Can computer design such an optic if we ask to create 32 steps grey chart out of 16 ?

Mustafa Umut Sarac · Aug 21, 2014

I thought genetic algorithms is the best toy in unthinkable design problems and asking the correct question to the computer is the most important.

Lets keep it basic , if we put to the image side 16 steps grey chart and ask to computer to cast 32 steps chart to the focal plane with designing a filter ,
3d printed lens or flat glass , what would be the answer. I want to keep the distances for 90 cms from the lens because I use my Leicas at that distance most successful.

I think may be answer would be a blur . May be not ?

What do you think ?

Umut

BrianShaw · Aug 21, 2014

How do you define a step?

Xmas · Aug 21, 2014

Hi Umut

I refer you to your signature line.
With the normal design software you give the software the design and it optimises it and tells how much tolerance you are allowed on glass and mechanical errors.
eg you start with a triplet.

Noel

Mustafa Umut Sarac · Aug 21, 2014

Brian ,

How do you define a step?

If I learn that computer can cope with ordinary grey chart , I would make it more complex later.

Noel ,

With the normal design software you give the software the design and it optimises it and tells how much tolerance you are allowed on glass and mechanical errors.
eg you start with a triplet.

I want to be at wild side , genetic algorithm could figure out not expected , IMHO it would depend on ray tracing or wave tracing difference .

ps. I posted an article on filter design for microscopy and use that technology by us , its a Leica invention , I want to ask the computer a 2d cut plate design single layer or multilayer in front of taking lens or between taking lens and film.

But I cant theorize it in my mind whether it works or not.

Let me hear your thoughts ?

I dont access to highly programmable code v software or zemax and I could not figure out how kde works.

An test program is needed. So I need to program the test software. If I could program to test a virtual design , I could may be ask to create virtual designs , software was kde or similar, I could not remember.

Lets hear from the experts and I will ask to few lens design software programmers.

Dan Fromm · Aug 21, 2014

Mustafa Umut Sarac said:
if we put to the image side 16 steps grey chart and ask to computer to cast 32 steps chart to the focal plane with designing a filter ,
3d printed lens or flat glass , what would be the answer.

In the world, as opposed to targets used for testing, tones can vary continuously. What's the point of the magic device you want to have? What will it do that the lens(es) you already have can't?

Or are you asking for a lens that reduces contrast? If so, you need a lens with severe veiling flare. You can make one, at the cost of ruining an otherwise perfectly good lens, by dismantling the lens and removing its barrel's interior blackening and baffles.

Mustafa Umut Sarac · Aug 21, 2014

Read , with the God's name , Read. First verse from Koran.

jp498 · Aug 21, 2014

I would think it be foolish to create more steps optically if there are computers involved, unless you have a particular artistic purpose. Just capture it straight and make a software function/filter for manipulating tones. (like determine the difference between steps and shift some tones half way between the difference or expand the range if that's the intent.)

Dan Fromm · Aug 21, 2014

Mustafa Umut Sarac said:
Read , with the God's name , Read. First verse from Koran.

If this is a response to the questions I asked, please answer them more clearly. And please don't bring religion into this forum.

You're right, I missed at least one possible reading of your question. Are you trying to produce a step target with 32 tones starting from one that has 16 tones? Are you trying to make art, whatever that means, by photographing targets with stepped tones?

The last isn't entirely a new idea, painters have done much the same starting with a blank surface.

DannL. · Aug 21, 2014

Have you considered using core memory for long term storage of the required algorithms. Of course if your not up on programing the mainframe using hex through a binary key panel, you can buy a book on that. Just be careful that you don't over-tweak the Johnson Rectifier or you may end up with too much reluctance in the core. That can be extremely dangerous. I know from experience. Adding extra flux capacitors and inverting amplifiers will undoubtedly make for smoother data flow. Just saying.

BrianShaw · Aug 21, 2014

The updated Johnson Rectifier is much more accomodating with over-flatulation than the original.

DannL. · Aug 21, 2014

BrianShaw said:
The updated Johnson Rectifier is much more accomodating with over-flatulation than the original.

I was just about to mention that when the Cray XMT died, interrupting my internet connection. Someone left the power cord on the floor and the kids tripped over it, yanking it from the wall. It's just frustrating.

E. von Hoegh · Aug 21, 2014

BrianShaw said:
The updated Johnson Rectifier is much more accomodating with over-flatulation than the original.

I don't think you can store the data in anything other than Tormat memory banks, though.

Mustafa Umut Sarac · Aug 21, 2014

Mustafa Umut Sarac · Aug 21, 2014

Phase-shift masks are photomasks that take advantage of the interference generated by phase differences to improve image resolution in photolithography. There exist alternating[1] and attenuated phase shift masks.[2]

A conventional photomask is a transparent plate with the same thickness everywhere, parts of which are covered with non-transmitting material in order to create a pattern on the semiconductor wafer when illuminated.

In alternating phase-shift masks, certain transmitting regions are made thinner or thicker. That induces a phase-shift in the light traveling through those regions of the mask (see the illustration on the left). When the thickness is suitably chosen, the interference of the phase-shifted light with the light coming from unmodified regions of the mask has the effect of improving the contrast on some parts of the wafer, which may ultimately increase the resolution on the wafer. The ideal case is a phase shift of 180 degrees, which results in all the incident light being scattered. However, even for smaller phase shifts, the amount of scattering is not negligible. It can be shown that only for phase shifts of 37 degrees or less will a phase edge scatter 10% or less of the incident light.

Plot of scattered light (normalized to incident light) as a function of the phase of a phase edge.
Attenuated phase-shift masks employ a different approach. Certain light-blocking parts of the mask are modified to allow a small amount of light to be transmitted through (typically just a few percent). That light is not strong enough to create a pattern on the wafer, but it can interfere with the light coming from the transparent parts of the mask, with the goal again of improving the contrast on the wafer.

Mustafa Umut Sarac · Aug 21, 2014

Above is the Leica's newest filter for creating 3d relief on film from transparent living organism.

I had been inspired by Leica but were not aware such a inventions other uses. It just came from photolithography.

Well , for wafers , that phase mask resolution is so high and calculating is difficult.

But our photography does not require exact precision but visual impact.

Leica does it with two rectangular , one per filter combination.

I am after more artistic , more Leicaisque relief out of ordinary lens or may be eyeglass with printing in 3D.

Chris Lange · Aug 21, 2014

What is the ultimate goal here?

EDIT: Simultaneous posting. There is no such thing as "Leicasque Relief".

Steaphany · Aug 21, 2014

The calculations would be somewhat involved to utilize an AI to achieve imaging optic design. I suggest you take a look at "Lens Design Fundamentals" by Rudolf Kingslake and R. Barry Johnson. The text has 17 chapters and is over 500 pages of "burn a hole through your brain" technical and engineering aspects, with lots of formulae, charts, and tables, for imaging lens system design.

BTW, if your lens criteria is dynamic range and step sensitivity, this is not an optic characteristic, but more of an imager / recording medium characteristic.

I have this text in my library, not for sleep inducing nights, this stuff is fun and keeps me awake, but to understand why camera and imaging lenses possess various characteristics and for me to eventually try my hand at DIY lens design.

Mustafa Umut Sarac · Aug 21, 2014

Chris , Leitz was one of the more than 1000 lens manufacturers in Germany more than 100 years ago. They started to business with microscopy and they have intense competition. Of course they are aware of not only technical but visual qualities needed in lens results. They say they carried that information to camera design , Max was a microscope lens designer also.

If you do buy 1930s to 80s lens magazines - I am reading them for 28 years and if you train your eye to differences , you find how superior your Leica is. I found that relief even at your images also. May be relief is wrong but traditional work but it is roundness and smoothness and placement of shadows and lighest tones and their border lines , transitions and place of the high frequency noise.

And you can see the muscles , bones , lots of borders of tones , dances of the high noise at night human pictures.

You must have photographs with all these qualities and I did not write on color but you dont know to discriminate these as qualities.

You must have a eye for classical paintings , Goya to Matisse. Buy catalogs and look them for hours. If you cant do this , its not your business.

Mustafa Umut Sarac · Aug 21, 2014

Chris,

If you send me your e mail address , I will try to send you some china ink bw drawen comics. Relief is that.

railwayman3 · Aug 21, 2014

Some of the posts in this thread remind me of the late, great, Professor Stanley Unwin. (Older UK readers may remember him.).

Mustafa Umut Sarac · Aug 21, 2014

Steaphany,

Do that book cover phase plates , I am not very much interested in spherical design but may be cylinder lenses due to easier production also.
I can send you least 30 ebook lens design books which very intense. If you want , pm me your address and prepare yourself for 40mb per bool or get a gmail and use it as library.

I am trying to involve in lens design

Thanks,

Umut

Mustafa Umut Sarac · Aug 21, 2014

Lets make a summary of all with current technologies and update the ideas

There are,

- Computational photography comes from MIT , where there is a black film with holes used as lens. No phase effect at these holes but a reversing the blurred image referencing to original pattern.

- There is pin sieve , told , there is phase relation between the holes but neither there is toning at the holes nor different thicknesses and result is weak.

- There is lithography and microscopy phase plates , phase relation is there , microscopy does it with 3 or so holes.

I am after the final one,

basic design and production with inkjet oor laser printing ,

different tones for holes

no computationaly reversing the blur

sharp image

huge variable and controllable final image and may be a program with simulator

phase relation

Lets see , how that is done at world , I will put papers and continue the discussion.

Dan Fromm · Aug 21, 2014

Mustafa Umut Sarac said:
I am trying to involve in lens design

Good for you. There are free on-line courses. Here's one: http://ocw.mit.edu/courses/mechanical-engineering/2-71-optics-spring-2009/index.htm

There are also many books on optical design. Google will find many for you.

Buy a book and work through it or take a course. You'll learn more that way than you will from random answers to random questions on a bulletin board.

DannL. · Aug 21, 2014

Mustafa Umut Sarac said:
Come on folks , there are only snakes and indians in oklohama or whereever it at the hell and the terrorists

Umut, You have a strange way of addressing technical issues. The smiley face you attached to your post isn't working. I would recommend that we address "our peoples" in a respectful manner, in the future. I hope you will agree.

E. von Hoegh · Aug 21, 2014

Mustafa Umut Sarac said:
Chris , Leitz was one of the more than 1000 lens manufacturers in Germany more than 100 years ago. They started to business with microscopy and they have intense competition. Of course they are aware of not only technical but visual qualities needed in lens results. They say they carried that information to camera design , Max was a microscope lens designer also.

If you do buy 1930s to 80s lens magazines - I am reading them for 28 years and if you train your eye to differences , you find how superior your Leica is.

snip

Mustafa, for the last time this is arrant rubbish and nonsense. Reading magazines for 28 years?? You're funny!! I've been making and printing negatives for 40. Two M3s and a IIIG, a Leitz CL, an R3; an assortment of Leitz glass. It's superb. It isn't magical. I no longer have Leitz gear - I'll take one as a gift, if offered, but the gear I now have serves my needs and wants better.

Buy a Nikkormat for $25, stick a $40 50/2 Nikkor H on it, get some T-Max 100 and take some pictures. Print them 8"x12" and matt them say 16"x20". You'll be stunned at the quality, that is if you can actually use a camera and darkroom. You'll be a lot happier and you won't be bothering us with nonsensical threads which betray your ignorance and laziness.
( bye, back to ignore.)

Can computer design such an optic if we ask to create 32 steps grey chart out of 16 ?

Privacy & Transparency

Privacy & Transparency