New sensor doubles brightness

[jebrady03]

Apologies if this has been posted already. Thoughts? http://www.diginfo.tv/v/13-0021-r-en.php

 

[nightbreath]

There's no way to tell if the technology really worth our attention (lots of processing power required, unknown drawbacks of the approach, etc.) We'll need to look at real world results when they appear

 

[mb66energy]

Apologies if this has been posted already. Thoughts? http://www.diginfo.tv/v/13-0021-r-en.php

It's a very good idea to split the light spectrum by interference filters (that's what they do IMO) into two parts without substantial loss of light - compared to RGB filter arrays.

It is a little bit like the FOVEON sensor principle which evaluates spectral components of light in different depths of the silicon - so there is in principal no loss of photons. But I think that you have some absorption of photons in FOVEON sensors because I see similar behaviour of noise for high iso settings at dpreview ( http://www.dpreview.com/reviews/sigmasd1/19 ).

Perhaps it is possible to build 3D chip structures where the light is split by interference filters for ONE SQUARE PIXEL, focused by lenses on three or four different photosites which lay behind the filter system - 1 full color pixel from 3 or 4 spectral ranges with zero light loss ... even 20 full color MPixels on a full frame sensor will provide spectacular IQ. If it costs 20 seconds to "develop" an image - it would be worth it. A preview of each 4th pixel (in each dimension) would result in 1.5 seconds development time and should be sufficient for fast previews.

But let's wait 5 (optimum) or 10 years (most probable) to see such sensors in sub 3 KiloEuro/-Dollar cameras ...

 

[jebrady03]

I simply find it very encouraging that companies are continuing to develop new processes to capture light. It's funny, so many people say things along the lines of "this will be the last camera I'll buy" (I saw that in reference to the 5d3 and 1Dx a LOT) because they're so impressed with the camera. But as always... "you don't know what you don't know" and if this technology, or even something more impressive, is available in 5 years, who wouldn't WANT to upgrade to it if it works as expected? I know I would - regardless of how happy I was with my current camera.

Imagine this in a camera without an anti-aliasing filter (because software corrects it) along with eye-control focus, etc. and everything else on people's wish list!!! SUPER BRIGHT, SUPER SHARP, CORRECTLY FOCUSED pics! Of course, DxO would kill all the enthusiasm because they'd say it was the worst sensor since the original Rebel...

 

[RGF]

There's no way to tell if the technology really worth our attention (lots of processing power required, unknown drawbacks of the approach, etc.) We'll need to look at real world results when they appear

Wait a few years. processing power increases, new algorithms develop, ...

Could be a major improvement - worthy of watching

 

[mb66energy]

[...]
It's funny, so many people say things along the lines of "this will be the last camera I'll buy" (I saw that in reference to the 5d3 and 1Dx a LOT) because they're so impressed with the camera. But as always... "you don't know what you don't know" and if this technology, or even something more impressive, is available in 5 years, who wouldn't WANT to upgrade to it if it works as expected? I know I would - regardless of how happy I was with my current camera.
[...]

Good point! - In the last days I reproduced some of my old slides. The Kodachromes are amazing viewed by a self built reproducing unit: The Visible DR is whatever kodachrome delivers - around 2000:1. What I see on the display of my 40D (it isn't the worst TFT) has perhaps 200:1 and the images suffer from that.

If I look my images with my beamer, a simple DLP model from NEC, 1024x768 pixels, they look great despite of the low resolution: A contrast of 2000:1 matches the contrast range of the original scene much better then all LCD-/TFT-Displays including very good IPS and PVA-Displays from EIZO.

Back to the slides: The depth of these slides is incredible so I am shure that there is a lot of headroom for digital development going closer and closer to recording reality perfectly. It is not the sole process of photography but it is always better to record as much meaningful data as possible and select it later!

 

[Mt Spokane Photography]

I think it was discussed so long ago that many don't remember

http://www.dpreview.com/news/2013/02/04/panasonic-promises-high-sensitivity-sensors-using-micro-color-splitters