Albert Theuwissen publishes the second part of his ISSCC review. This part covers CMOSIS 391MP sensor and Canon 250MP APS-H imager, as well as Toshiba and TSMC papers.
It is a custom lens made by Zeiss that we use for the sensor. 92mm focal length F/5.6 Look at: http://leica-geosystems.com/products/airborne-systems/cameras-large-format/leica-dmciii
If I understand correctly its not 391 Mp in a single sensor, but in a 3x6 stitched array of sensors.
"The sensor is using stitching : 6 x 3 blocks are stitched in the active area, with 4352 x 5000 pixels in each stitched block." Source: http://harvestimaging.com/blog/?p=1536
There is a major difference between butting (= what you are referring to) and stitching (= what I am referring to). This sensor is stitched and is made out of 1 tile of silicon.
Very good job from technical committee ;). Dual gain amplifier first shown by Kawahito's group at ISSCC 2003. 13 years ago!! Now shows up again as main "novelty" from the work of Canon. Submission rules of ISSCC say novelty is needed, not only nice results. N.O.V.E.L.T.Y.!
First of all this : I am no longer involved in any ISSCC paper selection or so, don't shoot at the piano player. Secondly : 250 Mpixel (1.5 um, 0.13 um) forced into an APS-size device is an achievement as well. Dual gain amplifier is not new indeed, but 4T pixels neither ... and so many other techniques/architectures/methods.
I found this paper from Kawahito's group. http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=1425722&newsearch=true&queryText=column%20gain%20cmos%20image%20sensor It looks 4T pixels....
391MP
ReplyDeleteIs this the CMOS highest resolution sensor record?
I bet outside airborne application, some will build a giga pixel camera with it.
if my computation is correct, it's 26000x15000 => 100mmx58mm for the pixel array. What lens can support this???
It is a custom lens made by Zeiss that we use for the sensor.
Delete92mm focal length F/5.6
Look at:
http://leica-geosystems.com/products/airborne-systems/cameras-large-format/leica-dmciii
The chip size is indeed very large, 6 dies per 300 mm wafer.
ReplyDeleteIf I understand correctly its not 391 Mp in a single sensor, but in a 3x6 stitched array of sensors.
Delete"The sensor is using stitching : 6 x 3 blocks are stitched in the active area, with 4352 x 5000 pixels in each stitched block." Source: http://harvestimaging.com/blog/?p=1536
There is a major difference between butting (= what you are referring to) and stitching (= what I am referring to). This sensor is stitched and is made out of 1 tile of silicon.
DeleteVery good job from technical committee ;). Dual gain amplifier first shown by Kawahito's group at ISSCC 2003. 13 years ago!! Now shows up again as main "novelty" from the work of Canon.
ReplyDeleteSubmission rules of ISSCC say novelty is needed, not only nice results. N.O.V.E.L.T.Y.!
First of all this : I am no longer involved in any ISSCC paper selection or so, don't shoot at the piano player. Secondly : 250 Mpixel (1.5 um, 0.13 um) forced into an APS-size device is an achievement as well. Dual gain amplifier is not new indeed, but 4T pixels neither ... and so many other techniques/architectures/methods.
DeleteI found this paper from Kawahito's group.
Deletehttp://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=1425722&newsearch=true&queryText=column%20gain%20cmos%20image%20sensor
It looks 4T pixels....