Welcome! These forums will be deactivated by the end of this year. The conversation continues in a new morph over on Discord! Please join us there for a more active conversation and the occasional opportunity to ask developers questions directly! Go to the PS+ Discord Server.

What do flat lens cameras look like?

9 posts / 0 new
Last post
Grim G Grim G's picture
What do flat lens cameras look like?
So, according to the game, cameras use flat lens technology. I researched this and it seems pretty legit, but I have yet to find a picture of a flat lens that isn't magnified to the microscopic level. What exactly do they look like? Do they have to be in a specific color to catch the light properly? Can lens spotters actually reflect off them?
Trappedinwikipedia Trappedinwikipedia's picture
This is the best picture I
This is the best picture I can find of a modern microlens array. It's quite small, so the whole camera is a postage stamp (potentially a large one) of fuzzy looking glass. http://home.eps.hw.ac.uk/~ryf1/T27_files/image004.jpg
Grim G Grim G's picture
Trappedinwikipedia wrote:This
Trappedinwikipedia wrote:
This is the best picture I can find of a modern microlens array. It's quite small, so the whole camera is a postage stamp (potentially a large one) of fuzzy looking glass. http://home.eps.hw.ac.uk/~ryf1/T27_files/image004.jpg
You think light can reflect off of something like that?
ScorpionOneNiner ScorpionOneNiner's picture
Article, and link to the
Article, and link to the specific image. This looks to be what EP uses. http://www.bbc.com/news/science-environment-36438686 http://ichef-1.bbci.co.uk/news/624/cpsprodpb/E1B0/production/_89867775_i... But, there are other ways of shaving that cat and I'm not convinced the EP countermeasures would affect them. (I know just enough optics to be dangerous) In general, quantity is better than quality, and with enough 'nano' and 'software' all things are possible. Look at a cell phone camera, especially the front (you-facing) one. A couple mm diameter. But that's for a bunch of pixels. One pixel only takes 0.05 mm (based on 500 ppi). That is 0.05 mm with current, non-future-tech cameras. One pixel is hard to notice, so spread a couple thousand pixels over an entire wall. Array based telescopes is sorta this. Or, take an actual cell phone camera, then use nanites to build custom, real-time tailored Fresnel lenses. That gives you a flat lens. Probably can only focus on a small area at a time, but you can move that point around by adjusting the segments. Actually, this is how those flat lens cameras in the BBC article work. Their 'trick' is using metamaterial to get a fresnel-like effect on something very very thin. And then there is focus stacking, and other software algorithms. Basically, the 'camera' just takes in all light with no lens, you get a blurry mess, but with enough pixels & computer horsepower you can bruteforce to recreate images at whatever focus distance. https://en.wikipedia.org/wiki/Focus_stacking So, how does a Dazzler (or other covert thingy that didn't turn up right away when I searched) detect these? And how do they blind? My best guess would be: shoot a laser at a white wall, anything that reflects differently than the wall is a camera. Which would have reliability: none. When was the last time your wall was perfectly consistent, no markings or anything? And if it is a lack of reflection, well human eyes also absorb light. So much for 'only camera or sensor systems'. My guess at how a countermeasure COULD work? Combine the laser disco ball with your muse. Either have the laser not fire when it would point at your eyes, and/or sync your cameras with the laser so camera turns off while laser fires. This can't detect cameras, but would definitely blind them. Random technical info I found from google. Pdf of a PhD dissertation, very long. Might not be relevant. http://graphics.stanford.edu/~wilburn/wilburn_thesis.pdf
How can you challenge a perfect immortal machine?
Grim G Grim G's picture
ScorpionOneNiner wrote:
ScorpionOneNiner wrote:
So, how does a Dazzler (or other covert thingy that didn't turn up right away when I searched) detect these? And how do they blind? My best guess would be: shoot a laser at a white wall, anything that reflects differently than the wall is a camera. Which would have reliability: none. When was the last time your wall was perfectly consistent, no markings or anything? And if it is a lack of reflection, well human eyes also absorb light. So much for 'only camera or sensor systems'. http://graphics.stanford.edu/~wilburn/wilburn_thesis.pdf
Well considering that nanofabbing is a thing, I'd say pretty consistent. Potentially. Other than that, there's the fact that the cameras are probably the most consistent thing on a wall, so you could look for that. But other than that, what are you saying? Light can't reflect off of them?
ScorpionOneNiner ScorpionOneNiner's picture
Grim G wrote:Well considering
Grim G wrote:
Well considering that nanofabbing is a thing, I'd say pretty consistent. Potentially. Other than that, there's the fact that the cameras are probably the most consistent thing on a wall, so you could look for that. But other than that, what are you saying? Light can't reflect off of them?
One non-EP story I read had cleaner nanites built into the wall, so there is that. Some or most light will not reflect off the cameras. Photons enter a camera and end up absorbed into film, or onto a light detector (CCD for modern cameras). (oh, also if light didn't get lost in the camera then a dazzling laser wouldnt get in there either) Only a small percentage of total light though, since the camera is so small. So what a detector would look for is places that are slightly darker than their surroundings. While telling the difference between 'dark camera spot', 'multicolor pattern painted on the wall' and 'person standing in the room being darker than the wall'. Might be easier if it could also look for electricity use, and data flow between the camera and the image processor. But then you want to find cameras, and ignore all the other electronic 'stuff' going on. In the end, the easiest way is maybe to hack the hab network and download a list of all camera locations. Which misses any not registered with the hab. Hmm. A Sherlock AI? Calculate the most likely locations for a hidden camera, given room dimensions, locations of objects, historical info on the hab residents......
How can you challenge a perfect immortal machine?
Grim G Grim G's picture
ScorpionOneNiner wrote:Grim G
ScorpionOneNiner wrote:
Grim G wrote:
Well considering that nanofabbing is a thing, I'd say pretty consistent. Potentially. Other than that, there's the fact that the cameras are probably the most consistent thing on a wall, so you could look for that. But other than that, what are you saying? Light can't reflect off of them?
One non-EP story I read had cleaner nanites built into the wall, so there is that. Some or most light will not reflect off the cameras. Photons enter a camera and end up absorbed into film, or onto a light detector (CCD for modern cameras). (oh, also if light didn't get lost in the camera then a dazzling laser wouldnt get in there either) Only a small percentage of total light though, since the camera is so small. So what a detector would look for is places that are slightly darker than their surroundings. While telling the difference between 'dark camera spot', 'multicolor pattern painted on the wall' and 'person standing in the room being darker than the wall'. Might be easier if it could also look for electricity use, and data flow between the camera and the image processor. But then you want to find cameras, and ignore all the other electronic 'stuff' going on. In the end, the easiest way is maybe to hack the hab network and download a list of all camera locations. Which misses any not registered with the hab. Hmm. A Sherlock AI? Calculate the most likely locations for a hidden camera, given room dimensions, locations of objects, historical info on the hab residents......
Technically most camera spimes are accessible to the public. Then again only the private ones would need a lens spotters. That being said I don't think hacking into the network for private sensor spimes is really necessary, can't you just look for their stealthed signal?
ScorpionOneNiner ScorpionOneNiner's picture
the future is now
http://spectrum.ieee.org/consumer-electronics/gadgets/inside-the-develop... Speaking of teeny tiny cameras, camera arrays, image combining software, depth of focus, etc etc. News in my IEEE today; this is basically a handheld version of all that. Just add nano magic and spread the cameras over a wall. Re stealthed signal, dunno. That's outside my area of expertise. :) Probably depends on GM's ruling, and an approach more akin to hacking than anything optical?
How can you challenge a perfect immortal machine?
Trappedinwikipedia Trappedinwikipedia's picture
If they're not already public
If they're not already public, encrypting the feed is trivial, so even if you can spot the stealth signal, it looks like garbage without the right access.