spherical aberration solved?

[Bruce Watson]

Looks like the 2000 year old question of how to eliminate spherical aberration (https://science.slashdot.org/story/19/07/08/2214254/physicist-solves-2000-year-old-optical-problem) in camera lenses might actually be solved.

Some of you are good at lens math. What say you?


Looks like this is old news from last year. Well, maybe someone will find it interesting. Has this been reported on LFP.info before? If so I must have missed it.



Published in Applied Optics, they'll let you read the abstract for free (https://www.osapublishing.org/ao/abstract.cfm?uri=ao-57-31-9341) anyway.

If you can make a singlet lens that's free of spherical aberration it would seem you could eliminate some elements in a view camera lens, making it sharper, lighter weight, and cheaper. If anyone were still making large format lenses that is. Maybe Cooke will make some just for fun as they have done before.

But even if they did, and this eliminated some of the sharpness problems of shooting wide open, would we? Shoot wide open I mean. Because in LF, there are other reasons to stop down. But more choice would be better than less choice, yes?

[Kevin J. Kolosky]

I have to start eating more bread with Nutella spread on it!

[ic-racer]

193164

[Pere Casals]

IMHO spheric aberration cannot be considered as the single factor, because we also want field flatness, low distortion, to correct primary and secondary chromatic aberrations, low fall-off... we also want the lens to work well for a range of magnifications... and we also may want a good bokeh...

That paper should have importance for scholars, but IMHO it is no industrial revolution, it's an analytical method... but I'd say that current designs are optimized with software.

Of course aspheric surfaces are useful to correct spheric aberration, but manufaturing high quality aspheric surfaces has been industrially challenging, what is an industrial revolution is the way that manufacturing has improved, specially in costs.

An remarkable industrial leap forward was compound aspherical elements. In the cheapo (now old) Nikon 28-80mm 3.3-5.6 kit lens the front element is a compound aspherical. "This means Nikon glues a thin plastic corrector over another spherical glass element". The result was an atonishing performance in a crazy cheap lens made in mass production conditions.

https://kenrockwell.com/nikon/28-80mm-g.htm


Amazigly, manufacturers don't play care in the important/easy things, like a round iris or the spheric aberration correction strategy in the OOF that delivers a pleasing bokeh.


This gives an advantage to true LF artists, they have little competition when beauty counts because they have the true good glass for capturing beauty. Today industry is too focussed in dxo ratings and sadly ignoring beauty.

[Mark Sawyer]

For all practical purposes, spherical aberration was eliminated by WWII or earlier. This is sort of like someone saying they've just designed the first apochromatic lens.

[Drew Wiley]

Toy store talk. Look up Tinsley Labs. There's a little film showing the operation on their website (coherent.com) It's just up the road about ten minutes from me. You identify what kind of lens you need, you can have it with absolutely no compromises. Just remember to bring along your NASA or NSA credit card, with no spending limit of course.

[Pere Casals]

Mark, yes but by wwii that required several elements, now it is done with a bi asph singlet, analytically.

[Havoc]

Question remains: will my photos be any better because of it?

[Pere Casals]

What is "better" in the photography art?

[Mark Sawyer]

Mark, yes but by wwii that required several elements, now it is done with a bi asph singlet, analytically.

Then they've got chromatic aberration, coma, curved field, and distortion to replace the spherical aberration. Yay...

[Pere Casals]

Then they've got chromatic aberration, coma, curved field, and distortion to replace the spherical aberration. Yay...

Yes, of course... this is what it is, an analytic way to calculate asph singlets that are free fom sph aberration, so this is math.

[Drew Wiley]

It all depends on the glass type. Do you saw off the bottom of a Coke bottle or a Pepsi bottle?

[Bruce Watson]

For all practical purposes, spherical aberration was eliminated by WWII or earlier. This is sort of like someone saying they've just designed the first apochromatic lens.

Not with a singlet -- a single piece of glass. It always took at least two, which is perhaps why lenses of the period used so many cemented doublets and triplets.

What makes this interesting to me is that it can be done with a singlet. Which should allow replacement of a cemented doublet or triplet with a singlet, which should result in a lighter and physically smaller camera lens. Not to mention somewhat better light transmission because of less glass, and less internal reflections because of fewer surfaces, etc. And of course it might be cheaper as well, because less glass even if the singlet is more expensive to make.

So... lighter, smaller, faster, cheaper. What's not to like?

Bottom line for me is that I could hike farther, and make photographs where I could not before. And others like me could do the same. And that, might make it worth doing. IDK.

And lets not leave out the fun of solving an "unsolvable" problem. :cool:

[Mark Sawyer]

So... lighter, smaller, faster, cheaper. What's not to like?

The distortion, curved field, chromatic aberration and coma. Zeroing out one aberration at the expense of ignoring others may be an interesting math problem, but not a practical solution to anything.

[Bruce Watson]

The distortion, curved field, chromatic aberration and coma. Zeroing out one aberration at the expense of ignoring others may be an interesting math problem, but not a practical solution to anything.

This is the difference between science and engineering. Science is about figuring it out, engineering is about using it to build things. The scientists in this case have figured out how to solve spherical aberration in a singlet. Good on 'em. Really, they've done great work here.

Engineers haven't yet used this to design lenses AFAIK. Maybe the optical engineers will eventually, maybe not. But if they do, my bet is this will show up first in cell phone cameras, not lenses for us. It may never show up in lenses for LFers.

[Ulophot]

Not that we'll likely see much of this in LF, but it may be of interest to some. It caught my eye today, so I'm passing it along.

https://www.techspot.com/news/81357-scientist-solves-centuries-old-problem-lead-cheaper-sharper.html

[Oren Grad]

Threads merged.

[Nodda Duma]

Incorrect. In fact, spherical aberration can be corrected with a single surface. Sir Isaac Newton corrected spherical aberration, lateral color, spherochromatism, astigmatism, and distortion with the paraboloidal surface in the 1600s.

The single lens of a CD reader corrects spherical aberration entirely.

Pere is right, this was a well-known academic problem and good for him solving it, but it is purely academic. The physicist worked out a mathematical solution — which is great — for a problem that is solved via other methods every day by lens designers. That’s a great way to earn his doctorate. But in reality it can actually take longer to set up the initial conditions in Zemax or Code V than to run the optimization itself that drives sph ab to zero ..which is not much faster for a simple lens than doing the Seidel coefficient calculations by hand.

So leading to cheaper, simpler solutions... I shudder at the thought of trying to convince an optical shop to turn a surface like that, or even a molded optics shop to do the same. More expensive and more complex.. definitely.

Not knocking what he did, don’t get me wrong. I have issue with the sensationalist misleading reporting that followed. It is just a bit too much.





Not with a singlet -- a single piece of glass. It always took at least two, which is perhaps why lenses of the period used so many cemented doublets and triplets.

What makes this interesting to me is that it can be done with a singlet. Which should allow replacement of a cemented doublet or triplet with a singlet, which should result in a lighter and physically smaller camera lens. Not to mention somewhat better light transmission because of less glass, and less internal reflections because of fewer surfaces, etc. And of course it might be cheaper as well, because less glass even if the singlet is more expensive to make.

So... lighter, smaller, faster, cheaper. What's not to like?

Bottom line for me is that I could hike farther, and make photographs where I could not before. And others like me could do the same. And that, might make it worth doing. IDK.

And lets not leave out the fun of solving an "unsolvable" problem. :cool:

[Jac@stafford.net]

Incorrect. In fact, spherical aberration can be corrected with a single surface. Sir Isaac Newton corrected spherical aberration, lateral color, spherochromatism, astigmatism, and distortion with the paraboloidal surface in the 1600s.

Half of a regular paraboloid is shown in the picture below. It is an interesting feature from nature.

For a news story in 1973 I photographed Prof. Riccardo Levi-Setti, University of Chicago, holding a hollow spun aluminum cone-like shape (no lens!) which uses the dimensions of a crab's (or trilobite) eye. It is a poor print reproduction which does not show that at the small end is a cup of water very quickly brought to a boil, even through a dirty window on a bright overcast day.

He found the dimensions using a scanning ion microscope to discern one cone of a crab's eye
Picture here. (https://cdn.news.uchicago.edu/sites/default/files/styles/full_width/public/images/2018-11/Levi_3.jpg?itok=2QHc9SnD)

More about the professor. (https://news.uchicago.edu/story/riccardo-levi-setti-physicist-and-trilobite-collector-1927-2018)

[Nodda Duma]

Very cool, jac

[Jac@stafford.net]

Very cool, jac

I was impressed! Something nature figured out 500 million years ago! I wonder if we could make micro-lenses for digital sensors like that.

[Nodda Duma]

I was impressed! Something nature figured out 500 million years ago! I wonder if we could make micro-lenses for digital sensors like that.

Nah there’s easier ways. It’s all about the right tool for the job.

The form or I guess you could say the design approach you show is used for focusing x-rays and other very high energy wavelengths. In that application it is the easiest approach to implement.

[Bernice Loui]

Nature did not "Figure this out", Nature wrote ALL the rules. With the passage of time using essentially trial and error various creatures and sometimes humanity also figures out how to make the best symbolic deal to solve a problem that completely fits the rules set down by Nature.

The better one understands the way Nature really is, the better a symbiotic deal can be made. Try to get around the ways of Nature comes a cost with no exceptions.

Doing technical design work often instill a very serious appreciation of how unforgiving Nature really is.


As for this Academic paper presented, caused cringing as this was more of a math exercise than a real world solution. Previously mentioned, any lens grinder that would attempt to make the complex curvature of this proposed solution would face some very serious challenges and problems to solve. Real world solutions must be producible and replicated with not too much difficulty and complexity or the cost per item grows rapidly in a very non-linear way. Designs that have endured the test of time have survived the real world test of solving and meet the specific need well enough at a reasonable cost and ease of production or product evolution would have caused extinction early on in the products life.


Bernice



I was impressed! Something nature figured out 500 million years ago! I wonder if we could make micro-lenses for digital sensors like that.

[Jac@stafford.net]

Nature did not "Figure this out", Nature wrote ALL the rules. With the passage of time using essentially trial and error various creatures [...]
Have you wondered about the necessity of feedback to inform Nature when it has optimized the design, or does change continue to possibly make it worse? Is the feedback simply survival?


The better one understands the way Nature really is, the better a symbiotic deal can be made. Try to get around the ways of Nature comes a cost with no exceptions.

Doing technical design work often instill a very serious appreciation of how unforgiving Nature really is. That's why it is called Mother Nature. :)


As for this Academic paper presented, caused cringing as this was more of a math exercise than a real world solution. Previously mentioned, any lens grinder that would attempt to make the complex curvature of this proposed solution would face some very serious challenges and problems to solve. If you are referring to Levi-Seti's device, it has no glass. It is hollow, spun in a machine shop. This was decades before 3D printing. I met with the machinist.

[Dan Fromm]

Jac, given time to work natural selection can be counted on to find "at least good enough for the situation." It prunes off badly sub-optimal solutions, can't be counted on to find the optimum optimorum. This because of physical limits on what's possible and because situations change and what's at least good enough changes with them. Re this second point, I give you the humble but very well-studied Trinidadian guppy.

[Mark Sawyer]

Looks like this single-element correction is already allowing Canon to come up with some very simplified new designs...