There will be more VCM hybrid prime lenses, but not anytime soon

This is also my perception, but I'd wish for a real optical expert to tell us what the facts are. A "perception" is no fact.
The opinions expressed by most forum members rely on hearsay or subjectivity. This, of course, includes me. I do need a proof I'm either right or wrong.
Presently, given the choice between a digitally and an optically corrected lens, I'd go for the optically corrected one.
This makes sense to me as well: with optically-corrected lenses you have more information to play with as opposed to digitally corrected lenses that leave some pixels out.

Before @neuroanatomist chastises me as well :eek: , I am aware that what makes sense to me may not be reality. But, to play devil advocate with the examples he has offered, do we know for sure that the digital corrections do not include additional sharpening? I mean, could we get to at least parity if the optically-corrected images were sharpened?

Regardless, I do not know and I am not going to try. I do have and use the RF 10-20 4L and 35 1.4L VCM. They do work and do what says on the tin, but somehow they do not impress me like the 85 1.2L and 50 1.2L and 135 1.8L do. It is entirely possible that this stems from the fact that I am mainly a tele shooter who doesn't appreciate wide much (just printed large a big pano taken from the Presena glacier with the 100-500 and I love it, all the work it took to take the images and stitch them notwithstanding). But I will keep hanging to my romantic notion that glass is better... so Canon bring out my RF 35 1.2L already! :ROFLMAO:
 
Upvote 0
This makes sense to me as well: with optically-corrected lenses you have more information to play with as opposed to digitally corrected lenses that leave some pixels out.

Before @neuroanatomist chastises me as well :eek: , I am aware that what makes sense to me may not be reality. But, to play devil advocate with the examples he has offered, do we know for sure that the digital corrections do not include additional sharpening? I mean, could we get to at least parity if the optically-corrected images were sharpened?

Regardless, I do not know and I am not going to try. I do have and use the RF 10-20 4L and 35 1.4L VCM. They do work and do what says on the tin, but somehow they do not impress me like the 85 1.2L and 50 1.2L and 135 1.8L do. It is entirely possible that this stems from the fact that I am mainly a tele shooter who doesn't appreciate wide much (just printed large a big pano taken from the Presena glacier with the 100-500 and I love it, all the work it took to take the images and stitch them notwithstanding). But I will keep hanging to my romantic notion that glass is better... so Canon bring out my RF 35 1.2L already! :ROFLMAO:
I'm absolutely certain you'll love your optically corrected 35mm f/1,2! :)
While I keep using my RF 35 f/1,4 II and the fantastic Summilux M 35 f/1,4 Asph.
PS: I'm not always evil...
 
Last edited:
Upvote 0
Canon will do what makes economic sense to them, so milking the hybrid market with VCM first (many potential buyers) and coming afterwards with a better (and faster?) lens at a higher price (most likely), weight and dimension (and fewer buyers) might make sense to them. With the RF 85/1.2 L (BR optics) they have shown that they can produce excellent glass for photographers.
I think the lenses can co-exist.. like the RF 50mm 1.2L & 1.4L VCM version.... and Canon making super lenses are not in question... it is a matter of how many lungs or liver or which organ you have to sell.... :LOL:
 
Upvote 0
Nothing is perfect. Everything is a compromise.
When they try to target 2 segments, they have to compromise more than focusing on only 1 segment....

In some way, I find the Z & VCM L designs are quite smart..... They have the same dimension, same parts, save money, and also, do not need to readjust when mounting on gimbal....

Canon have a few overlapping L lenses.... RF 70-200mm, RF 50mm... and I think they will continue to overlap and milk the photographers dry......
 
  • Like
Reactions: 1 user
Upvote 0
When they try to target 2 segments, they have to compromise more than focusing on only 1 segment....

In some way, I find the Z & VCM L designs are quite smart..... They have the same dimension, same parts, save money, and also, do not need to readjust when mounting on gimbal....

Canon have a few overlapping L lenses.... RF 70-200mm, RF 50mm... and I think they will continue to overlap and milk the photographers dry......
"Segments" are just a way of conceptualising the market, they aren't eternal truths of the universe. There are now 3 50mm RF primes. They're all as good as or better than their EF predecessors, no? So whatever compromises have been made, the outcome is good. What's missing, in your view? I don't go along with this "stills vs video" dichotomy mentality incidentally.
 
Upvote 0
  • Like
Reactions: 1 user
Upvote 0
This makes sense to me as well: with optically-corrected lenses you have more information to play with as opposed to digitally corrected lenses that leave some pixels out.
How good is the information? Keep in mind that the corners of 'optically corrected' lenses deliver much lower resolution than the central portions. As I keep saying, there's no free lunch. So which is better, correcting the distortion with optical elements that lead to reduced resolution and are subject to manufacturing tolerances, or using a simpler optical design that leaves a lot of distortion in then correcting that distortion based on a mathematical model of the lens?

Before @neuroanatomist chastises me as well :eek: , I am aware that what makes sense to me may not be reality. But, to play devil advocate with the examples he has offered, do we know for sure that the digital corrections do not include additional sharpening? I mean, could we get to at least parity if the optically-corrected images were sharpened?
Interesting point. I presume that lens profiles to correct distortion do not add local sharpening in the corners. I would think if that were the case, the aliasing artifacts that result from oversharpening would show up more strongly in the corners, and they don't...so I don't think it's happening.

Regardless, I do not know and I am not going to try. I do have and use the RF 10-20 4L and 35 1.4L VCM. They do work and do what says on the tin, but somehow they do not impress me like the 85 1.2L and 50 1.2L and 135 1.8L do. It is entirely possible that this stems from the fact that I am mainly a tele shooter who doesn't appreciate wide much (just printed large a big pano taken from the Presena glacier with the 100-500 and I love it, all the work it took to take the images and stitch them notwithstanding). But I will keep hanging to my romantic notion that glass is better... so Canon bring out my RF 35 1.2L already! :ROFLMAO:
The EF 35/1.4L II has 0.4% barrel distortion. The RF 35/1.4L has closer to 4%. After correction, the corners of the latter are sharper. Notions are nice, personally I prefer reality. Telephoto lens designs are quite different from wide angle designs, obviously.
 
  • Like
Reactions: 1 user
Upvote 0
How does the AF motor choice (VCM) impact the optical performance?
Although the 24/1.4 would have a bigger entrance pupil vs 20/1.4, astro depends on what is the field of view that you are wanting to shoot.
Deep sky then okay.
If entrance pupil is all that matters, 600/4 FTW. But as you say, there are other considerations... In the case of @Malm's comment that the 24/1.4 lets in more light, as you suggest that's true but is balanced by the fact that for a single exposure you can leave the shutter open longer with a 20/1.4 than with a 24/1.4.
 
  • Like
Reactions: 1 user
Upvote 0
How good is the information? Keep in mind that the corners of 'optically corrected' lenses deliver much lower resolution than the central portions. As I keep saying, there's no free lunch. So which is better, correcting the distortion with optical elements that lead to reduced resolution and are subject to manufacturing tolerances, or using a simpler optical design that leaves a lot of distortion in then correcting that distortion based on a mathematical model of the lens?
Don't know, but I'd have thought that "softer" pixels still carry more information than black ones.
Interesting point. I presume that lens profiles to correct distortion do not add local sharpening in the corners. I would think if that were the case, the aliasing artifacts that result from oversharpening would show up more strongly in the corners, and they don't...so I don't think it's happening.
Sharpening tweaked especially for a given lens may not show many oversharpening artifacts. But yes, I cannot prove this one way or another
The EF 35/1.4L II has 0.4% barrel distortion. The RF 35/1.4L has closer to 4%. After correction, the corners of the latter are sharper. Notions are nice, personally I prefer reality. Telephoto lens designs are quite different from wide angle designs, obviously.
Reality and facts are important. I try to teach that to my daughter :cool: But in the end this is a passion of mine and the livelihood of mine and of my family does not depend on it. So feelings are important as well (cop out warning)

I appreciate the civil reply in any case - seriously
 
  • Like
Reactions: 1 users
Upvote 0
Don't know, but I'd have thought that "softer" pixels still carry more information than black ones.
Fair point, but distortion correction is about computationally moving the pixels back where they belong, i.e. where the light came from in the real world before being bent by the lens.

I've been through this transition before with scientific imaging, when makers of early deconvolution microscopes (which make thin optical sections from thicker ones computationally) were trying to compete with confocal microscopes (which make thin optical sections with optics). People using confocal systems argued that the optical way had to be superior, simply because it was optical and not digital. It took a few decades, but now here's a statement from a microscopy primer on the National Cancer Institute website (part of the US NIH): In many cases, particularly with dimmer or light-sensitive specimens, deconvolved images are superior to confocal images. This is because in confocal microscopy the out-of-focus light is rejected, whereas in deconvolution microscopy the out-of-focus light is restored to its source of origin in the image.

At the end of the day, the results should speak for themselves. I've posted a bunch of actual examples of how the digitally corrected corners are at least as good and often better than optically corrected corners. I have seen lots of people claim that digital correction is worse, but I have yet to see anyone post an example that demonstrates that. Those people just keep repeating that optical correction must be better simply because it's optical and not digital. I didn't buy that argument in favor of confocal microscopy, and I don't buy it in favor of optical distortion correction. SHOW ME THE DATA.
 
  • Like
Reactions: 1 user
Upvote 0
"Segments" are just a way of conceptualising the market, they aren't eternal truths of the universe. There are now 3 50mm RF primes. They're all as good as or better than their EF predecessors, no? So whatever compromises have been made, the outcome is good. What's missing, in your view? I don't go along with this "stills vs video" dichotomy mentality incidentally.
I think Canon has enough data to know what the trends.... and the way consumers, prosumers, pros will keep wanting the BEST gear (mostly).... while another segment wants to pay a Ferrari at a VW/Ford price..... The changing trend where photographers also will take some videos... buyer prefers smaller lense... and after data analytics of the data, arrive at VCM L lenses...

They filled the gap with primes that are slightly cheaper... relatively... and the catalogue now looks more complete..... The advantage of VCM lenses are it is small, compact, and maintain RF lenses superiority in sharpness than EF and F-stop at F1.4....

Potentially, it leaves Canon another potential market segment, with F-stop at 1.2, optically corrected without digital magic, lesser CA, better bokeh, sharper across the frame, and who knows... they may add IS.... and bigger.... but they could touch this much later and fill out other range for now in their catalog....
 
  • Like
Reactions: 1 users
Upvote 0
Fair point, but distortion correction is about computationally moving the pixels back where they belong, i.e. where the light came from in the real world before being bent by the lens.

I've been through this transition before with scientific imaging, when makers of early deconvolution microscopes (which make thin optical sections from thicker ones computationally) were trying to compete with confocal microscopes (which make thin optical sections with optics). People using confocal systems argued that the optical way had to be superior, simply because it was optical and not digital. It took a few decades, but now here's a statement from a microscopy primer on the National Cancer Institute website (part of the US NIH): In many cases, particularly with dimmer or light-sensitive specimens, deconvolved images are superior to confocal images. This is because in confocal microscopy the out-of-focus light is rejected, whereas in deconvolution microscopy the out-of-focus light is restored to its source of origin in the image.

At the end of the day, the results should speak for themselves. I've posted a bunch of actual examples of how the digitally corrected corners are at least as good and often better than optically corrected corners. I have seen lots of people claim that digital correction is worse, but I have yet to see anyone post an example that demonstrates that. Those people just keep repeating that optical correction must be better simply because it's optical and not digital. I didn't buy that argument in favor of confocal microscopy, and I don't buy it in favor of optical distortion correction. SHOW ME THE DATA.
Your argument is rational and sound.
Regardless, should Canon finally announce a 35mm 1.2L which needs less digital corrections than the current 35 1.4L, I will still order the 1.2 and dump the 1.4, increased size and weight notwithstanding
 
  • Love
  • Like
Reactions: 1 users
Upvote 0
To be honest: I'm not impressed by the VCM line yet and would like to have more dedicated lenses for normal non-video photographers. What Canon in principal said is: VCM are very good lenses for video shooters which can also be used by normal photographers, as they sacrifice the ultimate quality by limiting size and weight (rig friendly). Video shooters don't need the ultimate quality (VCM lenses have no BR optics any more) but normal photographers do! Look at the RF 35/1.4 VCM which is only looking better on the paper than the super EF 35/1.4 II with BR optics. These VCM lenses seems to need a lot of 'pixel adjustment' in the RAW converter (vignette, distortion, chroma) to make them look good and that's not good for the quality. The RAW image of the RF24/1.4 VCM has black corners, so only a part of the sensor is used and pixels are shifted around. That degrades the quality!
The VCM lenses might be selling well at the moment as there are only very few high-quality RF lenses available, but that will chance if there will be more lenses for photographers become available. At the moment these lenses might be a money machine for Canon, but they are not looking very attractive to me. Just my 2 cents.
100%
 
  • Like
Reactions: 1 users
Upvote 0
the 24/1.4 lets in more light, as you suggest that's true but is balanced by the fact that for a single exposure you can leave the shutter open longer with a 20/1.4 than with a 24/1.4.
If you are referring to star trailing that that it correct (higher density sensors assuming 1:1 pixel peeping impacts the time) but the use of star tracking removes that limitation within reason.

A reasonable priced tracker should give you up to 2 minute exposure without trailing if reasonably aligned with the relevant polar axis of your hemisphere.

Taking one shot for foreground and a tracked one for the sky can be blended in starry landscape stacker (for Mac) for instance
 
  • Like
Reactions: 3 users
Upvote 0
Like how I would have bought a Ferrari, but my coffee mug is mauve and they don’t sell a Ferrari in that color.
If you want red or yellow, at least back in 2008, you had to choose one of the official Ferrari reds or yellows. For any other color they'll do whatever you want, though. I got mine in BMW color 894 instead of the similar Grigio Silverstone. (I know this wasn't your main point, but it's an excuse to dig up this old photo.)1730549511676.png
 
  • Like
Reactions: 1 user
Upvote 0