Patent: Canon is actively developing fast mirror lenses

[entoman]

With software being able to manipulate OOF areas better and better it is not unthinkable that a software solution can "do away" with the donuts.

For these wide-aperture cats to be worthwhile for "normal" photography, they'll need to have a means of increasing depth of field, and that means having a physical aperture stop-down. The donut effect would vary considerably according to aperture, focal length and subject distance. Consequently the AI correction software would need to recognise different diameters of the "donut hole", and account for different donut diameters and different degrees of donut overlap.

Perhaps suitable post-processing software might appear in 10 years time, but I don't think it's anywhere on the horizon yet. Maybe something that NASA or the military might have, but not the general photographic community.

My guess is that if Canon is "actively developing fast mirror lenses", they won't be general sports/wildlife lenses, but are more likely to be for astro or satellite use, where depth of field is less critical due to the much greater subject distances. I'd also guess that the donut effect will be less of a problem, although the astrophotography guys here would be in a better position than I am to answer that one.

 

[TAF]

Tripod is not an option for a contrail or even for a long distance aircraft photography, due to the very narrow field of view, combined with the realtive position of the target and the relatively fast moving target. Just imagine to make a photo with 2000mm about a continously moving airplane, what you have to find by your own naked eye first. A video pan head has its own initial angle limitation, which is really slows down the reaction time. Previously, I made a photo about the ISS with the single f/10 1000mm mirror lens, handheld, because of the position of the object (right above my head, on the Zenith). So these circumstances are not really god for a tripod-related photography. I also tried a monopod, but was not so helpful at all. The slowest shutter speed ever was 1/640, but the finest sweetspot is around 1/1000 - 1/1250, depending on the actual relative angle and position of the path of the aircraft. And the shutter speed is crucial, since the reflex mirror has a lack of sharpness, combined with a really narrow depth of field. And the fact, that you could rotating over the infinite distance mark the tube, you could easly miss the right focal point. By a native 1000mm mirror focal length (the extenders are really does not matter) the depth of field is extremly narrow, so between an 1000 or an 1500 meter, I have to set the focus, with the possibility of the overrotating.

I'm curious. How do you deal with atmospheric turbulence, haze, and defraction?

I do OK in winter, but the rest of the year the air is too 'cruddy' to get a decent image.

I am presently experimenting with the RF 600mm/f11, as a replacement for the 500mm/f6.3 mirror tele (a cheap Opteka) I was using. The bokeh is much better, but the air still isn't very clear.

 

[entoman]

I'm curious. How do you deal with atmospheric turbulence, haze, and defraction?

I do OK in winter, but the rest of the year the air is too 'cruddy' to get a decent image.

I am presently experimenting with the RF 600mm/f11, as a replacement for the 500mm/f6.3 mirror tele (a cheap Opteka) I was using. The bokeh is much better, but the air still isn't very clear.

With long focal lengths, atmospheric haze and heat haze can cause a lot of image degradation, regardless of the subject matter. The best time to shoot is very early in the morning, when temperatures are still low. After rain, the air is usually at its clearest.

 

[AJ]

I'm really surprised that Canon is looking at mirror lenses.
The main advantage of mirror lenses is the short physical length due to the collapsed optical path.
However, Canon has already solved this a different way with the 600/11 and 800/11 - a way that does not result in funky bokeh.
I wonder if the same design principles of the 600/11 and 800/11 could be applied to a shorter focal length with an aperture larger than f/11.

 

[scottburgess]

Dealing with bokeh and aperture concerns is actually pretty simple: build the lens in a Z configuration [it increases the size of the housing and looks non-standard, but who cares?]. While these patents illustrate the common ring configuration, arranging in a double barrel fashion, larger on top, allows for no ring bokeh and an aperture in the lower barrel. One might even engineer in a Canon 52mm drop-in (er, push-up?) filter. Such a lens could still be fat, but also short and light.

The important question to me is whether adequate IQ can be delivered. I would not be shocked if Canon has engineered some improvements to mirror lens technology. But as the target market for the displayed design is more likely middle-class amateurs, IQ could be similar to the mediocre 800/11 with ring bokeh added, but with a price around $1,000. It's not an unreasonable tradeoff for a casual hobbyist. I would be more interested in looking at a Z-design, short, durable, pro-quality 1000/5.6 based on the equipment I already have.

I'm guessing Canon is pulling out all their design ideas to create hardware for younger photographers in an effort to build up their amateur market. Some percentage of those will eventually become more serious or get GAS, or both.

 

[Fischer]

For these wide-aperture cats to be worthwhile for "normal" photography, they'll need to have a means of increasing depth of field, and that means having a physical aperture stop-down.

People used these lenses for years with a fixed aperture and DOF - do not see any reason at all it would make them irrelevant as long as they can offer good IQ for a fraction of a normal lens price. Lots of people only use their fast prime lenses wide open "99%" of the time, myself being one. Otherwise I might as well shot a zoom lens.

 

[entoman]

People used these lenses for years with a fixed aperture and DOF - do not see any reason at all it would make them irrelevant as long as they can offer good IQ for a fraction of a normal lens price. Lots of people only use their fast prime lenses wide open "99%" of the time, myself being one. Otherwise I might as well shot a zoom lens.

Sure, there are a lot of people including myself who commonly use long teles at max aperture, but the very narrow d.o.f. of a 300mm fixed F2.4 limits the scenarios in which it could be used (and the donut effect limits it even further).

If such a lens was cheap and very light it might appeal, but I think the very wide max aperture strongly hints that these will be very expensive pro-quality lenses aimed at astrophotography or satellite use, or possibly for night surveillance cameras.

But Canon has shown that they like to experiment with new designs (e.g. 600mm F11 and dual fish-eye), so anything is possible.

 

[usern4cr]

Dealing with bokeh and aperture concerns is actually pretty simple: build the lens in a Z configuration [it increases the size of the housing and looks non-standard, but who cares?]. While these patents illustrate the common ring configuration, arranging in a double barrel fashion, larger on top, allows for no ring bokeh and an aperture in the lower barrel. One might even engineer in a Canon 52mm drop-in (er, push-up?) filter. Such a lens could still be fat, but also short and light.

The important question to me is whether adequate IQ can be delivered. I would not be shocked if Canon has engineered some improvements to mirror lens technology. But as the target market for the displayed design is more likely middle-class amateurs, IQ could be similar to the mediocre 800/11 with ring bokeh added, but with a price around $1,000. It's not an unreasonable tradeoff for a casual hobbyist. I would be more interested in looking at a Z-design, short, durable, pro-quality 1000/5.6 based on the equipment I already have.

I'm guessing Canon is pulling out all their design ideas to create hardware for younger photographers in an effort to build up their amateur market. Some percentage of those will eventually become more serious or get GAS, or both.

I'm curious what you mean by "Z-design, short, durable, pro-quality 1000/5.6 based on the equipment I already have"?
If you're thinking of a standard folded mirror & lens design with only a circular off-axis opening used, then I'm sure everyone else has considered this and found that it is a bigger snowcone-ish shaped width with inferior optical quality. As you go further off-axis, the image degrades drastically and instead of a modest OOF donut, I'm guessing that you would get a much worse OOF snowcone shape, similar to what coma looks like. In addition, the in-focus image toward the outer edge furthest from the optical axis would probably have severe coma do to the 2 to 3x further off-axis distance, so the in-focus image itself would be poor.

 

[MihalyLaczko]

I'm curious. How do you deal with atmospheric turbulence, haze, and defraction?

I do OK in winter, but the rest of the year the air is too 'cruddy' to get a decent image.

I am presently experimenting with the RF 600mm/f11, as a replacement for the 500mm/f6.3 mirror tele (a cheap Opteka) I was using. The bokeh is much better, but the air still isn't very clear.

It is depending on the angle of the shooting. Here is a mid-summer contrail photo (almost 180 degrees zenith position target), taken with the MTO 1000mm f/10, multiply by an old M42 Soligor 2x teleconverter:

https://flic.kr/p/2kV8BnG

As you ca see, here not the haze, but the image noise level was the biggest enemy, what I could decrease by a good IBIS and a better resolution/noise level ratio sensor.

This is an EF 24-105 at 24mm, please see the distance tower at the top of the hill:

Facebook

www.facebook.com

Here is the same tower at 4000mm, an MTO f/10 1000mm reflex mirror with a pair of old M42 Soligor and Takumar 2x converters, handheld:

Facebook

www.facebook.com

Here is a photo taken almost horizontal, hot summer, ambient temperature 38 degrees Celsius, EF 100-400 Mk2 with a Canon MkIII 2X extender at ~600mm:

Facebook

www.facebook.com

Here is the same at a few minutes later, second fly by, 1400mm, MTO 1000mm reflex mirror with a Canon 1.4x MkIII extender:

Facebook

www.facebook.com

Same configuration at 1-2 degrees Celsius, fogy winter condition (Lightroom dehaze tool was slightly applied):

Facebook

www.facebook.com

 

[MihalyLaczko]

Some additional examples for the 1000mm reflex mirror plus a single 2x extender, 2000mm:

https://www.facebook.com/laczkophotography/photos/a.3162074307379548/3162074220712890

Facebook

www.facebook.com

https://www.facebook.com/laczkophotography/photos/a.3162074307379548/3162077354045910

And for the 1000mm plus a double 2x teleconverters, 4000mm:

Facebook

www.facebook.com

EF 24-105 at 24mm:

Facebook

www.facebook.com

4000mm:

Facebook

www.facebook.com

 

[juan55]

I may have misunderstood your post, but I get the impression that you think Canon's "DO" lenses include a mirror?

As you can see from the diagrams of the EF 400mm F4 DO and the RF 600mm F11, "DO" lenses do not have mirrors, only lenses .

View attachment 207379

View attachment 207380

Let's say I did refer to the old "catadioptric" mirror system solution versus the new "refraction" lens

Canon EF 400 mm f/4 DO IS II USM review - Introduction - LensTip.com

Best digital cameras and lens reviews. If you are looking for the information about digital cameras and lenses you are in a right place. We have many professional tests of digital photography equipment.

www.lenstip.com

 

[tapanit]

Tripod is not an option for a contrail or even for a long distance aircraft photography, due to the very narrow field of view, combined with the realtive position of the target and the relatively fast moving target. Just imagine to make a photo with 2000mm about a continously moving airplane, what you have to find by your own naked eye first. A video pan head has its own initial angle limitation, which is really slows down the reaction time. Previously, I made a photo about the ISS with the single f/10 1000mm mirror lens, handheld, because of the position of the object (right above my head, on the Zenith). So these circumstances are not really god for a tripod-related photography. I also tried a monopod, but was not so helpful at all. The slowest shutter speed ever was 1/640, but the finest sweetspot is around 1/1000 - 1/1250, depending on the actual relative angle and position of the path of the aircraft. And the shutter speed is crucial, since the reflex mirror has a lack of sharpness, combined with a really narrow depth of field. And the fact, that you could rotating over the infinite distance mark the tube, you could easly miss the right focal point. By a native 1000mm mirror focal length (the extenders are really does not matter) the depth of field is extremly narrow, so between an 1000 or an 1500 meter, I have to set the focus, with the possibility of the overrotating.

Have you tried a gimbal head? I find it way faster than any other type of head with long lenses, fast enough to track a flying bird. I've never tried it with aircraft though, so I don't know how well it'd work there.

 

[AlanF]

Let's say I did refer to the old "catadioptric" mirror system solution versus the new "refraction" lens

Canon EF 400 mm f/4 DO IS II USM review - Introduction - LensTip.com

Best digital cameras and lens reviews. If you are looking for the information about digital cameras and lenses you are in a right place. We have many professional tests of digital photography equipment.

www.lenstip.com

The "D" in DO refers to "diffraction" optics.

 

[justaCanonuser]

No doubt about bokeh quality, that is definitely one aspect putting these lenses in a niche market — at least based on the information we have at this point, which is far from the entire picture.

Poor eagle, looks like she or he is ashamed about the impressively ugly background bokeh I think mirror tele lenses need a completely smooth background, this image would only work with a dark background w/o any texture. So in most settings such lenses are useless ...

 

[David_E]

Bad wrap? Please make that bad “rap.”