Canon Develops World's Largest CMOS Sensor

[Canon Rumors]

<div id="attachment_4917" class="wp-caption aligncenter" style="width: 585px"><img class="size-full wp-image-4917" title="cmos" src="http://www.canonrumors.com/wp-content/uploads/2010/08/cmos.jpg" alt="" width="575" height="360" /><p class="wp-caption-text">The new sensor beside a 35mm Full Frame Sensor</p></div>
<p><strong>Canon succeeds in developing world’s largest CMOS image sensor, with ultra-high sensitivity</strong></p>
<p><strong>TOKYO, August 31, 2010</strong>-Canon Inc. announced today that it has successfully developed the world’s largest*1 CMOS image sensor, with a chip size measuring 202 x 205 mm. Because its expanded size enables greater light-gathering capability, the sensor is capable of capturing images in one one-hundredth the amount of light required by a professional-model digital SLR camera.</p>
<p>At 202 x 205 mm, the newly developed CMOS sensor is among the largest chips that can be produced from a 12-inch (300 mm) wafer, and is approximately 40 times the size of Canon’s largest commercial CMOS sensor.*2</p>
<p>In the past, enlarging the size of the sensor resulted in an increase in the amount of time required between the receiving and transmission of data signals, which posed a challenge to achieving high-speed readout. Canon, however, solved this problem through an innovative circuit design, making possible the realization of a massive video-compatible CMOS sensor. Additionally, by ensuring the cleanest of cleanroom environments during the production process, the sensor minimizes image imperfections and dust.</p>
<p>Because the increased size of the new CMOS sensor allows more light to be gathered, it enables shooting in low-light environments. The sensor makes possible the image capture in one one-hundredth the amount of light required by a 35 mm full-frame CMOS sensor, facilitating the shooting of 60 frame-per-second video with a mere 0.3 lux of illumination.</p>
<p>Potential applications for the new high-sensitivity CMOS sensor include the video recording of stars in the night sky and nocturnal animal behavior.</p>
<p>Through the further development of distinctive CMOS image sensors, Canon will break new ground in the world of new image expression, in the area of still images as well as video.</p>
<p><strong><span style="color: #ff0000;">c</span>r</strong></p>
<p class="facebook"><a href="http://www.facebook.com/share.php?u=http://www.canonrumors.com/2010/08/canon-develops-worlds-largest-cmos-sensor/" target="_blank" title="Share on Facebook">Share on Facebook</a></p>

 

[AprilForever]

Do we have a large format capture back here in the makings?

 

[kubelik]

this could be really amazing for applications in telescopes or satellite-based imaging of planet earth... don't think I'd want to lug around a camera body built around that thing

 

[Sebastian]

Strange that Canon doesn't give any information on the MP count or on pixel density... ???


Regards,

Sebastian

 

[Jon Gilchrist]

Although no ISO sensitivity is stated, we can make some assumptions and get an idea of where it may be. They say that it will shoot in 1/100th the light of the 5DII. If we take that as ISO 6400, the highest standard setting on the 5DII, then 1/100th would be about 6.5 stops faster or somewhere above 409,600.

6400
12,800 (1-stop)
25,600 (2-stops)
51,200 (3-stops)
102,400 (4-stops)
204,800 (5-stops)
409,600 (6-stops)
(this sensor)
819,200 (7-stops)

 

[NormanBates]

ok, I know I want bigger sensors with more pixels, but this is INSANE

 

[unruled]

rolling shutter must be pretty bad on it

 

[that1guy]

That is totally awesome! Quick, someone grab a Sinar Or maybe digital IMAX?

Now before someone goes and tells me that there is no way this will ever make it into production...I know...and I don't care. Canon seems to enjoy design excercises (50 or 120mp sensor anyone?), and I just think that it is cool that they are pushing the limits. Wow, that is really neat!

 

[afrank99]

Although no ISO sensitivity is stated, we can make some assumptions and get an idea of where it may be. They say that it will shoot in 1/100th the light of the 5DII. If we take that as ISO 6400, the highest standard setting on the 5DII, then 1/100th would be about 6.5 stops faster or somewhere above 409,600.

6400
12,800 (1-stop)
25,600 (2-stops)
51,200 (3-stops)
102,400 (4-stops)
204,800 (5-stops)
409,600 (6-stops)
(this sensor)
819,200 (7-stops)

Hm, let me make a guess... Should that be 640,000 maybe? (6400 * 100) :

 

[MadButcher]

Is this the "*BIG* Anouncement that Vincent Laforet meant?

 

[that1guy]

Whoa, just did the math and realized that is over 8" across! Forget the 4x5, where did you put the 8x10 view camera

 

[epsiloneri]

You would need a monster lens to supply that sensor with a big enough image circle. Sounds perfect for wide-field telescopes, but I'm curious what other applications. Probably just a technical demonstration with no plans to put it into production.

I find it annoying, though, that they keep perpetuating the myth that larger sensors are more light sensitive, when it's (almost) all in the aperture of the optics. Larger sensors merely samples a larger field of view.

 

[/dev/null]

Wow!
But I am almost certain that this sensor will never go into any consumer product. Maybe professional astronomy, medial x-ray sensors and the like, where up to now very expensive CCD sensors are used - although I am not aware of one *that* big.
I wonder what the readout time is like...

 

[lotus]

Great news! This is a new milestone for Canon. They are leaders in making sensors and not just cameras!

 

[unexposure]

..., but I'm curious what other applications.

As people always complained that xxxD bodys beeing too tiny for their hands, and that 1000D-body is even smaller, canon decided to make 2000d a camera for people with big hands. that's what this sensor is for. ;-)

 

[Cyrax]

I had one geeky reaction when reading this. Especially after looking at that picture...

LOL ;D

 

[that1guy]

You would need a monster lens to supply that sensor with a big enough image circle. Sounds perfect for wide-field telescopes, but I'm curious what other applications. Probably just a technical demonstration with no plans to put it into production.

I find it annoying, though, that they keep perpetuating the myth that larger sensors are more light sensitive, when it's (almost) all in the aperture of the optics. Larger sensors merely samples a larger field of view.

Actually, lenses from 8x10 view cameras would work quite nicely on that. They aren't as huge as you might think...of course, they aren't point and shoot lenses either :

 

[Jon Gilchrist]

Sounds perfect for wide-field telescopes, but I'm curious what other applications. Probably just a technical demonstration with no plans to put it into production.

The application is moving large quantities of data off the sensor. I can see that trickling down to most every camera they make.

Or maybe they'll actually make an 8" square view camera sensor. The news release says it's 40x the size of the 5DMKII sensor. Anyone have an idea of the cost of that sensor? Not the replacement cost, the part cost. Some rough numbers: Camera cost=$2500. Profit of 50%, make the sensor 25% of the remaining cost and you have a sensor cost of about $312.50, probably way high. Multiply that by 40 and you have $12,500 in sensor cost. A Hassy 39MP back currently sells for $13,995 on B&H, so an 8" square back that sold for $50k+ would not be out of line. Hmmm....

 

[NormanBates]

@Jon:

I think you're not too far off... except you didn't include yields in your calculation, and that makes a HUGE difference: for such a big chip, it has to be the main cost-driving factor

a 12" wafer on a high-end CMOS process costs $3K to $5K; maybe this one is not so high-end, but still: the problem here is how many working chips you get out of a wafer

if your numbers are correct and they pay $5K per wafer, that means only 40% of the chips actually work (if they pay $3K, it's 24%); if that's your yield and you try to build a chip that is 40 times larger, you can expect to get one that works out of every 10^16 chips you manufacture

I think they have vastly better yields, but even at 90% (which would be my guess here) that gives you only 1.5% working chips when you build a chip that is 40 times larger (=0.9^40); at 95%, it goes to 13%

bottom line: this thing is massively expensive to produce, unless you can make it with enough redundancies that you get super-high yields

 

[Jaszek]

Although no ISO sensitivity is stated, we can make some assumptions and get an idea of where it may be. They say that it will shoot in 1/100th the light of the 5DII. If we take that as ISO 6400, the highest standard setting on the 5DII, then 1/100th would be about 6.5 stops faster or somewhere above 409,600.

6400
12,800 (1-stop)
25,600 (2-stops)
51,200 (3-stops)
102,400 (4-stops)
204,800 (5-stops)
409,600 (6-stops)
(this sensor)
819,200 (7-stops)

Hm, let me make a guess... Should that be 640,000 maybe? (6400 * 100) :

It doesn't work that way. The previous post is correct.