Sammy can try , but I don't think it will be a good idea.
P.S thank you thinness race ! you make smartphones "better".
Bigger sensor + Britecell tech (isocell 2.0) + good aperture + OIS = over 90% on dxomark and famed best smartphone camera on the planet.
Smart WDR looks horrible. Apparently no one at Samsung head of overexposure. The image is too bright and washed out. Where "Conventional" has good contrast and proper colors.
Eske Rahn, 18 Nov 2015A little more in detail. Traditionally at each green cell with a value G, you do an extrapola... moreThis was exactly what I was about to post. (R+B) - W = G. But that's a bad idea, it's only good in paper. What happens if one channel is clipping? What if the color white takes the value of 255 but in reality it's let's say 298 and the color blue and red don't clip? When you will do the math it will result into a different color, which will give the visual result of a Green or Magenta tinted area. I don't like the idea.
Eske Rahn, 18 Nov 2015A little more in detail.
Traditionally at each green cell with a value G, you do an extrapola... moreThanks guys, this is what I figured.
Still I (as a consumer) will need to proof this is indeed better than traditional "photon" capture.
Colors ARE made from RGB, so any calculation instead of direct input of those basic colors, makes me sceptical.
Eske Rahn, 18 Nov 2015The Idea is that W=R+G+B, and if you can estimate R and B from neighbouring cells, you will ge... moreA little more in detail.
Traditionally at each green cell with a value G, you do an extrapolation from the neighbouring cells, to get a R and a B value. (Typically a mean of the closest ones), This way you end up with RGN=(R,G,B)
With this approach you calculate the G from the W and then have RGB=(R,W-R-B,B)
Of course it is a bit more complicated, since the filters are not ideal, so you will need some weight-factors and not just 1.0* on each of them.
NULUSIOS, 18 Nov 2015Can someone explain how white CCD pixels will work? The will collect unfiltered brightness "d... moreThe Idea is that W=R+G+B, and if you can estimate R and B from neighbouring cells, you will get G=W-R-B
Can someone explain how white CCD pixels will work?
The will collect unfiltered brightness "data" and combined with R and B pixels will the G be "calculated"?
I don't get it.
Anybody can (really) explain how this works?
well, shapness is important, but i think 20 mp is ok, in my opinion it will be better to improve the optical image stabilization and the low light sensor.
Anonymous, 18 Nov 2015LED flash < Xenon. Not really any more, with the latest improvements in LED technology and natural color LEDs.
At least not for xenon of the size that smartphones have (the few that do).
xTREAMER, 18 Nov 2015is all blablabla from sammy, they do the ISOCELL cameras but dont use it on theyr devices, why... moreyou are wrong, Carl Zeiss do the best, the nokia 808 is really awesome in all aspects, its better that all samsungs and sonys phones (taking about camera of course).
Apple logic. They will use that technology for their own design benefit. Sure they pay for it but nobody cares. As long as it is from Apple it will be good. Like Apple A9 and DDR4 RAM.
At least they improved ISOCELL to BRITECELL. This might be better than Sony sensor but who knows. We need to see real life performance first.
xTREAMER, 18 Nov 2015if they are as good as Sony why sammy dont use it? even in theyr low end devices?! LOL! Stupid... moreDo u know note 4, 5, s6 series has either sony imx 240 or samsung isocell? Sammy used two different sensor cause of shortage of imx 240 and both r eqally good. Know first then coment
Redmi Note 2 used the technology already. Samsung u can get out from the technology world. Copy other phone.... samsung lack research and development team.... no idea why a big company cannot invented something better....