rosuna Posted January 22, 2009 Share #1 Posted January 22, 2009 Advertisement (gone after registration) It is likely that Sony presents a A800 camera with a APS-C CMOS back-illuminated sensor during the PMA show. You can read about this new technology here: 1) http://www.photonics.com/Content/ReadArticle.aspx?ArticleID=34685 2) Sony Global - Press Release - Sony develops back-illuminated CMOS image sensor, realizing high picture quality, ne . Link to post Share on other sites More sharing options...
Advertisement Posted January 22, 2009 Posted January 22, 2009 Hi rosuna, Take a look here The new frontier: Back-illuminated sensors. I'm sure you'll find what you were looking for!
kdemas Posted January 22, 2009 Share #2 Posted January 22, 2009 Very interesting technology and a nice description of the benefits and challenges on the Sony site. This may bring about another nice leap forward in technical image quality. Thanks for posting! Kent Link to post Share on other sites More sharing options...
like_no_other Posted January 22, 2009 Share #3 Posted January 22, 2009 No. Small sensors with small photo sites benefit from this technology (-> compact cameras). Put a black sheet of paper on your notebook screen and remove it. Then do the same with your 24' screen. You will see the difference. Same situation as with small and large sensors. Link to post Share on other sites More sharing options...
ho_co Posted January 29, 2009 Share #4 Posted January 29, 2009 ...Put a black sheet of paper on your notebook screen and remove it. Then do the same with your 24' screen. You will see the difference. Same situation as with small and large sensors. lno-- I have neither a notebook screen nor a 24-foot screen (nor even a black sheet of paper, for that matter ), so I can't perform your experiment. What would be the difference between uncovering a smaller screen and uncovering a larger one? Link to post Share on other sites More sharing options...
rosuna Posted January 29, 2009 Author Share #5 Posted January 29, 2009 All CMOS sensors will benefit from this technology, which reduces the differences between CMOS and "full-frame transfer" CCDs. Link to post Share on other sites More sharing options...
adan Posted January 29, 2009 Share #6 Posted January 29, 2009 Actually, the general principles involved have BIG implications for the Leica M, because they directly address two things that are troublesome with the M wideangles when used with a digital sensor. If you look at the Sony link and diagrams, especially the green layer, you will see that light has to travel down a vertical shaft to get to the sensitive spot, like sunlight reaching a street between tall office buildings (the gray circuit blocks), in the traditional sensor. If the light is coming from the side, as in the corners of an image from a 35 cron or 21 Super-Angulon, the "buildings" tend to shade the "street", causing vignetting as one gets further from the center of the lens. Microlenses can correct this, but unless they are offset progressively as one gets closer to the sensor corners, as in the M8, they also project their refocused light more and more onto the sides of the "buildings" rather than straight down the shaft. A backward sensor design like this makes the imaging surface more like film - a flat surface that can see and react to light coming from a wide range of directions, rather than just a "high-noon" overhead light source. Are there technical issues still out there? Sure. Will it work with 6-micron pixels as well as 1.6-micron pixels? Is a real-world chip's imaging surface actually as flat and smooth as Sony's schematic? Does Sony have patents that mean Kodak can't turn around and do this for Leica tomorrow? Can the idea even be adapted to CCD architecture and signal paths? But Leica has said all along that what M lenses required to work well for digital as originally designed - was a significant change in sensor architecture. This is a significant change. Link to post Share on other sites More sharing options...
like_no_other Posted January 29, 2009 Share #7 Posted January 29, 2009 Advertisement (gone after registration) lno-- I have neither a notebook screen nor a 24-foot screen (nor even a black sheet of paper, for that matter ), so I can't perform your experiment. What would be the difference between uncovering a smaller screen and uncovering a larger one? ino? The small experiement should demonstrate the principle if you remove the wires from the front side of the sensor to the back side. The wires cover a much larger fraction of the possible light sensitive area on sensors used in small compact cameras than they do in DSLR's. If you remove them in compact cameras the efficiency of removal regarding light sensitivity is much larger than in DSLR's. The effect in small cameras is about the equivalent of 1 stop. In DSLR's much less. Link to post Share on other sites More sharing options...
rosuna Posted January 30, 2009 Author Share #8 Posted January 30, 2009 Actually, the general principles involved have BIG implications for the Leica M, because they directly address two things that are troublesome with the M wideangles when used with a digital sensor. If you look at the Sony link and diagrams, especially the green layer, you will see that light has to travel down a vertical shaft to get to the sensitive spot, like sunlight reaching a street between tall office buildings (the gray circuit blocks), in the traditional sensor. If the light is coming from the side, as in the corners of an image from a 35 cron or 21 Super-Angulon, the "buildings" tend to shade the "street", causing vignetting as one gets further from the center of the lens. Microlenses can correct this, but unless they are offset progressively as one gets closer to the sensor corners, as in the M8, they also project their refocused light more and more onto the sides of the "buildings" rather than straight down the shaft. A backward sensor design like this makes the imaging surface more like film - a flat surface that can see and react to light coming from a wide range of directions, rather than just a "high-noon" overhead light source. Are there technical issues still out there? Sure. Will it work with 6-micron pixels as well as 1.6-micron pixels? Is a real-world chip's imaging surface actually as flat and smooth as Sony's schematic? Does Sony have patents that mean Kodak can't turn around and do this for Leica tomorrow? Can the idea even be adapted to CCD architecture and signal paths? But Leica has said all along that what M lenses required to work well for digital as originally designed - was a significant change in sensor architecture. This is a significant change. I am not 100% sure but I think CCD full frame sensors have not so much wires and circuits as CMOS have. Back-Illuminated sensors are "bottom-circuits" designs. The circuits are just below the silicon-light sensible layer, instead of on top of it. I think FFT CCDs are different, because there aren't so many circuits to be placed at the bottom. CCDs are more like back-illuminated CMOS sensors... Link to post Share on other sites More sharing options...
rosuna Posted January 30, 2009 Author Share #9 Posted January 30, 2009 Several improvements we can expect: The Online Photographer: (Still) Keeping the 'X' in 'Xmas' Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.