Canon's Blue Spectrum Refractive Element

[Canon Rumors]

<strong>MELVILLE, N.Y, August 27, 2015</strong> – Canon U.S.A., Inc., a leader in digital imaging solutions, today announced that its parent company, Canon Inc., has developed Blue Spectrum Refractive (BR), a new optical element for use in camera lenses that corrects chromatic aberrations at an extremely high level to achieve superb imaging performance.</p>
<p>The new Canon-developed BR optical element offers characteristics that significantly refract blue light, which lies within the short-wavelength range, to achieve impressive levels of chromatic aberration correction for outstanding imaging performance. The BR optical element, positioned between two glass lens elements to create a BR lens, will make its debut in the new EF 35mm f/1.4L II USM wide-angle fixed-focal-length lens, which is scheduled to go on sale in October 2015.</p>
<p>Natural light, or white light, comprises a spectrum of wavelengths, or colors, each of which realizes a unique refractive index when passing through a lens. As all colors do not converge on the same point, this disparity causes chromatic aberrations, or color fringing, to occur in an image.</p>
<p>Canon develops optical elements by reviewing organic optical materials, beginning with the design of molecular structures, with the aim of achieving optimal chromatic aberration correction that suppresses color fringing. With the successful development of the BR optical element, which offers unique light-dispersion characteristics that significantly refract blue light—a wavelength that, until now, had proven particularly difficult to converge to a specific focal point—Canon is able to develop lenses that result in outstanding imaging performance by correcting chromatic aberrations at an exceptionally high level.</p>
<p><strong>Preorder EF 35 f/1.4L II $1799: <a href="http://www.bhphotovideo.com/c/product/1180801-REG/canon_9523b002_35mm_f_1_4l_ii_usm.html/BI/2466/KBID/3296" target="_blank">B&H Photo</a> | <a href="http://www.adorama.com/CA35142.html?utm_term=UbK24x0al34oSlvW4eT8QxjoUkX3mDVXeWC-Ug0&utm_medium=Affiliate&utm_campaign=Other&utm_source=rflaid64393&cvosrc=affiliate.64393" target="_blank">Adorama</a> | <a href="http://amzn.to/1Uehm5w" target="_blank">Amazon</a>  </strong></p>

 

[Chaitanya]

Re: Canon Develops New Camera-Lens Optical Element

I hope they use this element while developing new macro lens either a 60mm or 180mm focal length options.

 

[Camerajah]

Re: Canon Develops New Camera-Lens Optical Element

This is BIG

 

[Machaon]

Re: Canon Develops New Camera-Lens Optical Element

Canon's been doing some pretty impressive stuff with its new L glass. I can't wait to see what sort of CA performance this delivers.

I'd especially like to see this in an EF 85mm f/1.2L III USM before too long.

 

[stefsan]

Re: Canon Develops New Camera-Lens Optical Element

Sounds like a great thing for image quality. But CA is already handled very well in a lot of lenses and can be completely eliminated in post processing. One area where innovation would be most welcome is coma aberrations in (ultra)wide angle lenses – wouldn't it be cool to take a shot of the night sky and getting the stars as spheres and not as triangles? But that is probably much harder to achieve, especially for zoom lenses like the 16-35…

 

[HighLowISO]

Re: Canon Develops New Camera-Lens Optical Element

Congradulations to the team at Canon on this achievement. Looks like it's doing a great job in the new 35mm L, maybe the 50mm L is not that far from an update.

I'm thinking this will also help low light color fidelity.

 

[NorbR]

Re: Canon Develops New Camera-Lens Optical Element

Bingo.

Now, Canon, put this in an 85mm lens, and watch me throw money at you ... :-X

 

[LonelyBoy]

Re: Canon Develops New Camera-Lens Optical Element

<sarcasm>If it doesn't expand DR 3 stops at least, it's worthless!!! Sony is only innovative company!</sarcasm>

Came here to say this, exactly. Superb innovation.

 

[m8547]

Re: Canon Develops New Camera-Lens Optical Element

Sounds like a great thing for image quality. But CA is already handled very well in a lot of lenses and can be completely eliminated in post processing. One area where innovation would be most welcome is coma aberrations in (ultra)wide angle lenses – wouldn't it be cool to take a shot of the night sky and getting the stars as spheres and not as triangles? But that is probably much harder to achieve, especially for zoom lenses like the 16-35…

CA correction in post processing does not always work well. For example, in an image with grass in the foreground, the algorithm seems to get confused if you ask it to correct green CA, and you end up with gray blocks around the grass. At least that's what happened to me in a couple images I processed with Lightroom. When I'm using my $100 18-55 lens I'll gladly correct CA in post processing, but if I'm paying 18x as much for a L lens, I expect better performance.

 

[Famateur]

Re: Canon Develops New Camera-Lens Optical Element

Sounds like a great thing for image quality. But CA is already handled very well in a lot of lenses and can be completely eliminated in post processing. One area where innovation would be most welcome is coma aberrations in (ultra)wide angle lenses – wouldn't it be cool to take a shot of the night sky and getting the stars as spheres and not as triangles? But that is probably much harder to achieve, especially for zoom lenses like the 16-35…

CA correction in post processing does not always work well. For example, in an image with grass in the foreground, the algorithm seems to get confused if you ask it to correct green CA, and you end up with gray blocks around the grass. At least that's what happened to me in a couple images I processed with Lightroom. When I'm using my $100 18-55 lens I'll gladly correct CA in post processing, but if I'm paying 18x as much for a L lens, I expect better performance.

While correcting the *color* of chromatic aberration is often easy in post (exceptions acknowledged), the frustrating thing for me is that the fringe is still visible, even if devoid of color. Haven't seen an easy way to correct that.

I'm excited to see Canon engineers continuing to innovate in optical technology...

 

[rrcphoto]

Re: Canon Develops New Camera-Lens Optical Element

Sounds like a great thing for image quality. But CA is already handled very well in a lot of lenses and can be completely eliminated in post processing.

not really to both points. LoCA on very fast lenses is not handled very well unless you want to pay for it.

while lateral CA can be pretty much handled easily enough, longitudinal is much more complex.

 

[NancyP]

Re: Canon Develops New Camera-Lens Optical Element

"Organic" just means that there's a carbon atom in there somewhere - as opposed to a purely silicon-based compound (standard glass). I wouldn't make any assumption about the longevity of the organic lens material.

 

[chromophore]

Re: Canon Develops New Camera-Lens Optical Element

Sounds like a great thing for image quality. But CA is already handled very well in a lot of lenses and can be completely eliminated in post processing. One area where innovation would be most welcome is coma aberrations in (ultra)wide angle lenses – wouldn't it be cool to take a shot of the night sky and getting the stars as spheres and not as triangles? But that is probably much harder to achieve, especially for zoom lenses like the 16-35…

CA correction in post processing does not always work well. For example, in an image with grass in the foreground, the algorithm seems to get confused if you ask it to correct green CA, and you end up with gray blocks around the grass. At least that's what happened to me in a couple images I processed with Lightroom. When I'm using my $100 18-55 lens I'll gladly correct CA in post processing, but if I'm paying 18x as much for a L lens, I expect better performance.

Indeed. Chromatic aberration is not something that is always correctable in post, and even when it is, it is not a substitute for apochromatic lens design: once CA appears, information is lost, especially in the case of longitudinal/axial chromatic aberration, in which the plane of sharpest focus is a function of the wavelength.

One of the most powerful ways to achieve apochromatic correction is to have at one's disposal a variety of refractive materials of different Abbe numbers (i.e., dispersion), but even more so, to have a materials whose dispersion in one wavelength range is markedly different than in another wavelength range. The reason for this is that although you can have high- and low-dispersion glass, this corresponds mainly to the slope of the refractive index curve plotted against wavelength, with the high-dispersion glass having a steeper slope than the low-dispersion glass. But if the *shape* of these curves are largely similar, you still have a hard time making an apochromat no matter how many elements of each glass you use or how curved they are.

That's why fluorite does so nicely in supertelephoto lenses: fluorite exhibits low dispersion but also anomalous dispersion, just like this "blue refractive" technology. The problem with fluorite, though, is that it's expensive to grow, expensive to grind and polish in mass quantities, and for lenses that need high refractive power due to their shorter focal length (supertelephotos don't need strong curvatures), fluorite has too low a refractive index, meaning it would need to be very strongly curved in a normal focal length design.

So, how do lensmakers like Zeiss and Leica make apochromats? Well, money. These are limited production lenses, expensive as hell, and incorporate a large number of elements with tight tolerances, not to mention their own specially formulated glass. Canon, being a large manufacturer of lenses, basically can't afford to make 10-20x the volume at 50% of the price, and even if they were on par with price, these things would need more handmade attention than Canon can devote....

It's just easier, then, to invent a new technology that can be relatively cheaply but reliably implemented and automated. That's what this BR stuff (hopefully) achieves. If it is really successful, we very might well see it implemented in ALL EF L fast aperture primes, or even the entire EF lens range.

A final comment about the nature of lens design: it is always a game of balancing out compromises. What you gain in one area (say, center resolution) is always at the expense of some other constraint (e.g., corner resolution). A good design either intentionally optimizes in one area and sacrifices another to be a specialist design (85/1.2L II), or it tries to achieve a balance across the board (24-70/2.8L II). If "BR" works as well as it does, this in theory would permit the lens design to be less constrained in other ways. The more tools in a lens designer's toolbox, the more they can do to improve overall performance. So if he/she can correct axial color more effectively and gain apochromatic performance with this technology, this allows more of the other elements of the lens to be devoted to things like correcting comatic aberration, for example. It's a gross simplification but that is the essence of how lenses are designed.

 

[jrista]

Re: Canon Develops New Camera-Lens Optical Element

A final comment about the nature of lens design: it is always a game of balancing out compromises. What you gain in one area (say, center resolution) is always at the expense of some other constraint (e.g., corner resolution). A good design either intentionally optimizes in one area and sacrifices another to be a specialist design (85/1.2L II), or it tries to achieve a balance across the board (24-70/2.8L II). If "BR" works as well as it does, this in theory would permit the lens design to be less constrained in other ways. The more tools in a lens designer's toolbox, the more they can do to improve overall performance. So if he/she can correct axial color more effectively and gain apochromatic performance with this technology, this allows more of the other elements of the lens to be devoted to things like correcting comatic aberration, for example. It's a gross simplification but that is the essence of how lenses are designed.

This is SO not true. Just look at all of Canon's latest lens designs. And many of Nikon's latest. And Zeiss' Otus. There is absolutely no requirement that tradeoffs MUST be made in order to improve something in a lens. Canon's recent optics have been improving both center and corner performance simultaneously. Corner performance in particular has been improving considerably. Just look at the MTF of the new 35mm f/1.4 II. Improvement across the board, radical improvement in the corners.

This is naive and archaic thinking. Tradeoffs are made FOR COST, not because they MUST be made. If cost is no issue, then there is no reason to accept any tradeoff. The Otus line, for example, comes very close to delivering optimal performance from corner to edge, with only a slight loss in PSF quality in the corners.

 

[chromophore]

Re: Canon Develops New Camera-Lens Optical Element

A final comment about the nature of lens design: it is always a game of balancing out compromises. What you gain in one area (say, center resolution) is always at the expense of some other constraint (e.g., corner resolution). A good design either intentionally optimizes in one area and sacrifices another to be a specialist design (85/1.2L II), or it tries to achieve a balance across the board (24-70/2.8L II). If "BR" works as well as it does, this in theory would permit the lens design to be less constrained in other ways. The more tools in a lens designer's toolbox, the more they can do to improve overall performance. So if he/she can correct axial color more effectively and gain apochromatic performance with this technology, this allows more of the other elements of the lens to be devoted to things like correcting comatic aberration, for example. It's a gross simplification but that is the essence of how lenses are designed.

This is SO not true. Just look at all of Canon's latest lens designs. And many of Nikon's latest. And Zeiss' Otus. There is absolutely no requirement that tradeoffs MUST be made in order to improve something in a lens. Canon's recent optics have been improving both center and corner performance simultaneously. Corner performance in particular has been improving considerably. Just look at the MTF of the new 35mm f/1.4 II. Improvement across the board, radical improvement in the corners.

This is naive and archaic thinking. Tradeoffs are made FOR COST, not because they MUST be made. If cost is no issue, then there is no reason to accept any tradeoff. The Otus line, for example, comes very close to delivering optimal performance from corner to edge, with only a slight loss in PSF quality in the corners.

The cost factor is assumed because it is the only constraint that differentiates lenses into broad performance classes.

 

[Mt Spokane Photography]

Re: Canon Develops New Camera-Lens Optical Element

"Organic" just means that there's a carbon atom in there somewhere - as opposed to a purely silicon-based compound (standard glass). I wouldn't make any assumption about the longevity of the organic lens material.

The life of a "Organic Material" is the first thing that came to mind. The varied extreme environments that lenses go thru is pretty tough on organic materials like the greases and lubricants, but they seem to have solved that issue.

Fortunately, accelerated life tests for such materials can predict the life fairly accurately. I've done a lot of that type of testing. Manufacturers of such compounds tend to exaggerate, but a good test reveals all.

 

[keriboi]

Re: Canon Develops New Camera-Lens Optical Element

Why all the trouble? isnt CA fixed one click in photoshop?

 

[LetTheRightLensIn]

Re: Canon Develops New Camera-Lens Optical Element

Sounds like it may be an interesting new way to create full-on APO designs!

Canon sure charges forward with lens design!!!! If only their darn sensor and body division had the same pride and hadn't become such a bunch of timid, tepid, follower, milkers.

But their lens division is still charging full steam ahead.

 

[Famateur]

Re: Canon Develops New Camera-Lens Optical Element

Why all the trouble? isnt CA fixed one click in photoshop?

Kind of. Some aberration is easier to "fix" than others.

Even when de-colored with a single click, the fringe remains. When that fringe is on the edge of back-lit Landscape Arch in an image created with three bracketed exposures, it stands out against the sky and makes it look artificial. Only solution is to carefully remove that fringe in Photoshop. That's a pain.

If the lens produces no fringe in the first place, so much less work...

 

[LetTheRightLensIn]

Re: Canon Develops New Camera-Lens Optical Element

Sounds like a great thing for image quality. But CA is already handled very well in a lot of lenses and can be completely eliminated in post processing.

Not really. I think you are thinking of lateral CA and this will fix longitudinal CA which is entirely different and rather resistant to any correction in post other than trying to use advanced algorithms to hunt for green/purple outlines and shift spectrum and kill saturation on those parts and unless the algorithm is ultra advanced it tends to often still leave a fringe ghost behind even when it does work.

And you can never get the ultimate color performance from a lens when you need to correct in post, even lateral correction which works quite well, leaves very subtle bits of trouble behind.

Tbis is a pretty cool thing they have come up with.