Light L16 Camera - What are your thoughts?

Interesting concept altogether! It reminds me of jumping spiders with their collection of severla eyes, from wide angle to telephoto.

The old path of progress towards better image quality and higher detail resolution has so far been bigger sensors and bigger lenses. I wonder if a cluster of small cameras conjoined in this kind of image construction computer could produce images as good as one big traditional sensor and lens? If so, would it be possible for the cluster to be smaller than the big camera?
 
If only we could see the exif data for the camera that took the product images of The Light ... £10 says it's a nikon or canon ... :-)
 
This is about the new Light L16 camera. The company's website is here, there's an unbelievably fatuous promo video here, and there's a remarkably shallow interview with one of the founders here.
 
chris malcolm said:
The old path of progress towards better image quality and higher detail resolution has so far been bigger sensors and bigger lenses. I wonder if a cluster of small cameras conjoined in this kind of image construction computer could produce images as good as one big traditional sensor and lens? If so, would it be possible for the cluster to be smaller than the big camera?
Click to expand...
Yes. There will be money poured into this line of development as it fits into a smartphone.

Once they have gathered enough image information, including stereoscopic and multiscopic sets, there are many ways to process it to take full control of quality, exposure, focus and depth of field.

Results of multi camera technology will definitely be a dominant competitor to the results of large sensors, as the technology matures.
 
Last edited:
I did search and couldn't find anything else posted about this

I came across this on a Facebook site, the Light L16 camera, certainly a different take on how to design/build a compact camera, not sure about the claims DSLR quality photos from a compact though, but an interesting concept nonetheless

http://spot.light.co/light-l16-camera-launch/
 
Last edited:
ianp5a said:
Yes. There will be money poured into this line of development as it fits into a smartphone.

Once they have gathered enough image information, including stereoscopic and multiscopic sets, there are many ways to process it to take full control of quality, exposure, focus and depth of field.

Results of multi camera technology will definitely be a dominant competitor to the results of large sensors, as the technology matures.
Click to expand...

Agree :)

The question for me is, is there a high level of consumer demand for the improved image quality and increased control this could deliver - given the inevitable cost and increased bulk if added to a smartphone? Smartphones already seem to be plenty good enough for the vast majority, with unrivalled convenience and connectivity. If it doesn't catch on in a big way, it will become 'just another expensive camera'.

I hope it doesn't develop along the same lines as Lytro, veiled with marketing blulcrap (and Nimslo, for those with longer memories). I don't think it will, if only because the underlying science is sound.
 
Weird to think that what we consider photography today could be dead in 10 years thanks to hardware like this.

Oh, and advanced software, of course. I mean, depth of field and bokeh is going to be totally adjustable in post-production, isn't it?
 
I'm sure all sorts of things will be possible with the right information. Image stabilisation, although amazing today, will be even better. Lens aberration calibration too. So you can get a better image from a flawed lens than you can with a traditional sensor with a good lens.
HoppyUK said:
The question for me is, is there a high level of consumer demand for the improved image quality and increased control this could deliver -
Click to expand...
I remember the futuristic Nimslo. 3D is the future!
Smartphone makers are always looking for ways to differentiate their phones. And innovative camera manufacturers are trying to reduce weight and size of both cameras and lenses. I suspect those with the software talent will make most headway.
 
Last edited:
Looks interesting, but at $1600, I think I'll give it a miss.
 
Yeah, 16 lenses and 16 sensors doesn't come cheap, can't see it catching on at that sort of price though
 
furq said:
Weird to think that what we consider photography today could be dead in 10 years thanks to hardware like this.
Click to expand...
So long as you don't want to take photos when the sun is out! Did you notice that, in the promo video, not a single person was using the camera in bright light? There's a fundamental problem with using a big glass screen as a viewfinder and control panel....
 
Yes. But if it takes off, EVFs and better controls can be added to later models.
 
Just looked at this, idea seems fine, I would imagine it has phone sized sensors behind the lenses, would probably make a great P&S if the price was sensible.
 
ianp5a said:
Yes. There will be money poured into this line of development as it fits into a smartphone.

Once they have gathered enough image information, including stereoscopic and multiscopic sets, there are many ways to process it to take full control of quality, exposure, focus and depth of field.

Results of multi camera technology will definitely be a dominant competitor to the results of large sensors, as the technology matures.
Click to expand...
There was one article I read that said they wouldn't actually apply this to a smartphone due to the complexity of and it was still a very long way off (the summer 2016 date being very optimistic)?
 
Last edited:
I was referring to the future of this type of technology. It's a cheap way to gather a lot of image and multidimensional data.
 
ianp5a said:
I was referring to the future of this type of technology. It's a cheap way to gather a lot of image and multidimensional data.
Click to expand...
It's not cheap and won't be for quite a while :)

That's part of the issue, it's very expensive tech, and it's difficult to package.

It will be awesome but expensive, with the basic unit alone costing nearly £2k.

They could try and make it smaller so phone fabrication and packaging is possible (it would be very thick as a phone apparently) and the will get cheaper (but not for quite a long time yet), but then we're back to square one - a not as good camera for phones due to the inevitable extra compromises. Also, as it stands, the file sizes alone would kill off Apple's memory restrictions!

I hope they don't go down the phone route and just concentrate on making a great camera (though the purist in me still thinks it's cheating somewhat!).
 
Last edited:
There are no exotic components required for this. Small sensors are mass produced and much cheaper than the large ones in the cameras we use. And as the lenses are very small it fits into a small camera body easily. And with further development, into a smartphone package.
It's this potential that will attract further investment to them, or a rival. Where all the kinks can be ironed out and Pro and Mini variants will be considered. The clever bit is all done in software. For the image quality it can already achieve its not too expensive. But that will only get better.
 
Last edited:
ianp5a said:
There are no exotic components required for this. Small sensors are mass produced and much cheaper than the large ones in the cameras we use. And as the lenses are very small it fits into a small camera body easily. And with further development, into a smartphone package.
It's this potential that will attract further investment to them, or a rival. Where all the kinks can be ironed out and Pro and Mini variants will be considered. The clever bit is all done in software. For the image quality it can already achieve its not too expensive. But that will only get better.
Click to expand...

Do you see this as a standalone camera? I think it'll struggle to make much headway there.

At the end of the day, it makes a bigger sensor out of lots of smaller ones, but the USP is it does that with a very slim profile - small enough to fit into a smartphone, though probably with less lens/sensor units but still with a very handy jump in image quality.

It takes more than 50 typical 1/3" smartphone sensors to match the area of full-frame, and it is physical sensor area that ultimately defines image quality. It's about light-gathering and lens performance, not how many pixels. IMHO if this device is a camera in its own right, instead of being incorporated into a smartphone, then it looses all its appeal.
 
It's already a standalone camera. And if this technology does a better job than traditional sensor technologies, without some of today's physical limitations, it can replace the current standalone camera sensors.

But, as I wrote, it'll be very interesting for a smartphone manufacturer. They want to make their phone stand out from the rest. So I suspect a smartphone manufacturer, rather than a camera manufacturer will be first to take a close interest.
 
Last edited:
ianp5a said:
But, as I wrote, it'll be very interesting for a smartphone manufacturer. They want to make their phone stand out from the rest. So I suspect a smartphone manufacturer, rather than a camera manufacturer will be first to take a close interest.
Click to expand...
I'm sorry, but I don't see how you can pack multiple lenses and sensors into a smartphone. Unless it's a very big smartphone of course, but most people don't want them to be very big.
 
Last edited:
If they can fit one 20MP sensor and lens in a normal sized smartphone like they do today, they can put another one next to it. And so on.
The current device shown is not much bigger than a smartphone. And that yeilds 53Mp. I'd be quite happy to slum it with a 20Mp version in my phone. I'd probably get my fingers in the way of half of the lenses though.
 
Last edited:
Couldn't seem to find any real specs - what ISO can it handle? Shutter speeds? Actual aperture? They say f/1.2 but I guess they just mean "can produce depth of field similar to using an f1.2 lens". Would love to actually try one out!
 
StewartR said:
I'm sorry, but I don't see how you can pack multiple lenses and sensors into a smartphone. Unless it's a very big smartphone of course, but most people don't want them to be very big.
Click to expand...
That's what to was trying to say the point of which I think was missed.

They have already stated they are not looking at the smartphone route due to its complexity and fabrication issues - with the phone part, and extra large battery it would be very thick. Certainly a lot bigger than capable enthusiast compacts which could probably do just as good a job!
 
ianp5a said:
If they can fit one 20MP sensor and lens in a normal sized smartphone like they do today, they can put another one next to it. And so on.
The current device shown is not much bigger than a smartphone. And that yeilds 53Mp. I'd be quite happy to slum it with a 20Mp version in my phone. I'd probably get my fingers in the way of half of the lenses though.
Click to expand...
It's not that simple.

Once they start cutting back on lenses and sensors to make it part of a useable telephone and portable PC, it doesn't have the headline grabbing specs and would disappear into the already burdened smartphone category.

It really wouldn't be that special.
 
Last edited:
I'm referring to the potential of this type of technology. And not some particular device that we have not yet seen the specs for.

If version 1 has 53mp and all that they promise, just think of what can be done when they've proved it in the market, and big investments are made to take it beyond version 1.
 
Last edited:
ianp5a said:
If version 1 has 53mp and all that they promise
Click to expand...
If......

If the lens(es) really can resolve 52mp. If they really can get DSLR-type low-light performance from a bunch of smartphone sensors. If all the clever software works fast enough to make the camera usable.

Don't get me wrong, I think it's an intriguing concept, and it would be great to see a newcomer shake up the industry. But they're making some big claims. For me, the complete absence of any real specifications, plus the complete absence of any 52mp sample images on their website, suggests that they still have very significant engineering challenges to overcome.
 
Yes that device may be sh!t, expensive and slow. I'm not going to get one. But I am interested in getting high quality results in a small device. From them, or whoever has the best technology. It'll be fun to see people dump their big heavy cameras.
Luddite: "Well that's not going to happen for a long time"
It seems that this is largely software driven. Where iteration cycles are short. So it'll be interesting to see how fast this develops.
 
Last edited:
My enthusiasm, as I wrote above, is really just for the concept - using multiple small sensors and lens units to create a larger format camera. It would have one major advantage (and only one as far as I can see) and that is a very slim form factor, to fit easily into a pocket or handbag. That is a huge advantage denied to current designs.

It would be slim enough to fit into a smartphone, and that's where I think the big opportunity lies - though most likely as a scaled down version with maybe four, six or possibly eight lenses. As a standalone camera, that you would have to carry separately, I think it loses 90% of its appeal and immediately puts itself in competition with conventional cameras where it'll have a hard time.

BTW, this is not a new idea, and I'd be surprised if there isn't a whole bunch of patents to be navigated. I'm assuming the timing has something to do with technologies that have now evolved to a point where it becomes both feasible and affordable. Or if I was to be more cynical, it could just be a bunch of chancers looking make a fast buck. Not that I can see much evidence of that, but it's a very light weight marketing pitch with a lot of obvious questions unanswered. Anybody putting a deposit down now is taking a jump into the unknown.
 
ianp5a said:
It seems that this is largely software driven.
Click to expand...
That's part of what I don't quite get.

At the end of the day a digital camera is just a device for capturing photons, working out where they've come from, and counting them. There are only a finite number of photons, and existing cameras count them pretty well: modern DSLRs typically have a quantum efficiency around 50%. Improving that requires better hardware, not better software. Similarly, working out where the photons have come from requires a lens which delivers them accurately to the right portion of the sensor, regardless of the angle of incidence or the wavelength of the light: again, that's a hardware issue not a software one.

Or maybe the guys who are behind this are brighter and more creative than I am. That's possible.

ianp5a said:
It'll be interesting to see how fast this develops.
Click to expand...
Yes it will.
 
If you sample the same spot with multiple sensors you can interpolate the values to get a good quality result with a high dynamic range.
If you can calibrate the lens by comparing multiple images of the same calibration target, lens weaknesses can be corrected in software.
 
Last edited:
ianp5a said:
Yes that device may be sh!t, expensive and slow.
Click to expand...

Most things are when they are new. In 2003 I paid £1,300 for a Nikon D100 with a mere 6M pixels. My 24MP D3200 cost a few hundred and is much better.


Steve.
 
Steve Smith said:
Most things are when they are new. In 2003 I paid £1,300 for a Nikon D100 with a mere 6M pixels. My 24MP D3200 cost a few hundred and is much better.


Steve.
Click to expand...
The improvements go far further than MP though, of course :)
 
ianp5a said:
If you sample the same spot with multiple sensors you can interpolate the values to get a good quality result with a high dynamic range.
If you can calibrate the lens by comparing multiple images of the same calibration target, lens weaknesses can be corrected in software.
Click to expand...

This is true, but you can't defeat the laws of physics!! Small sensors have very small pixels, so have limited light gathering ability and consequently limits the dynamic range. Cheap optics, will always be cheap optics, there are no shortcuts here. These two things are fundamentally why there is a difference between mobile phone cameras, compact cameras, DSLRs, etc.

Software can be very clever, but it can only interpolate, ie make a best guess based on input data.

That said its an interesting concept, and yes I can see future generations being quite clever and sophisticated.
 
Mr Perceptive said:
This is true, but you can't defeat the laws of physics!! Small sensors have very small pixels, so have limited light gathering ability and consequently limits the dynamic range. Cheap optics, will always be cheap optics, there are no shortcuts here. These two things are fundamentally why there is a difference between mobile phone cameras, compact cameras, DSLRs, etc.

Software can be very clever, but it can only interpolate, ie make a best guess based on input data.

That said its an interesting concept, and yes I can see future generations being quite clever and sophisticated.
Click to expand...

Yes, the laws of physics... The image quality problem with phone cams is simply down to the tiny size of the sensors. Make the sensor bigger, and it collects more light (regardless of the number of pixels) and lenses come to life because the extreme resolution demands are automatically lifted. All aspects of image quality are transformed and that applies just the same to a bunch of smaller sensors joined together as it does to one bigger sensor.

Software can do the joining up bit easily - it's just like stitching a panorama or like one of those mega-panormas, eg http://www.dailymail.co.uk/sciencet...ION-pixel-image-created-David-Breashears.html
 
But you aren't stitching each shot together like a panorama though with this camera, each lens shoots the same scene, the camera decides out of the 16 different lenses which 10 it will use, but it only has lenses of 3 different focal lengths, so i suspect that there will be several lenses of the same FL but with different apertures thus allowing the end user to change the DoF post taking of the picture

Either that, or each of the same FL lenses shoots at a different distance and you end with something akin to focus staking for Macro photography
 
Last edited:
rpsmith79 said:
But you aren't stitching each shot together like a panorama though with this camera, each lens shoots the same scene, the camera decides out of the 16 different lenses which 10 it will use, but it only has lenses of 3 different focal lengths, so i suspect that there will be several lenses of the same FL but with different apertures thus allowing the end user to change the DoF post taking of the picture

Either that, or each of the same FL lenses shoots at a different distance and you end with something akin to focus staking for Macro photography
Click to expand...

You may be right, but that's not how I understood it.

In which case though, the science becomes a whole lot murkier and more software dependent. The problem though, is physics is physics and while the light gathering aspect of larger formats might be replicated by overlaying multiple images of the same thing, you're still asking lenses to resolve miniscule detail. Pixel-width of a typical 10mp smartphone is around 1.1 microns - that's tiny, many, many times smaller than anything in a decent camera, and to get 50-odd mp they're presumably just multiplying that by the number of active units. It doesn't work like that - the more resolution a lens has to deliver, the lower the image contrast becomes, so even if the detail is there (doubtful, even with the very best lenses) it's so faint you can't see it. Smartphones get around that, kind of, by pumping up the contrast, sharpening, saturation and noise reduction to horrendous levels, then compressing everything down to smooth out the mess (and losing even more detail). That might look okay on Facebook, but enthusiast photographers would reject it. Simple fact: sharpness and all aspects of image quality are directly related to format size.
 
You must log in or register to reply here.
Back
Top