Patent: Canon RF 135mm f/1.8L and other fast primes

Jul 21, 2010
31,096
12,857
Sorry for jewelry for LoCA is BS (no offense). One can see things like that with the naked eye since a gem acts itself like a lens (uncorrected). For a proper LoCA test one needs to not have additional “optical elements” in the frame.

Most accounts of the 50 f1.2 points to well controlled LoCA.

Anyway... the 55 sonar is a spectacular lens and came out at €1k and it showed some CA. I am sure that you can find examples of all lenses exhibiting LoCA with jewelry.
CA (both types) is most evident at high contrast transitions. Specular highlights (e.g., jewelry, chrome, sun on water, etc.) simply provides a near-maximum contrast. CA on the RF50 is certainly less than on the EF50/1.2, but really any lens with that wide a max aperture will suffer from noticeable LoCA.

To your earlier point, LoCA is notoriously bad on the 85/1.8.
 
  • Like
Reactions: 1 user
Upvote 0

navastronia

R6 x2 (work) + 5D Classic (fun)
Aug 31, 2018
853
1,073
CA (both types) is most evident at high contrast transitions. Specular highlights (e.g., jewelry, chrome, sun on water, etc.) simply provides a near-maximum contrast. CA on the RF50 is certainly less than on the EF50/1.2, but really any lens with that wide a max aperture will suffer from noticeable LoCA.

To your earlier point, LoCA is notoriously bad on the 85/1.8.

FWIW, the LoCA was infinitely worse on the Tamron 35/1.8 than the Canon 85/1.8, and harder to correct. Indeed, perhaps it's because it's a wider angle lens.
 
Upvote 0
CA (both types) is most evident at high contrast transitions. Specular highlights (e.g., jewelry, chrome, sun on water, etc.) simply provides a near-maximum contrast. CA on the RF50 is certainly less than on the EF50/1.2, but really any lens with that wide a max aperture will suffer from noticeable LoCA.

To your earlier point, LoCA is notoriously bad on the 85/1.8.

All I was saying is that for jewelry... care to provide an example of a lens that will produce no CA at all on a gem? Gems bend light that passes through it. With the nature of different wavelengths of wight light you can/will get color separation. You know, like light going through the edge of a glass door projecting the colors of the rainbow. So my comment was specifically to point out that you can't blame a lens when the physical event is determined by the glass/gem/jewel rather than an issue with the lens. As such jewelry images do not exactly inform me/us how well CA is controlled by the lens, as it might not be "high contrast" causing the effect.

I much rather take an image of branches against a bright sky, or power lines, and look at those for CA.
 
  • Like
Reactions: 1 users
Upvote 0
Jul 21, 2010
31,096
12,857
All I was saying is that for jewelry... care to provide an example of a lens that will produce no CA at all on a gem? Gems bend light that passes through it. With the nature of different wavelengths of wight light you can/will get color separation. You know, like light going through the edge of a glass door projecting the colors of the rainbow. So my comment was specifically to point out that you can't blame a lens when the physical event is determined by the glass/gem/jewel rather than an issue with the lens. As such jewelry images do not exactly inform me/us how well CA is controlled by the lens, as it might not be "high contrast" causing the effect.

I much rather take an image of branches against a bright sky, or power lines, and look at those for CA.
LoCA results from refraction of light by the lens elements and is characteristically seen as purple fringing in front of the focal plane and green fringing behind the focal plane (although in extreme cases both colors are seen both places, because refraction is occurring at both surfaces of the lens elements).

Are you saying that light refracted at the source, e.g. by a gem, would show up in the same way – purple foreground, green background? Reproducibly? LoCA is mitigated by stopping down the aperture. How would that resolve refraction occurring in a subject a significant distance from the lens? If you look at the jewelry image, there’s as much CA from the metal portions as from the inset gems. Does solid metal refract light? Perhaps you’re relying on a different physics than the rest of the universe.

I agree that real world examples, are informative, but when testing the performance of a lens (e.g.,for a review) it is often beneficial to test the extreme cases. White/black transitions at the edge of the frame for lateral CA, and specular highlights for longitudinal CA.
 
  • Like
Reactions: 1 user
Upvote 0
May 11, 2017
1,365
635
CA (both types) is most evident at high contrast transitions. Specular highlights (e.g., jewelry, chrome, sun on water, etc.) simply provides a near-maximum contrast. CA on the RF50 is certainly less than on the EF50/1.2, but really any lens with that wide a max aperture will suffer from noticeable LoCA.

To your earlier point, LoCA is notoriously bad on the 85/1.8.
Yes, the LoCA of the 85/1.8 is well known. Part of the reason may be that the 85 goes back to an era when a lot of photography was in black and white, where LoCA would be less obvious, so Canon would have worried about it less. LoCA may be hard to correct, but it can be avoided by stopping down the lens. As the user of an 85/1.8, I have to remember not to use it shoot pictures of motorcycle handlebars in bright sunlight with the lens wide open (unless, of course, I want to show the lens can produce impressive LoCA).
 
Upvote 0
There isn't that much difference between 85mm and 100mm?

Canon makes two EF 85mm lenses and at least two EF 100mm lenses, so apparently there's profit in making lenses of both focal lengths, in spite of not having much difference.

Situation is similar with Nikon, who has at least three 85mm lenses (1.8G, 1.4G, f/3.5G macro), and two 105mm lenses (f/1.4E, and f/2.8G macro).
 
Upvote 0
LoCA results from refraction of light by the lens elements and is characteristically seen as purple fringing in front of the focal plane and green fringing behind the focal plane (although in extreme cases both colors are seen both places, because refraction is occurring at both surfaces of the lens elements).

Are you saying that light refracted at the source, e.g. by a gem, would show up in the same way – purple foreground, green background? Reproducibly? LoCA is mitigated by stopping down the aperture. How would that resolve refraction occurring in a subject a significant distance from the lens? If you look at the jewelry image, there’s as much CA from the metal portions as from the inset gems. Does solid metal refract light? Perhaps you’re relying on a different physics than the rest of the universe.

I agree that real world examples, are informative, but when testing the performance of a lens (e.g.,for a review) it is often beneficial to test the extreme cases. White/black transitions at the edge of the frame for lateral CA, and specular highlights for longitudinal CA.

No, I am not relying on different physics. But as I was implying and as you pointed out refraction when it comes to optics like elements can contribute to what you see. I know what the definition of CA is. What is the exact physics behind it? Is it not different wave lens reaching different portions of the lens, especially when the light is bend aggressively and not compensated accordingly? So I when I say that generally I wouldn’t put optical elements in the frame to judge CA such. As gems or glass or ice etc it is to eliminate a potential contribution to the result. After all if one holds a gem up to the life and twist it this way and that one can general manipulate it to see a verity of colors. That includes purple and green those it not? So gems do bend light and can be a cause of a higher than “usual” incident light into the lens, which one can’t attribute.

Oh well, as an applied physicist, I suppose I have no idea about what I am talking about :p.

Of course if this was shown for those interested in taking pictures of jewelery specifically I get it can be useful. But as a blunt example of CA control... eh, grain of salt.
 
  • Like
Reactions: 1 users
Upvote 0

FramerMCB

Canon 40D & 7D
CR Pro
Sep 9, 2014
481
147
56
I seriously hope they will make one too. But after seeing the price of the 85, I’m not too eager anymore.
Keep in mind the 85 is an "f/1.2". Considering that this one is supposed to be f/1.8 it should be less than $2K at introduction - I'm thinking $1,599 - $1,699 range... We'll see. But I hear ya on the $$$ front. Pretty heady prices recently - but wow, the performance/metrics...
 
Upvote 0

FramerMCB

Canon 40D & 7D
CR Pro
Sep 9, 2014
481
147
56
Reasonable? Margin? I would hope Canon, and anyone in the world would have margin in mind at some point in their lives. That’s business.

Reasonable as compared to what? Sony & Nikon? Leica? Or just compared to your budget?

As far as I’ve seen, Canon’s lens pricing relative to the competition and factoring in quality and uniqueness is actually VERY reasonable. Browse through a BH catalog and compare list pricing and tell me Canon isn’t “reasonable”.
True - for example, just compare the Canon 70-200mm f/2.8L IS Mk III with the Sony GSS Master and Nikon's VR Mk II versions. A bargain. Now if you're comparing Canon lens pricing to 3rd party then that is a different story - but that's always been the case. That's why there's even a market for 3rd-party manufacturer's to "exploit"...
 
Upvote 0
Jul 21, 2010
31,096
12,857
No, I am not relying on different physics. But as I was implying and as you pointed out refraction when it comes to optics like elements can contribute to what you see. I know what the definition of CA is. What is the exact physics behind it? Is it not different wave lens reaching different portions of the lens, especially when the light is bend aggressively and not compensated accordingly? So I when I say that generally I wouldn’t put optical elements in the frame to judge CA such. As gems or glass or ice etc it is to eliminate a potential contribution to the result. After all if one holds a gem up to the life and twist it this way and that one can general manipulate it to see a verity of colors. That includes purple and green those it not? So gems do bend light and can be a cause of a higher than “usual” incident light into the lens, which one can’t attribute.
Yes, the physics is the same, but you're relying on physics at a vastly different scale. Refraction from lens elements results in separation of wavelengths on a µm-scale, refraction from an object in the field of view is not the same. From the standpoint of a subject LoCA results from reflection, not refraction. Consider: why is it that LoCA – no matter the source – shows up as purple and green? To (re)quote you (with typos corrected): "After all if one holds a gem up to the life light and twists it this way and that one can [in] general manipulate it to see a verity variety of colors." That's obviously true, so why does the LoCA observed in jewelry pictures* not show a variety of colors, but rather only two? Because the LoCA is not resulting from refraction through the subject, but only through the lens elements.

*There's a caveat there, that the gems need to be clear. If the gem is colored, the light from it will not be white and the way in which that light is refracted by the lens elements will be affected. But that's not the case here, as the jewelry under discussion contains only clear gems.


Of course if this was shown for those interested in taking pictures of jewelery specifically I get it can be useful. But as a blunt example of CA control... eh, grain of salt.
So if one is interested in sharpness, ISO 12233 charts are only useful for those interested in taking pictures of flat, printed subjects?

Regardless, look at the jewelry LoCA image, and notice that there are purple/green-fringed gems and that the setting is also purple/green-fringed, and that the intensity of the fringing is equivalent between the two. So let me ask you again, as an applied physicist – does solid metal refract light? If it does, then it's possible you're correct. But of course, we both know it doesn't.

As I stated previously, the 'extreme case' of LoCA is a specular highlight, and it really doesn't matter if that highlight is from a gem, polished metal, or another highly reflective surface.

Here's an example from a review of the RF 50/1.2. There is LoCA on the ring, no gems required.

184689

Having said that, I do agree that the LoCA on the RF 50/1.2 is very well controlled.
 
Upvote 0