Samsung's MEMS shutter could massively improve high megapixel cameraphones
By Darren Murph 
posted May 15th 2009 7:05PM
posted May 15th 2009 7:05PM
posted May 15th 2009 7:05PM
Now that we've thrown 'em off the trail, use the form below to get in touch with the people at Engadget. Please fill in all of the required fields because they're required.
Nokia phones have mechanical shutters.
Yeah, that's what makes the pictures look decent, unlike some wannabe smartphones.
Wow, a 453 megapixel camera phone. Must be some serious high image quality.
i thought 12 megapixels was enough for most applications....
And this is why you should write your representative and get them to get more funding to your local synchrotron facility. You get more cool shit like this.
453 megapixel Camera Phone will still look crap compare to DSLR.
Camera phones will catch up. DSLRs and the like will always be more advanced but it'd be foolish to think in the future we won't have some fantastic quality cameras in our phone. Different lenses might not be possible but it should be interesting times indeed.
Assuming you carry a cellphone.
is that a joke image or is it just coincidence that it looks like an eyeball?
Swon, First thing i thought when i read this was the same damn thing.
it's true, Samsung is the new Sony...
time has come to change reigns...
I'm slightly confused here, to me that looks a hell of a lot like an aperture and not a shutter. If this is the case it will not help avoid blurry photos. What is needed to avoid blurryness is a faster shutter speed, but current speed are fast enough, they are just unusable on phones due to poor lighting and sensor sensitivity. As the pixel density of the sensor increases the light sensitivity decreases (one reason why I love my ol' Canon G5, 5mp on a 1/1.8" sensor makes for pretty good sensitivity).
Basically there are three options:
1. Increase the aperture of the lens
2. Increase the lighting (better flashes)
3. Decrease pixel density (either bigger sensor, less pixels or both)
It's important to point out that increasing the aperture of the lens will be limited by the fact that most cameraphones are fixed focus. You can only get away with fixed focus as long as the aperture is small enough to yield a good depth of field. Open up the aperture far enough and anything that isn't at a specific distance range will look blurry.
Indeed, that options is more for p&s cameras that seem to be increasing in megapixel count yet reducing in size. With phones decreasing the pixel density is the only viable option.
I've jsut re-read the read link and it turns out this is a shutter (I was going more on the picture which give the impression the 'curl' of the blades would be adjustable). Apart from the fact that this type of shutter requires more power and power to stay closed, I don't see why this would be any better or faster than a traditional shutter (not that we need anything quicker, my before mentioned G5 goes well over 1/1000 of a second and that will require a flash gun, something you can't really stick on a phone).
Even though current SLR's produce higher quality images than compacts, their mechanical shutter design is fatally flawed when used with high density image sensors. The problem is that the physical movement of the shutter generates vibrations which undesirably causes blur, the latest research has shown that SLR technology has almost reached it's maximum pixel count as any further increase results in no improvement due to blur being caused by the shutter vibrations, testing was completed in strict conditions with the camera body fully secured.
The current solution is to either remove the physical shutter or reduce the pixel density, such as in full frame camera's where bigger pixels means less noticeable blur caused by vibrations, not a solution for camera phones! Alternatively, new shutter technology is required, such as MEMS as mentioned in the article.
@Richard - The idea of vibrations never crossed my mind, that is a possible advantage this would have over traditional shutters. However, the reason I think it didn't cross my mind is because it just doesn't matter and that's why it has remained relatively unchanged after so many years. The only time the vibrations would be noticeable are when the sensor has an extremely high resolution, the problem is to take photos at such high resolution on relatively small sensors a lot of light input is required as I have said. For times when this is not appropriate (most times unless you shoot in a studio for example) A decrease in resolution would be preferred as it would give increased sensitivity.
Does Samsung have a headquarters in Japan? Or did the article mean "Over in Korea"?
It's a Japanese tech journal, but since when does Samsung have a design and development center in Japan? I understand that they have design bureaus in Korea, the US and China, but Japan?
They have R&D centers in Japan. They have one in Germany too.
Samsung does have R&D facilities in Japan, Korea, the US and Europe, but NOT in third-worlds like China or Uganda Africa though I was told there is a manufacturing factory in China and Malaysia mainly for their cheap labors.
It still won't help the main problems of cellphone imagery: grainy/a lot of noise, poor color rendition, and a lack of real focus. Megapixels are mostly irrelevant when all of your precious megapixels are filled with noise and odd colors.
Well, the thing is, if I take a poor picture of 5MP and reduce it to fill my screen it actually looks better I found, unlike a poor picture of 300K that I blow up to fill the same screen.
You can process more on more pixels, noise reduction also reduces resolution by averaging neighbouring pixels, but the more pixels you have the better a final reduced size picture looks... sort of, theoretically. :)
Samsung is surpassing its Japanese counterparts and the Japs aren't too happy.
What I do not understand and unable to find is why they do not use LCD shutters. Kinda of what is used a LCD shutter glasses for virtual reality just adopt the technology for camera shutters, it is small, light, no moving parts, super cheap, and single color (black) LCD refresh rates can be extremely fast. I am obviously missing something can someone tell me what?
LCDs aren't anywhere close to 100% opaque. Hence why LCD televisions are incapable of producing true black. It's my understanding that this problem gets worse with faster switching times.
@Plothole
LCD TV's have a red, green, and blue pixels, along with the matrix wiring of each pixel packed very tightly which massively up's the complexity. I am talking about a large mono-pixel glass laminated liquid crystal to be only used as on\off as done in high end 3D glasses. Massively reducing the complexity allows the LCD to be engineered to up's the speed and light block potential. I have not been able to find much research that has been done on it, there seems to be some pattens out there. It just seems like an easier way to produce a low cost shutter for phones, etc.
Exactly how fast is extremely fast? I'm guessing that 3D glasses are around 1/60th of a second. That's a very slow shutter speed in terms of freezing motion. Such a shutter would only bend light coming in. I'm not sure how effectively it could do this in direct sunlight. It seems like a good idea, but I'm guessing that it might not be one that works (at least at the moment), otherwise, you'd think someone would have made one by now.
Hmm... if you assume a 1/1.8" sensor... and if I've figured all the variables correctly... the diagonal of each individual pixel on a 453MP sensor would be significantly shorter than the wavelength of blue light (440–490 nm).
They'll find a way.
Not gonna happen :P You will need a gigantic sensor to get ANY light with that small aperture.
I hope DSLR's become obsolete.
You mean in the sense that they get smaller and need no complex shutter and mirror and become all like micro four thirds? Or do you mean phones would replace them, because I can agree with the first but not the second, you'll always need a way to put some interchangeable decent lenses on a good camera.
In a sense that there is no real discernible boost in quality using a bigger sensor (of course it'll be better...but only under a microscope) and that they become cheap enough for everyone to use. And to stop the know-it-alls from posting their knowledgeable comments.
@superhobo - Sensor size can increase image quality dramatically. As I've said in my previous posts, if you have two 5mp sensors but one is twice as big as the other the bigger one will have half the pixel density and twice the sensitivity, allowing it to gather light more efficiently and so quicker shutter speeds and smaller apertures can be used. If you don't think any of that is important then a DSLR is obviously not right for you, but that doesn't meant here aren't more experienced photographers out there that do think it's important.
I know, but improvements in technology can allow for smaller sensor sizes while keeping the same quality.
Bigger lens will also reduce the chance of any defect or aberration or dustspeck from influencing final outcome though doesn't it? And it will capture more light, that's obvious, regardless of the sensitivity of the sensor, even if sensitivity was 'perfect'
Not if the sensors have black hole technology.
You guys just can't think outside the box :p
Sure you may be able to improve the quality of the sensor, making the pixels more sensitive to a certain extent, but then you can also have these super sensitive pixels on a larger sensor. There is absolutely no need to decrease the size of image sensors, if you need the camera to be so small that at least a 1/1.8" sensor won't fit then you shouldn't be taking photos.
Why would you want to use a bigger sensor if the smaller ones get so good that you can't notice the difference?
(Bear in mind we're talking about future technology)
Because a bigger sensor will always be better. A small sensor will
never be equal to a bigger sensor of the same resolution, that's jsut
the way it works.
@superhobo: Sure, we will happily use smaller sensors if they become good enough - but that assumes that smaller sensors improve enormously to match the quality of current DSLRs (maybe, over time, but see below) - and also that you're happy with the quality of DSLRs now. Personally, I'm not happy with the quality of my less-than-35mm-sensor DSLR (I mean, I'm not like complaining, for the time and price it's fine, but in terms of 'would I be happy if development stopped here and they just made sensors smaller instead'). It's fantastic in bright daylight but quality is noticeably worse once you need ISO 800. I'd like at least 3200 [which by the way I think the camera theoretically does, but there are lies, damn lies, and camera max ISO limits] in better quality than that. Sure it's already better than film but so what? I want it *more* better than film. :)
Also, dynamic range sucks (and that's not better than film) - as well as improving accuracy (low-light noise) why can't we have sensors that allow significantly brighter light before they clip. That seems like it should be possible.
So I think we need at least five, ten years more technology improvement before the large sensors get really good enough. Add time on top of that to make smaller ones better...
...and even with all that, last I heard there are also some fundamental limits on sharpness per sensor size due to diffraction or something, which phone cameras have already exceeded. I'm not sure if we are also near the fundamental limits on noise performance too, or if there's still room for improvement there.
All this said I would definitely appreciate a smaller camera - doesn't have to be phone-camera small, but micro 4/3 looks nice... except I don't like the aspect ratio, the sensors are too small (see above), and most importantly I want a proper viewfinder... yes, I realise this is impossible. The Leica rangefinder looks nice and has relatively small lenses but there are numerous problems with it, not least the price. So, for now, DSLRs it is. And my phone camera is convenient for taking pictures of things I want to remember when I get back home, or show people amusing things or whatever, so that's cool too even if the quality is horrible.
Nail in the coffin of the megapixel race? Please?
Why? You might rather have more sensitivity, but the wafers cost a lot and the smaller the sensor the cheaper and that means sensitivity is dependant on coincidental breakthroughs, where meanwhile the manufacturers have to make do and compete in phones with silly megapixel counts, and who cares if they do or not if in the end it doesn't change anything except that it sells the unknowing consumers new phones/cameras which then finance research that hopefully will lead to results that are more useful like increased sensitivity and deeper colorspace.
They'll more than likely put most of their funding into research for more megapixels rather than camera aspects that are actually useful.