New curved Camera sensors

I am NOT stating "it cannot be done"...I am simply raising issue to the NEED to make it happen repeatedly to the same form! Or else the flexible sensor to be used for both legacy plus curved sensor lenses cannot happen.
I previous did explicitly state, "The key issue is achieving the identical curvature every time, when put into the 'curved mode'" when dealing with the a flexible sensor, and not merely grinding away semiconductor substrate into a fixed form curvature.

Wilt wrote in post #19277627
I am NOT stating "it cannot be done"...I am simply raising issue to the NEED to make it happen repeatedly to the same form! Or else the flexible sensor to be used for both legacy plus curved sensor lenses cannot happen.
I previous did explicitly state, "The key issue is achieving the identical curvature every time, when put into the 'curved mode'" when dealing with the a flexible sensor, and not merely grinding away semiconductor substrate into a fixed form curvature.

Like adaptive optics do now.

Capn Jack wrote in post #19277629
Like adaptive optics do now.

But adaptive optics for image waveform shaping (the lens) seems inherently different than adaptive form curvature of a sensor.

Wilt wrote in post #19277650
But adaptive optics for image waveform shaping (the lens) seems inherently different than adaptive form curvature of a sensor.

They don't shape a lens with adaptive optics. They shape a surface, a mirror. Is that so different than shaping a surface with sensor elements on it?

I think cameras are full of things that seem too difficult to be made.

IBIS, particularly 5 Axis.
Sensors have to be precisely positioned for good results and IBIS moves the sensor to compensate for movement and returns the sensor to precisely the correct position when no movement detected IBIS is turned off.

Is bending a sensor any more difficult? I suspect manufacturing and design of a sensor that can bend may be the biggest challenge, not the technology of bending it accurately.

I don't know how many RX1R series cameras Sony sold but a similar camera with a curved sensor is possible.
They have small size and a fixed fantastic 35mm prime lens with a high price tag. A curved sensor would allow the lens to go from small to pancake and likely make it far more appealing to buyers than the original.
It's clear Fuji has had good success with the X100 series cameras with fixed 23mm f2 lens (on APS-C sensor).
That's another example of likely sales success with curved sensor.

Cameras don't have to appeal to you or seem likely to be sales success to be released.
eg above Sony RX1R and Fuji X100.
Sigma dp Quattro series or DP Merrill series.

Recent history indicates Sigma more likely to release a curved sensor camera, interchangeable lens or fixed.

Wilt wrote in post #19277488
The key issue is achieving the identical curvature every time, when put into the 'curved mode' vs. the 'flat mode' sensor configuration

Here is some technology that does just that. One minute into the video is a cool visualization.
https://youtu.be/4gvPl​3qWZIM

Yes, the technology does exist for adaptive optics, when the mirror is as large as an observatory telescope. Individualized panels that can each be adjusted is quite a different animal from altering the curvature of a monolithic piece of silicon upon which a sensor circuit is imprinteed...not individual segments that can be mechanically adjusted individually.
I am not inherently opposed to the idea of flexible sensors, simply realistic in comprehending there are some complex issues to need to be reliably solved in shrinking the adaptive optics principle for the micro sensels involved in camera sensors, not the large panels of 4' and larger telescopes

Wilt wrote in post #19278600
Yes, the technology does exist for adaptive optics, when the mirror is as large as an observatory telescope. Individualized panels that can each be adjusted is quite a different animal from altering the curvature of a monolithic piece of silicon upon which a sensor circuit is imprinteed...not individual segments that can be mechanically adjusted individually.
I am not inherently opposed to the idea of flexible sensors, simply realistic in comprehending there are some complex issues to need to be reliably solved in shrinking the adaptive optics principle for the micro sensels involved in camera sensors, not the large panels of 4' and larger telescopes

Or very small adaptive mirrors:
http://www.irisao.com/​product.ptt111.html
Why does the sensor needs to be made of silicon? Why does it need to be monolithic, if it is made of silicon? Why not MEMS to move it precisely?

infrared guy wrote in post #19278503
Here is some technology that does just that. One minute into the video is a cool visualization.
https://youtu.be/4gvPl​3qWZIM

Is it the same Ball that makes canning supplies?
Who would have thought it?
https://www.walmart.co​m …BEgINY_D_BwE&gc​lsrc=aw.ds

Capn Jack wrote in post #19278642
Or very small adaptive mirrors:
http://www.irisao.com/​product.ptt111.html
Why does the sensor needs to be made of silicon? Why does it need to be monolithic, if it is made of silicon? Why not MEMS to move it precisely?

The fundamental principle of semiconductors as a monolithic electrical substrate of doped silicon, upon which there are photoresistive layers that permit other dopants to penetrate the silicon selectively to form layers that make up transistors that make up memories or photosensors. Not a monolith, that adds the complexity of how to form electrical connection, millions of them, between the segments. If there are 6000 pixels in a single direction and 4000 rows of them, how would you move a single pixel when it is (36mm/6000) x (24mm/4000) in size?!

Wilt wrote in post #19278652
The fundamental principle of semiconductors as a monolithic electrical substrate of doped silicon, upon which there are photoresistive layers that permit other dopants to penetrate the silicon selectively to form layers that make up transistors that make up memories or photosensors. Not a monolith, that adds the complexity of how to form electrical connection, millions of them, between the segments. If there are 6000 pixels in a single direction and 4000 rows of them, how would you move a single pixel when it is (36mm/6000) x (24mm/4000) in size?!

You must work in management. Everything is impossible to you, you don't know what's being done, and you don't read the entire post! All the stuff you mentioned isn't very thick- only around 10 microns. And you don't need to move individual pixels to make a curved surface You just move the points of the surface that define the required curvature- this isn't something where there are singularities that prevent use of a smooth curve.
Here's some example of flexible silicon electronics:
https://www.idtechex.c​om …ock-their-potential/20717
https://spectrum.ieee.​org …or-flexible-silicon-chips

As for reading the entire post- I did ask why silicon? We're making organic circuits now that are flexible.

Capn Jack wrote in post #19278660
You must work in management. Everything is impossible to you, you don't know what's being done, and you don't read the entire post! All the stuff you mentioned isn't very thick- only around 10 microns. And you don't need to move individual pixels to make a curved surface You just move the points of the surface that define the required curvature- this isn't something where there are singularities that prevent use of a smooth curve.
Here's some example of flexible silicon electronics:
https://www.idtechex.c​om …ock-their-potential/20717
https://spectrum.ieee.​org …or-flexible-silicon-chips

As for reading the entire post- I did ask why silicon? We're making organic circuits now that are flexible.

No, everything is NOT 'impossible'. But a wise person understands what needs adequate resolution in order to SUCCEED, and develops a plan to overcome every one of the identified obstacles. The failure to do that adequately that leads to a failed businesses.

I did

• take a stanant business, identified the reasons for its stunted growth, came up with a plan to address the issues, and in five years had grown sales by 2.75X and profit to 5.5X (to $55 Million annual) the starting amount before I was hired
  
• I also took a business that GE had trield to sell off for years without success, developed a plan to improve the products via Engineering activity, and after 2 more years GE was able to sell that business to their #1 competitor in that industry segment.
  
• I am not an engineer but I can find engineers to solve technical problems. One project from the mid-1990s (updated with technical modifications over the interim) was still being purchased and used by physicians in 2020.
  

Wilt wrote in post #19278726
No, everything is NOT 'impossible'. But a wise person understands what needs adequate resolution in order to SUCCEED, and develops a plan to overcome every one of the identified obstacles. The failure to do that adequately that leads to a failed businesses.

I did

• take a stanant business, identified the reasons for its stunted growth, came up with a plan to address the issues, and in five years had grown sales by 2.75X and profit to 5.5X (to $55 Million annual) the starting amount before I was hired
  
• I also took a business that GE had trield to sell off for years without success, developed a plan to improve the products via Engineering activity, and after 2 more years GE was able to sell that business to their #1 competitor in that industry segment.
  
• I am not an engineer but I can find engineers to solve technical problems. One project from the mid-1990s (updated with technical modifications over the interim) was still being purchased and used by physicians in 2020.
  

You don't do any of that if starting with 'not possible' point of view. In shoirt, I grew/developed businesses. Now I am retired, but still consult for international companies.

Like I said, you are (were) management with a management mindset. You know the business reasons for not investing in such a sensor and so create artificial obstacles against it. I agree making such a flexible sensor would require R&D to make it work reliably. I also don't believe there is customer demand for such a sensor to justify the investment, speaking as a user (and not any sort of business or technical person). It's users that ultimately give the return on investment.

Capn Jack wrote in post #19278814
Like I said, you are (were) management with a management mindset. You know the business reasons for not investing in such a sensor and so create artificial obstacles against it. I agree making such a flexible sensor would require R&D to make it work reliably. I also don't believe there is customer demand for such a sensor to justify the investment, speaking as a user (and not any sort of business or technical person). It's users that ultimately give the return on investment.

You ASSume that reasons for not doing something are reasons to 'create artificial obstacles'...I did NOT say that in my previous post!

I said that if you know the obstacles that are in the way, you try to determine what might be done to overcome the obstacles in order to succeed. Most plans are not perfect, almost all have issues needing to be overcome. Sometimes the issue is training Sales in an area they are unfamilar with!
In evaluating opportunities, a wise business considers, at the most basic level,

1. 'Can we make the product?...is design and manufacturing feasible?
   
2. 'Is there a market for that product'...will customer buy it?
   
3. 'What obstacles exist within our company, for our company to be able to succeed with that product in those markets?'...internal training, ability to manufacture at a cost that permits sufficient profit?
   

Wilt wrote in post #19278918
You ASSume that reasons for not doing something are reasons to 'create artificial obstacles'...I did NOT say that in my previous post!

I said that if you know the obstacles that are in the way, you try to determine what might be done to overcome the obstacles in order to succeed. Most plans are not perfect, almost all have issues needing to be overcome. Sometimes the issue is training Sales in an area they are unfamilar with!
In evaluating opportunities, a wise business considers, at the most basic level,

1. 'Can we make the product?...is design and manufacturing feasible?
   
2. 'Is there a market for that product'...will customer buy it?
   
3. 'What obstacles exist within our company, for our company to be able to succeed with that product in those markets?'...internal training, ability to manufacture at a cost that permits sufficient profit?
   

When obstacles cannot be eliminated, you throw out that idea, and look for better opportunities. When obstacles CAN be eliminated, you continue to consider that plan, and put it into action what you can invest in.

So in my participation in the thread about flexible sensors, I first asked 'what is the benefit?' Then I identified one of the obstacles, 'controllable and repeatable bending' if you want to use both newly designed lenses as well as legacy lenses.
You identified that there might not be enough customer demand, another obstacle. Between us, we began to cover #1 and #2 in the list. Both obstacles might well make such a project a foolhardy one. I render no conclusion, I merely am engaged in discussion.

The bolded part is the real obstacle, in my opinion. The other obstacles, the technical ones you mentioned earlier in the thread seem to be at the point of being worth a try.

Perhaps I don't see your comments as "Is there a way to make a sensor flexible?", or "can we make a flexible surface small enough to be a camera sensor?". I see "The fundamental principle of semiconductors as a monolithic electrical substrate of doped silicon, upon which there are photoresistive layers that permit other dopants to penetrate the silicon selectively to form layers that make up transistors that make up memories or photosensors." as suggesting it can't be made flexible (and yet flexible silicon circuits exist). Likewise "Yes, the technology does exist for adaptive optics, when the mirror is as large as an observatory telescope." seems to preclude the possibility of the small ones you can buy now from Edmunds. A better sequence of questions is "How would you do XXX?", followed by "Can we do XXX?", which is what you are suggesting in your list of 3 items.