Canon Introduces Three New Lenses, Enhancing Still Photography and Video Production for Any Skill Level

[scyrene]

Except for those of us who own the 800 and just pre-ordered the 200-800.

Used 800 prices will crater.

They're already lower than most.

 

[roby17269]

Important thing to keep in mind is that optically correcting the corners of wide/ultrawide lenses (i.e. with the design of the glass in the lens) still stretches the corners and reduces resolution. From my testing, correcting the distortion digitally yields similar results in terms of resolution, but results in a smaller, lighter, and cheaper lens.

good point didn't think of it that way

 

[Bahrd]

In a place where I can test the new lens!

Canyon de Chelly - if I may suggest...

 

[H. Jones]

With all these announcements recently, it's kinda crazy to think there isn't a ton of stuff from the EF left to make native on the RF mount.

Right now the biggest gap for RF seems to be fast wide primes, so I feel like those will be due next year.

RF 35mm F/1.2, no doubt this is finally happening next year.

RF 200-500 F/4L, no doubt since there's been plenty of chatter about that.

RF 70-200mm F/2.8L Z, or something along those lines, could also be good to announce with the R1. It does also beg the question of how far will the Z lenses go? is there a RF 14-24mm F/2.8 Z coming? A RF 100-400 F/5.6 Z lens?

RF AF tilt shifts, probably, though not sure they would go with the R1, they feel like more of an EOS R5 Mark II sorta announcement.

MP-E replacement? I know it's super niche, but a super macro doesn't exist for the RF mount yet other than the 1.4x macro of the 100mm.

 

[Nemorino]

There seems to be no pattern to it and it is the same day that the RF 10-20 f/4 IS L was released.

IMO it is a statement they care about all customers, starting with Aps-c vloggers up to high professionals.

 

[scyrene]

I'll hazard a guess that one of the next lenses coming from Canon will be a RF 70-200mm F/2.8L Z lens, of some sort.

Maybe 70-240mm F/2.8L Z? Or 50-200mm F/2.8L Z? Looking at the size and shape of the RF 24-105mm F/2.8L Z, I almost wonder if they could copy the external body over to an internal zoom 70-200mm with the same servo attachment compatibility.

I've said it before: 70-200 f/2

 

[scyrene]

Thanks for hearing me out.

To be clear I don't necessarily mean that exact lens. I'd love to see an example of what you're talking about here, whichever lens it may be. It may not be possible but a full comparison would include:

1) center-exposed, software-corrected image with noise in the corners. Even that by itself would be very instructive.

2) corner-exposed, software-corrected image that shows blown-out highlights in the center. Again, even by itself would be very instructive, and show that simply exposing for the corner isn't an adequate solution.

3) image (1) or (2) but without software correction, as it'd be interesting to see if the corrected version is ever actually worse than uncorrected,

4) any alternative lens not depending heavily on software correction of vignetting (or distortion). Ideally they'd be same subject same time, but I understand that's practically impossible to have access to. (I do usually try to shoot some comparisons, though, when I'm upgrading to the latest generation, e.g. EF135 to RF135.)

It's a niche of a niche and I don't have any files to hand but when I used to do deep sky astro work, which involves pushing files to extremes, I usually had to discard the outer portion of the frame in part because of noise in the corners caused by vignetting. And that was after stacking etc. I wouldn't expect them to produce lenses for that purpose though, and my feeling is cropping is usually a good workaround - particularly starting with a very wide angle lens.

 

[roby17269]

Right now the biggest gap for RF seems to be fast wide primes, so I feel like those will be due next year.

Agreed

RF 35mm F/1.2, no doubt this is finally happening next year.

Would love it sooner rather than later... but at this point, I will need to see an official announcement before I get my hopes up again

RF 200-500 F/4L, no doubt since there's been plenty of chatter about that.

RF 70-200mm F/2.8L Z, or something along those lines, could also be good to announce with the R1. It does also beg the question of how far will the Z lenses go? is there a RF 14-24mm F/2.8 Z coming? A RF 100-400 F/5.6 Z lens?

Not on the market for any of these, but all interesting lenses

RF AF tilt shifts, probably, though not sure they would go with the R1, they feel like more of an EOS R5 Mark II sorta announcement.

MP-E replacement? I know it's super niche, but a super macro doesn't exist for the RF mount yet other than the 1.4x macro of the 100mm.

Would be very interested in new TS, especially if with AF. MP-E could be interesting if they do something even more extreme than the existing one.

All in all, while I am still reeling from the disappointment (no 35 1.2!), at least I am happy that Canon is showing that RF continues to be a vibrant mount

 

[koenkooi]

With all these announcements recently, it's kinda crazy to think there isn't a ton of stuff from the EF left to make native on the RF mount.

Right now the biggest gap for RF seems to be fast wide primes, so I feel like those will be due next year.

RF 35mm F/1.2, no doubt this is finally happening next year.

RF 200-500 F/4L, no doubt since there's been plenty of chatter about that.

RF 70-200mm F/2.8L Z, or something along those lines, could also be good to announce with the R1. It does also beg the question of how far will the Z lenses go? is there a RF 14-24mm F/2.8 Z coming? A RF 100-400 F/5.6 Z lens?

RF AF tilt shifts, probably, though not sure they would go with the R1, they feel like more of an EOS R5 Mark II sorta announcement.

MP-E replacement? I know it's super niche, but a super macro doesn't exist for the RF mount yet other than the 1.4x macro of the 100mm.

I’m missing an equivalent for the 180L macro as well. The MP-E is hard to improve on in it current form, but the 180L has lots of low hanging fruit.

 

[AJ]

AFAIK, there are 2 ways modern lenses / cameras deal with vignetting:

• traditionally the camera / software would brighten the corners to offset / mitigate vignetting. This could result in higher noise in the corners, and it would be more or less visible depending on exposure
• more recently, some lenses are made to be actually wider than declared fl, but with extremely dark corners. The camera / software crops / stretches the image eliminating the dark corners and increasing the fl to the declared fl of the lens. This will not affect noise but it will potentially reduce resolution in the corners due to stretching. Again, this may be very visible or very little. This approach is used with the RF 24-240 and 10-20 lenses, and possibly other wide zooms.

What the 24-240 and 16/2.8 and 14-35 do is this: the optics produce an image that is barrel-distorted. Barrel distortion doesn't really affect the focal length. On a vertical bisector the field-of-view is the same as for a perfect rectilinear lens, and going on a diagonal the field-of-view is wider.

With the barrel distortion in place, the far corners are completely black, and then there is a gradation from black to un-vignetted image. The distortion correction, then, pushes the completely black and the worst of the vignetting outside of the imaging area.

I too was worried about the corner stretch resulting in smear and blur, but it is not nearly as bad as you might expect. Like Neuro, I haven't noticed significant image degradation. The benefits of compact optics far outweigh the cons in this regard.

The remaining vignetting is quite prominent and needs correction. Even when well corrected, there can be significant noise. I've found the issue to be most noticeable with deep blue skies. I sometimes end up cloning to hide the corner noise.

To me, the lack of corner illumination is the biggest compromise when it comes to these types of lenses, rather than the distortion correction.

 

[neuroanatomist]

To me, the lack of corner illumination is the biggest compromise when it comes to these types of lenses, rather than the distortion correction.

This came up recently, so I’ll just quote myself:

Vignetting is not unique to digitally corrected lenses. The RF 14-35/4 has about 2.3 stops in the corners at 14mm f/4, that's much less than many EF lenses (e.g. the 16-35/4 has over 3 stops, the 16-35/2.8 III and 11-24/4 have >4 stops). While I agree that the exposure adjustment needed to correct that vignetting adds noise and reduces DR, that has always been the case.

 

[dbcooper]

800mm is certainly headline grabbing but this is just only half a stop faster than the 800 f11. Yes theirs is zoom but most will be using this at 800 most of the time.
Close focus is an advantage here, but then the fixed tripod leg will make it more cumbersome to pack. Also if note is this is slower at 600 than the Sony.

 

[mdcmdcmdc]

When did "fast lens" become "bright lens"?

"Fast" and "bright" have always been synonymous in describing lenses. Lenses with wider apertures transmit more light to the image plane (bright), thereby allowing shorter exposures (fast).

Did I misunderstand the context of this question? Or is it just that I've been doing this long enough to remember a time when few people outside of Japan knew what "bokeh" was?

 

[dbcooper]

Except for those of us who own the 800 and just pre-ordered the 200-800.

Used 800 prices will crater.

I have the 800 f11 and I don’t particularly like except for being light and easy to carry. The 200-800 is a lot heavier for just half a stop faster. When I first saw the new lens I was certain I was going to sell the f11 and get this but I already changed my mind. I also don’t like the non removable tripod leg which will take plenty of valuable space in my bag

 

[Quarkcharmed]

To be clear I don't necessarily mean that exact lens. I'd love to see an example of what you're talking about here, whichever lens it may be. It may not be possible but a full comparison would include:

1) center-exposed, software-corrected image with noise in the corners. Even that by itself would be very instructive.

Here's a recent milky way shot - EF 16-35 f4. Sorry I couldn't upload the full sized one (8192x5464) as the forum couldn't digest it.

So here's a 3000x2000 one.

Lightroom lens vignette correction has been applied, but not the distortion correction. You can see the higher noise in the corners, except the bottom right one where it was light polluted (still noisy in the very bottom right tip).

2) corner-exposed, software-corrected image that shows blown-out highlights in the center. Again, even by itself would be very instructive, and show that simply exposing for the corner isn't an adequate solution.

It's not an adequate solution because it's generally meaningless and not practical. In makes no sense in astro and it makes no sense when doing ETTR.

3) image (1) or (2) but without software correction, as it'd be interesting to see if the corrected version is ever actually worse than uncorrected,

The same image without vignette correction.

 

[neuroanatomist]

Here's a recent milky way shot - EF 16-35 f4.

Thanks for the example. The debate/argument here is about the complaints from some people that the RF lenses designed to require digital correction of the distortion are ‘horrible’ or ‘unacceptable’. That your example of this terrible, no good, very bad digital corner mangling comes from an EF lens effectively pours cold water on their (il)logic.

 

[Quarkcharmed]

Thanks for the example. The debate/argument here is about the complaints from some people that the RF lenses designed to require digital correction of the distortion are ‘horrible’ or ‘unacceptable’. That your example of this terrible, no good, very bad digital corner mangling comes from an EF lens effectively pours cold water on their (il)logic.

The images are still usable after corrections, but note it's 16mm, and the lens in question is 24mm at its wide end. So we may expect less vignetting, probably. But the fear is it can be worse - we just need to wait for the proper tests/measurements on that lens.

If we're concerned about the RF lenses being worse that the RF lenses, it's a bit different question. In this case proper comparisons should be made between, say, EF 16-35 f2.8 and RF 15-35 f2.8.

The RF 15-35 at 15mm f2.8 has 4.6 stops of vignetting:

Canon RF 15-35mm f/2.8 USM L IS - Review / Test Report - Analysis

Canon RF 15-35mm f/2.8 USM L IS - Review / Test Report

www.opticallimits.com

The same 4.6 stops in EF 16-35 at 16mm f2.8:

Canon EF 16-35mm f/2.8 USM L III - Review / Test - Analysis

Canon EF 16-35mm f/2.8 USM L III - Review / Test

www.opticallimits.com

 

[mxwphoto]

Though I am not the target audience, I do feel the 24-105 2.8 is too expensive considering that to use the power zoom it looks like you need to attach an extra $1000 accessory. That brings the lens' full functionality price to $4k (or $4.3k for remote shooting).

Considering Canon was selling the PZ E1 for $150 in 2016 for the 18-135 lens, charging $1000 for the same thing now is a bit ridiculous, especially when competing brands add a powerzoom rocker on the lens itself.

 

[Australisblue]

Is that massive tripod mount on the 200-800 really not removable? Definitely not a showstopper for me but I’d much prefer to have the option to take it off when not needed. Makes it much more bulky in a bag.

 

[H. Jones]

Though I am not the target audience, I do feel the 24-105 2.8 is too expensive considering that to use the power zoom it looks like you need to attach an extra $1000 accessory. That brings the lens' full functionality price to $4k (or $4.3k for remote shooting).

Considering Canon was selling the PZ E1 for $150 in 2016 for the 18-135 lens, charging $1000 for the same thing now is a bit ridiculous, especially when competing brands add a powerzoom rocker on the lens itself.

This lens is most directly equitable to the CN-E 18-80mm T4.4 servo lens. That lens sells with a built in rocker for $4600, and then needs a servo grip for $450 if you want better handling. so this lens being a full frame compatable 24-105mm F/2.8 servo lens is a steal at $4000 versus the CN-E 18-80, which is super 35 only for $5000.

Add in a bunch of tech and features, I'd say the price point makes a lot of sense. Cinema lenses are far more expensive than regular lenses, and this Z seems to be setting up a middleground between the high end cinema stuff and regular stills line.

My last job before I got back into working for newspapers was videography for a college, and we would have scooped this lens up immediately with the servo unit, we actually were budgeting for the CN-E 18-80 for our C70s when I left, I imagine this will save some cash for them when they get this instead.