more →
Actually, I haven't, but I find this interesting. My experience with the 24 f/1.8 STM is that Adobe exactly matches the Canon corrections, resulting in a narrower FOV compared to other 24mm lenses, whereas DXO leaves considerably more image area, which I prefer.
I had assumed that Adobe would simply follow Canon guidance on this, but maybe not. I will look at this, thanks.
melcat wrote:
I’m not sure we really do know that. All the reviewers seem to think it’s a good thing to test on a high resolution camera because it’s “more demanding”. It might be a different story if you start with only 24Mpx and then stretch those relatively fewer corner pixels.
One might get some idea of how it works on, say, an R3, by taking the uncorrected R5 image, downres-ing to 24Mpx, and only then correcting the barrel distortion.
Yes, not all reviews are in, but the published MTF charts do suggest this is the case. Maybe I am optimistic.
Re. the lower MP bodies, that is an interesting thought that I had not considered. It seems illogical at first that a lower resolution sensor would show worse relative performance given that the lens can comfortably outresolve it, but you are right that there is less data for the resampling. Would be a good thing to test.
PhilH wrote:
But. This lens does have a fair bit of distortion, which already is being "not shown or discussed much" publicly. In camera corrections and post distortion corrections rectifies that. I think that's part of the reason the newer lens is actually closer to probably a 32 or 33 FOV, btw, as that was likely part of the design consideration. For some, this is a non-issue, for others, they will want a good high quality corrected optic, though it would be bigger and more expensive.
My general POV is Canon is dancing the dance of a balancing act with the new hybrid design lenses to keep them small, lightweight, perform very good optically, providing minimal focus breathing (thus far), but with the compromise of distortion being the concession.
What is a medium frustration is there is an "optical gap" in mindset and quality set by the rather good 50mm f/1.2L, 85mm f/1.2L, and 138mm f/1.8L RF, which are on par or exceed Zeiss Otus quality generally speaking. But on the wide side there is no such lens in the current RF lineup in terms of well corrected, low-to-no CA, and all of that.
I think that designing a lens to be wider than the planned use is a really good idea. It shows that Canon set a performance goal for the corner output of this lens, and used every means available to them, at the planned price point, to achieve this goal. Sure, today we can design exceptional 35mm lenses, the Leica SL Summicron for example, but even it has some digital correction. What matters is the output, if it's sharp, with low distortion and CA, then it's a win. Lighter weight is much appreciated. Mine is on order.
rscheffler wrote:
Gary, have you compared SOOC jpeg images with your 24-105Z against the same images, but raw versions, processed in Lightroom with corrections applied?
The 24-105Z has fairly strong distortion at 24mm and requires aggressive correction. With the few images I've processed from it so far, at 24mm the SOOC corrected image has a considerably narrower angle of view than the LRC distortion corrected version, which suggests to me that at least for this lens, Canon applies a fairly aggressive crop after distortion correction. At least at the wide end - I haven't yet compared other focal lengths. Perhaps this is also true for other lenses? ...Show more → garyvot wrote:
Actually, I haven't, but I find this interesting. My experience with the 24 f/1.8 STM is that Adobe exactly matches the Canon corrections, resulting in a narrower FOV compared to other 24mm lenses, whereas DXO leaves considerably more image area, which I prefer.
I had assumed that Adobe would simply follow Canon guidance on this, but maybe not. I will look at this, thanks.
It got me more curious about the 24-105 Z but don't want to derail this thread further. In the 24-105 Z discussion I posted three versions of an image from the 24-105 Z at 24mm. One SOOC with Canon's distortion corrections applied, one exported from LRC with Adobe's corrections applied, and one exported from LRC without distortion correction applied.
Making the lens wider, then cropping and fixing distortion in post robs acuity. No way around it.
The question is if it's 'worth it'; sensor resolution can mask the issue in terms of output, but on lower-resolution sensors, it may not be worth the compromise.
Depending on image content, one can opt to disable corrections in non-Canon software to retain fine detail acuity. But I agree, with the increased reliance on software corrections this will be a compromise in general use. For some a smaller lens is a bigger gain than a fractional loss of absolute image quality and this is likely where Canon placed priority.
johnctharp wrote:
Making the lens wider, then cropping and fixing distortion in post robs acuity. No way around it.
The question is if it's 'worth it'; sensor resolution can mask the issue in terms of output, but on lower-resolution sensors, it may not be worth the compromise.
That may or may not be true vs the alternative of designing a pure 35mm lens to this price point, size and weight. The wider lens may exhibit a superior mtf at the crop point vs the pure 35mm lens at the edge. We often see this when cropping. Resolution could be higher with the benefit of lower distortion and CA at the crop point of the wider lens. Correspondingly, this would be easier to correct in camera.
As I mentioned, it is possible to design such lenses, but they are much bigger & heavier. The excellent Sigma 35mm F1.2 springs to mind. The Leica is one stop slower and $6K new...
I see it as a compromise on compactness; in every other way, you're sacrificing potential resolution as any non-destructive corrections will rob acuity due to being magnified (literally).
As I said above, this concern is lessened with higher-resolution sensors; drop down to an R6 or R8, and now you've taken a serious bight out of the potential of the system.
Where I see it more positive, and in parallel to the higher-resolution R5, is in use with APS-C cameras. I'd actually be interested in seeing what that looks like.
numbertwo wrote:
A bit disappointing and explains why it’s cheaper than the rf 50/85 and doesn’t have the blue sphere coatings as the EF 35 II…
It will be interesting to see what other reviewers and new owners response to this new lens will be.
BTW, "blue sphere coatings"? I think you meant BR (Blue Spectrum Refractive Optics) that this new lens does not have that the EF 35mm f/1.4L II USM does have. The new lens does have ASC (Air Sphere Coating).
I experienced that the Canon RF 50mm f1.2 L looses some contrast when backlit, but that didn't hurt the image for me. However, judging by the images in the video, the Canon RF 35mm f1.4 L seems to be sub-par regardless of price.
I consider the reviewer to be impartial over the years and straight to the point in a German sense. He likes other RF L lenses, and loves the RF 85mm f1.2 L.
A bit disappointing and explains why it’s cheaper than the rf 50/85 and doesn’t have the blue sphere coatings as the EF 35 II…
Yes the review is so bad… I just sold my rf 135 mm and am wondering what lens to buy for my canon r6ii. I have a7rv, a9, zve1, with 14mm, 24mm, 35mm, Gm lens, plus sigma 35mm f1.2, 85mm f1.4.
My dream canon lens is 85mm ds but that is so heavy and expensive. I love the 35mm f1.2 Bokeh and am hoping the new canon might be better than the Sony 35mm Gm for bokeh but now I don’t think so.
Yeah, a leveled up sigma 35 1.2 was what I was hoping for as well. Canon has been releasing hybrid lenses where they also have a high end L lenses (like the rumored forthcoming 70-200 hybrid), so I still think we will eventually see a new beast RF 35 L.
Why did you sell the 135?
wind30 wrote:
Yes the review is so bad… I just sold my rf 135 mm and am wondering what lens to buy for my canon r6ii. I have a7rv, a9, zve1, with 14mm, 24mm, 35mm, Gm lens, plus sigma 35mm f1.2, 85mm f1.4.
My dream canon lens is 85mm ds but that is so heavy and expensive. I love the 35mm f1.2 Bokeh and am hoping the new canon might be better than the Sony 35mm Gm for bokeh but now I don’t think so.
A bit disappointing and explains why it’s cheaper than the rf 50/85 and doesn’t have the blue sphere coatings as the EF 35 II…
Thanks for sharing!
I found the review very detailed and the criticism fair.
It seems like canon heavily prioritized the lack of focus breathing (important for video, not so much for photo).
For me the two most annoying things if I were to buy the lens would be:
1.) The rattle when the camera is off. He mentioned that you can hear it clanking from inside your backpack. That could get quite irritating on a quiet hike.
2.) The backlit performance. The flares are VERY strong and it seems like they would be hard to clean up in post.
The digitally corrected distortion and vignetting personally don't bother me. I am OK with those if it means the lenses are going to be smaller.
But now I am kinda hoping Canon will still make a ultra premium 35mm (like 50mm and 85mm RF).
Taken on Canon R5, mixed indoor lighting. I am seeing a lot of optical vignetting (cat's eye bokeh) and a ton of barrel distortion and vignetting. This was not corrected adequately using the RF 35mm f/1.8 profile or with manual correction options in LR. Sharp as sharp can get, but definitely needs some profiling to be reasonable for stills. I am not so sure yet about this lens. We need a proper LR profile at the very least, to address barrel distortion.
On a positive note, the lens is very light and feels much more compact, easy to handle than prior EF models. I don't think the VCM rattle would bother me much as I'd rarely be swinging a camera around with enough force, velocity, or angular momentum for it to come into play for either stills or video.
To me, as a stills shooter who does some video, I think much will depend on the profiles avaiable in LR, the Bokeh characteristics in settings outside of a camera store, and the sharp/OOF transitions. The lens focuses fast enough for me, and does not feel like it hesitates at all.
ashwinrao1 wrote:
Taken on Canon R5, mixed indoor lighting. I am seeing a lot of optical vignetting (cat's eye bokeh) and a ton of barrel distortion and vignetting. This was not corrected adequately using the RF 35mm f/1.8 profile or with manual correction options in LR.
This may have to do with the fact that the RF 35/1.8 and RF 35/1.4L are different lenses with different profiles. LR13.4, that came out two days ago, has the correct profile included. You should try it out, I am getting much better results than what you're showing, my straight lines are actually straight, and vignetting isn't visible.
comotionfilms wrote:
Yeah, a leveled up sigma 35 1.2 was what I was hoping for as well. Canon has been releasing hybrid lenses where they also have a high end L lenses (like the rumored forthcoming 70-200 hybrid), so I still think we will eventually see a new beast RF 35 L.
Why did you sell the 135?
I brought the R6ii + 135mm, zve1, a7r5, a9, 14mm, 24mm, 35mm gm on my last vacation. I found the r6+135mm combo very heavy and I used it like 10% of the time and the photos did not turn out too nice. I think 135mm is just too hard to use on vacation and cannot justify its weight. I am thinking of bringing my 85mm dg dn instead on my vacations.
now I am thinking of selling the r6ii or get another lens... considering rf 70-200mm f2.8, 85mm ds (super expensive locally though)... (I had the sony 70-200mm gm2 and sold it as I did not like its bokeh and size)
There is another first review &lc=Ugy8eB49fPhs9k_edIB4AaABAg.A5HRusQ3gKsA5J0Plqb5YK
In the video the reviewer didn't cover potential flaring. I asked him in the comments and he did some testing afterwards with a flash directly into the lens and he didn't identify severe flaring.
