Patent: Canon Curved Sensor to Reduce Vignetting

[Canon Rumors]

It looks like Canon is continuing to research curved sensors, this patent is geared towards light fall off on the sides of the sensor.</p>
<p>Patent Publication No. 2016-197663 (Google Translated)</p>
<ul>
<li>Published 2016.11.24</li>
<li>Filing date 2015.4.3</li>
<li>The central portion of the light falloff is not noticeable to the flat shape</li>
<li>The periphery light falloff is noticeable a curved shape</li>
</ul>

		<style type='text/css'>
			#gallery-1 {
				margin: auto;
			}
			#gallery-1 .gallery-item {
				float: left;
				margin-top: 10px;
				text-align: center;
				width: 50%;
			}
			#gallery-1 img {
				border: 2px solid #cfcfcf;
			}
			#gallery-1 .gallery-caption {
				margin-left: 0;
			}
			/* see gallery_shortcode() in wp-includes/media.php */
		</style>
		<div id='gallery-1' class='gallery galleryid-27518 gallery-columns-2 gallery-size-thumbnail'><dl class='gallery-item'>
			<dt class='gallery-icon portrait'>
				<a href='http://www.canonrumors.com/wp-content/uploads/2016/11/2016_197663_fig03.png'><img width="168" height="168" src="http://www.canonrumors.com/wp-content/uploads/2016/11/2016_197663_fig03-168x168.png" class="attachment-thumbnail size-thumbnail" alt="2016_197663_fig03" /></a>
			</dt></dl><dl class='gallery-item'>
			<dt class='gallery-icon landscape'>
				<a href='http://www.canonrumors.com/wp-content/uploads/2016/11/2016_197663_fig02.png'><img width="168" height="149" src="http://www.canonrumors.com/wp-content/uploads/2016/11/2016_197663_fig02-168x149.png" class="attachment-thumbnail size-thumbnail" alt="2016_197663_fig02" /></a>
			</dt></dl><br style="clear: both" />
		</div>

<span id="pty_trigger"></span>

 

[Don Haines]

Before everyone gets too excited, this is probably for a P/S camera and not a DSLR.....

 

[YuengLinger]

Ha ha, now we have to buy new bodies to make the ef 16-35 f/2.8 III work right.

 

[Alex_M]

Any existing aps-c body will serve you well in that regard

Ha ha, now we have to buy new bodies to make the ef 16-35 f/2.8 III work right.

 

[Luds34]

Any existing aps-c body will serve you well in that regard

Ha ha, now we have to buy new bodies to make the ef 16-35 f/2.8 III work right.

Oh-oh, inside joke I'm missing? Is there some issue with the new 16-35 f/2.8??? I haven't been keeping up on my camera news lately so the last I saw was initial test charts showing the new f/2.8 beating the f/4 IS.

Is there some softness/issue in the corners? And if so I assume they are fine stopped down?

 

[3kramd5]

Any existing aps-c body will serve you well in that regard

Ha ha, now we have to buy new bodies to make the ef 16-35 f/2.8 III work right.

Oh-oh, inside joke I'm missing? Is there some issue with the new 16-35 f/2.8??? I haven't been keeping up on my camera news lately so the last I saw was initial test charts showing the new f/2.8 beating the f/4 IS.

Is there some softness/issue in the corners?

It loses about 4 stops to vignetting in the corners wide open at 16mm. So there you go, a real world case for 4-stop pushes!

 

[ahsanford]

Any existing aps-c body will serve you well in that regard

Ha ha, now we have to buy new bodies to make the ef 16-35 f/2.8 III work right.

Oh-oh, inside joke I'm missing? Is there some issue with the new 16-35 f/2.8??? I haven't been keeping up on my camera news lately so the last I saw was initial test charts showing the new f/2.8 beating the f/4 IS.

Is there some softness/issue in the corners? And if so I assume they are fine stopped down?

Consider yourself caught up:

The good news: https://www.lensrentals.com/blog/2016/10/canon-16-35mm-f2-8l-mark-iii-optical-bench-tests/

The not so good news: http://www.the-digital-picture.com/Reviews/Lens-Vignetting-Test-Results.aspx?FLI=0&API=0&FLIComp=0&APIComp=0&Lens=1073&Camera=979&LensComp=986

Yes, it's fine if you stop down, but why would you buy an f/2.8 lens if you have to stop it down? For some folks (reportage, sports, astro, events) this lens is shot wide open often, so stopping down is a luxury they may not be able to support.

- A

 

[9VIII]

Before everyone gets too excited, this is probably for a P/S camera and not a DSLR.....

Because EOS-M still uses Powershot serial numbers.

Short flange distance is the best application for this tech. If a Full Frame EOS-M ever comes out, it had better use a curved sensor.

 

[tron]

Before everyone gets too excited, this is probably for a P/S camera and not a DSLR.....

Because EOS-M still uses Powershot serial numbers.

Short flange distance is the best application for this tech. If a Full Frame EOS-M ever comes out, it had better use a curved sensor.

How much curved? If it will be an interchangeable lens camera then there will be an obvious issue...

 

[BeenThere]

Before everyone gets too excited, this is probably for a P/S camera and not a DSLR.....

Because EOS-M still uses Powershot serial numbers.

Short flange distance is the best application for this tech. If a Full Frame EOS-M ever comes out, it had better use a curved sensor.

How much curved? If it will be an interchangeable lens camera then there will be an obvious issue...

Will require lenses designed to match the curved sensor. Like a fixed lens camera for instance.

 

[gmrza]

Before everyone gets too excited, this is probably for a P/S camera and not a DSLR.....

Because EOS-M still uses Powershot serial numbers.

Short flange distance is the best application for this tech. If a Full Frame EOS-M ever comes out, it had better use a curved sensor.

I have a feeling that this may be the "dirty little little secret" of the EOS M. Photozone.de raises this issue in some of its EF-M lens tests. Very fast lenses, like the 85mm f1.2L are notorious for the amount of fall-off in peripheral illumination, and that is without even having the shorter flange distance that would be associated with a mirrorless design!

I have a sneaking suspicion that this is an issue which has been hampering Canon in the mirrorless space. Canon needs a way to may sensors more sensitive to light with a very oblique angle of incidence.

A curved sensor would probably also make it easier to address the issue of focal plane curvature (which was a problem with lenses like the 24-70 f/2.8L mkI). A different problem in this space is that a lot of lenses address focal plane curvature optically....

 

[Don Haines]

Before everyone gets too excited, this is probably for a P/S camera and not a DSLR.....

Because EOS-M still uses Powershot serial numbers.

Short flange distance is the best application for this tech. If a Full Frame EOS-M ever comes out, it had better use a curved sensor.

I have a feeling that this may be the "dirty little little secret" of the EOS M. Photozone.de raises this issue in some of its EF-M lens tests. Very fast lenses, like the 85mm f1.2L are notorious for the amount of fall-off in peripheral illumination, and that is without even having the shorter flange distance that would be associated with a mirrorless design!

I have a sneaking suspicion that this is an issue which has been hampering Canon in the mirrorless space. Canon needs a way to may sensors more sensitive to light with a very oblique angle of incidence.

A curved sensor would probably also make it easier to address the issue of focal plane curvature (which was a problem with lenses like the 24-70 f/2.8L mkI). A different problem in this space is that a lot of lenses address focal plane curvature optically....

The problem is, different focal lengths require different curvatures.... This is excellent for a cell phone camera with a fixed lens, sort-of-OK on a P/S camera where you have some degree of control over the focal length, but horrendous for DSLRs and interchangeable lens mirrorless cameras where one could be using an 8mm fisheye or an 800mm lens with a 2X teleconverter... 8 to 1600 is quite a range.... or if you include second party and specialty lenses and your range goes from 4.5mm to 4000mm...

 

[Mt Spokane Photography]

All digital cameras with Bayer Filters have issues with light fall off due to the angle of the light striking the sensor. A curved sensor will lessen that issue. The camera firmware applies extra gain to the outer photosites to help compensate, which means more noise in those areas.

A layered sensor like a Foveon will see the effect even more.

A separate but closely linked issue is curvature of field that is a lens property. This makes it difficult to use a curved sensor on a ILC because of the different properties of each lens.

As noted, its best for a fixed lens camera, but that's a moving target.

 

[waldi72]

Before everyone gets too excited, this is probably for a P/S camera and not a DSLR.....

Because EOS-M still uses Powershot serial numbers.

Short flange distance is the best application for this tech. If a Full Frame EOS-M ever comes out, it had better use a curved sensor.

I have a feeling that this may be the "dirty little little secret" of the EOS M. Photozone.de raises this issue in some of its EF-M lens tests. Very fast lenses, like the 85mm f1.2L are notorious for the amount of fall-off in peripheral illumination, and that is without even having the shorter flange distance that would be associated with a mirrorless design!

I have a sneaking suspicion that this is an issue which has been hampering Canon in the mirrorless space. Canon needs a way to may sensors more sensitive to light with a very oblique angle of incidence.

A curved sensor would probably also make it easier to address the issue of focal plane curvature (which was a problem with lenses like the 24-70 f/2.8L mkI). A different problem in this space is that a lot of lenses address focal plane curvature optically....

The problem is, different focal lengths require different curvatures.... This is excellent for a cell phone camera with a fixed lens, sort-of-OK on a P/S camera where you have some degree of control over the focal length, but horrendous for DSLRs and interchangeable lens mirrorless cameras where one could be using an 8mm fisheye or an 800mm lens with a 2X teleconverter... 8 to 1600 is quite a range.... or if you include second party and specialty lenses and your range goes from 4.5mm to 4000mm...

How about this: "A patent showing curved sensor development, this one deals with using a magnetic system to deform the sensor
http://www.canonrumors.com/patent-canon-continues-to-work-on-curved-sensor-technology/

 

[Don Haines]

Before everyone gets too excited, this is probably for a P/S camera and not a DSLR.....

Because EOS-M still uses Powershot serial numbers.

Short flange distance is the best application for this tech. If a Full Frame EOS-M ever comes out, it had better use a curved sensor.

I have a feeling that this may be the "dirty little little secret" of the EOS M. Photozone.de raises this issue in some of its EF-M lens tests. Very fast lenses, like the 85mm f1.2L are notorious for the amount of fall-off in peripheral illumination, and that is without even having the shorter flange distance that would be associated with a mirrorless design!

I have a sneaking suspicion that this is an issue which has been hampering Canon in the mirrorless space. Canon needs a way to may sensors more sensitive to light with a very oblique angle of incidence.

A curved sensor would probably also make it easier to address the issue of focal plane curvature (which was a problem with lenses like the 24-70 f/2.8L mkI). A different problem in this space is that a lot of lenses address focal plane curvature optically....

The problem is, different focal lengths require different curvatures.... This is excellent for a cell phone camera with a fixed lens, sort-of-OK on a P/S camera where you have some degree of control over the focal length, but horrendous for DSLRs and interchangeable lens mirrorless cameras where one could be using an 8mm fisheye or an 800mm lens with a 2X teleconverter... 8 to 1600 is quite a range.... or if you include second party and specialty lenses and your range goes from 4.5mm to 4000mm...

How about this: "A patent showing curved sensor development, this one deals with using a magnetic system to deform the sensor
http://www.canonrumors.com/patent-canon-continues-to-work-on-curved-sensor-technology/

Now THAT makes things interesting and greatly increases the potential!