6 ms·

Flat lens offers a perfect image

128 points by co_pl_te 14y ago

cs702 14y ago
As an amateur photographer with a small collection of DSLR and MILC camera lenses at home, I'm quite excited by this: "In the future we can potentially replace all the bulk components in the majority of optical systems with just flat surfaces." In other words, this will allow large-sensor cameras to get a lot smaller and thinner. Edit: WRONG! As sbierwagen points out[1], even though the press release states that the technology is "completely scalable" [to different wavelenghts of light], it actually works only with wavelengths greater than that of visible light. Too bad :-( sbierwagen: Thanks for the clarification. [1] http://news.ycombinator.com/item?id=4433970 http://news.ycombinator.com/item?id=4433970
- djmdjm 14y ago
  Sorry, read sbierwagen's comment for the bad news: metamaterials like those used to construct this lens are tuned for very specific wavelengths so unless your photographic aesthetic is centered on a very particular colour of IR then you will be out of luck for quite a while.
  - sbierwagen 14y ago
    Metamaterial lenses, if you can make them in visible-light frequencies, might be useful in the microlens array for light-field cameras, where they could act as both the bayer color filter pattern and as focusing element, and where their excellent optical qualities would be really helpful. (The Lytro light-field camera's lousy optical performance is mostly due to the microlens array: they have to be made of conventional materials, and the only lenses that would fit are singlets, which have chromatic abbertation and barrel distortion out the ass.) But as the primary light gathering element, no.
mrich 14y ago
As a myopic geek I am looking forward to ultra flat glasses :)
sbierwagen 14y ago
Operating at telecom wavelengths (i.e., the range commonly used in fiber-optic communications), the new device is completely scalable, from near-infrared to terahertz wavelengths, and simple to manufacture. Ho ho, very clever use of words there. Terahertz is far-far infrared, (1mm to .1mm (1,000,000-100,000nm) wavelength) while NIR is around 1,000nm. Visible light is 740-380nm. To the best of my knowledge, (this isn't my field) nobody has ever made a visible-light metamaterial. The antennas are too small. There's no real physical reason we can't make red-light metamaterials, the problem is just fabrication. (A parallel can be drawn here between metamaterials, desktop nanofactories, and fusion reactors) Additionally, it would be tricky to make a panchromatic metamaterial lens: metamaterials are tuned very precisely to one frequency, (color) constrained by the physical size of the antennas, and they can't transmit any others.
- soup10 14y ago
  I don't get why every science article gets negative comments like this. Just because something doesn't have immediate practical applications doesn't mean its not interesting research. Of course it's annoying when they are overly optimistic(either because of journalists trying to make the finding interesting, or researchers trying to get more funding), but that's to be expected.
  - DenisM 14y ago
    Where headline is not accurate, it bears pointing out. I had no idea that this is not in fact a visible light lense.
    - jasonwatkinspdx 14y ago
      The headline is accurate (unless hacker-news had one that has one that has now been edited away). If you had read the article you would know precisely what wavelengths the device is applicable to (3rd paragraph even), and why it's an exciting research development.
      - 14y ago
blacksmythe 14y ago
From the summary article, it looks like they have made a Fresnel lens, which can also be quite flat but works over a limited wavelength range of light. In order to achieve broadband operation, the phase change vs wavelength of the metamaterial needs to approximate a large delay. If a large delay is not feasible to implement, this summary is overstating the importance of this result compared to prior technology.
co_pl_te 14y ago
Hi, sbierwagen. This is certainly beyond my field of expertise, but I don't see how the inability of this new lens to focus in the range of visible light detracts from the main point of it being a distortion-free lens whose "focusing power approaches the ultimate physical limit set by the laws of diffraction." Although I agree that the press release makes it appear as if this will have direct implications on imaging in the lay sense (this isn't going to make a future iPhone the perfect camera), there are certainly applications of such technology that aren't consumer-focused and don't require visible light that stand to benefit from this research. As blacksmythe notes, the lens created is not your stereotypical refractive lens, but is more along the lines of a Fresnel lens or zone plate. I'd be interested in hearing more about the differences between the two, especially with regard to the limitations of each. Would anyone who read the write up in Nano Letters or has a better understanding care to comment more on the implications? I agree the press release is a tad sensationalistic and possibly (intentionally) misleading — but what do you expect from a press release? Although the findings presented may not be as earth-shattering as a mass-produceable visible-light metamaterial, I still think it's a noteworthy development that shouldn't be dismissed so easily.
- ehsanu1 14y ago
  OT/Meta: Please use the reply link to post replies, so that coherent threads can be maintained.
th0ma5 14y ago
Anyone play with a pinhole on a CCD or CMOS sensor, or otherwise "no" lens? For a non-lens-distorted picture?
- vladoh 14y ago
  As far as I know the problem is that the amount of light that goes trough the hole is very low and you cannot see much or you need a very strong light source.
  - Sharlin 14y ago
    Or you have to expose for a long time.
    - Luc 14y ago
      Or use more than one pinhole... E.g. http://pinhole.stanford.edu/zp.html http://pinhole.stanford.edu/zp.html
JoshTko 14y ago
Couldn't you just put one of these lenses in front of every single CCD subpixel to create a flat lens?
- yread 14y ago
  The article says they have to be tuned to a specific frequency so it wouldn't work for CCDs. But I think that it could be used for projectors. They probably can't zoom though
- jws 14y ago
  You would end up with very little light gathering capacity. The number of photons exposing each pixel would be only the ones emitted from the source that managed to strike the pixel directly. The same as a pinhole camera. The thing about a lens it is lets you grab a relatively huge area of light and sort it all out so the right bits go to the right pixels. I already did arithmetic once today (erroneously as it has transpired), so I'll pass on this, but although a metric dumptruck load of photons fly off ov things, when you are 10 meters away, and have an image element the size of a CMOS pixel, there aren't a lot of photons per second to let you sort out 256 levels of intensity without random noise.
wamatt 14y ago
One word: Cellphones
danvk 14y ago
Will this be useful for Astronomy?

128 points by co_pl_te 14y ago

cs702 14y ago

As an amateur photographer with a small collection of DSLR and MILC camera lenses at home, I'm quite excited by this: "In the future we can potentially replace all the bulk components in the majority of optical systems with just flat surfaces." In other words, this will allow large-sensor cameras to get a lot smaller and thinner. Edit: WRONG! As sbierwagen points out[1], even though the press release states that the technology is "completely scalable" [to different wavelenghts of light], it actually works only with wavelengths greater than that of visible light. Too bad :-( sbierwagen: Thanks for the clarification. [1] http://news.ycombinator.com/item?id=4433970 http://news.ycombinator.com/item?id=4433970

djmdjm 14y ago
Sorry, read sbierwagen's comment for the bad news: metamaterials like those used to construct this lens are tuned for very specific wavelengths so unless your photographic aesthetic is centered on a very particular colour of IR then you will be out of luck for quite a while.
- sbierwagen 14y ago
  Metamaterial lenses, if you can make them in visible-light frequencies, might be useful in the microlens array for light-field cameras, where they could act as both the bayer color filter pattern and as focusing element, and where their excellent optical qualities would be really helpful. (The Lytro light-field camera's lousy optical performance is mostly due to the microlens array: they have to be made of conventional materials, and the only lenses that would fit are singlets, which have chromatic abbertation and barrel distortion out the ass.) But as the primary light gathering element, no.

mrich 14y ago

As a myopic geek I am looking forward to ultra flat glasses :)

sbierwagen 14y ago

Operating at telecom wavelengths (i.e., the range commonly used in fiber-optic communications), the new device is completely scalable, from near-infrared to terahertz wavelengths, and simple to manufacture. Ho ho, very clever use of words there. Terahertz is far-far infrared, (1mm to .1mm (1,000,000-100,000nm) wavelength) while NIR is around 1,000nm. Visible light is 740-380nm. To the best of my knowledge, (this isn't my field) nobody has ever made a visible-light metamaterial. The antennas are too small. There's no real physical reason we can't make red-light metamaterials, the problem is just fabrication. (A parallel can be drawn here between metamaterials, desktop nanofactories, and fusion reactors) Additionally, it would be tricky to make a panchromatic metamaterial lens: metamaterials are tuned very precisely to one frequency, (color) constrained by the physical size of the antennas, and they can't transmit any others.

soup10 14y ago
I don't get why every science article gets negative comments like this. Just because something doesn't have immediate practical applications doesn't mean its not interesting research. Of course it's annoying when they are overly optimistic(either because of journalists trying to make the finding interesting, or researchers trying to get more funding), but that's to be expected.
- DenisM 14y ago
  Where headline is not accurate, it bears pointing out. I had no idea that this is not in fact a visible light lense.
  - jasonwatkinspdx 14y ago
    The headline is accurate (unless hacker-news had one that has one that has now been edited away). If you had read the article you would know precisely what wavelengths the device is applicable to (3rd paragraph even), and why it's an exciting research development.
    - 14y ago

blacksmythe 14y ago

From the summary article, it looks like they have made a Fresnel lens, which can also be quite flat but works over a limited wavelength range of light. In order to achieve broadband operation, the phase change vs wavelength of the metamaterial needs to approximate a large delay. If a large delay is not feasible to implement, this summary is overstating the importance of this result compared to prior technology.

co_pl_te 14y ago

Hi, sbierwagen. This is certainly beyond my field of expertise, but I don't see how the inability of this new lens to focus in the range of visible light detracts from the main point of it being a distortion-free lens whose "focusing power approaches the ultimate physical limit set by the laws of diffraction." Although I agree that the press release makes it appear as if this will have direct implications on imaging in the lay sense (this isn't going to make a future iPhone the perfect camera), there are certainly applications of such technology that aren't consumer-focused and don't require visible light that stand to benefit from this research. As blacksmythe notes, the lens created is not your stereotypical refractive lens, but is more along the lines of a Fresnel lens or zone plate. I'd be interested in hearing more about the differences between the two, especially with regard to the limitations of each. Would anyone who read the write up in Nano Letters or has a better understanding care to comment more on the implications? I agree the press release is a tad sensationalistic and possibly (intentionally) misleading — but what do you expect from a press release? Although the findings presented may not be as earth-shattering as a mass-produceable visible-light metamaterial, I still think it's a noteworthy development that shouldn't be dismissed so easily.

ehsanu1 14y ago
OT/Meta: Please use the reply link to post replies, so that coherent threads can be maintained.

th0ma5 14y ago

Anyone play with a pinhole on a CCD or CMOS sensor, or otherwise "no" lens? For a non-lens-distorted picture?

vladoh 14y ago
As far as I know the problem is that the amount of light that goes trough the hole is very low and you cannot see much or you need a very strong light source.
- Sharlin 14y ago
  Or you have to expose for a long time.
  - Luc 14y ago
    Or use more than one pinhole... E.g. http://pinhole.stanford.edu/zp.html http://pinhole.stanford.edu/zp.html

JoshTko 14y ago

Couldn't you just put one of these lenses in front of every single CCD subpixel to create a flat lens?

yread 14y ago
The article says they have to be tuned to a specific frequency so it wouldn't work for CCDs. But I think that it could be used for projectors. They probably can't zoom though
jws 14y ago
You would end up with very little light gathering capacity. The number of photons exposing each pixel would be only the ones emitted from the source that managed to strike the pixel directly. The same as a pinhole camera. The thing about a lens it is lets you grab a relatively huge area of light and sort it all out so the right bits go to the right pixels. I already did arithmetic once today (erroneously as it has transpired), so I'll pass on this, but although a metric dumptruck load of photons fly off ov things, when you are 10 meters away, and have an image element the size of a CMOS pixel, there aren't a lot of photons per second to let you sort out 256 levels of intensity without random noise.

wamatt 14y ago

One word: Cellphones

danvk 14y ago

Will this be useful for Astronomy?