Current Cameras with No AA Filters (a running thread)

**RobA_Oz** · 09-29-2012, 03:34 PM

Originally posted by kh1234567890

A birefringent filter does not blur the image, it doubles it.

Birefringence Demonstrated - YouTube

Doubling the image will have the concomitant effect of halving the light that falls on each sensor site, ignoring losses. On the surface of it, this means that removing such a filter may improve low-light performance, for a given sensor.

newarts · 09-29-2012, 03:55 PM

Originally posted by kh1234567890

A birefringent filter does not blur the image, it doubles it.

Birefringence Demonstrated - YouTube

Thanks, I was aware of that from Wikipedia Anti-aliasing filter - Wikipedia, the free encyclopedia.

Following up with Falk's work LumoLabs -- Nikon D800 AA filter vs. D800E -- Whitepaper it is clear that the net effect of the birefringent AA filter is to broaden an edge profile a bit. I don't yet understand his work well enough to compare his result to the corresponding effective ESF increase.

Dave

kh1234567890 · 09-29-2012, 04:20 PM

Originally posted by RobA_Oz

Doubling the image will have the concomitant effect of halving the light that falls on each sensor site, ignoring losses. On the surface of it, this means that removing such a filter may improve low-light performance, for a given sensor.

Aaargh !

At a risk of a considerable oversimplification (it is much easier to explain this with some overhead transparency and coloured felt tip pens) :

Imagine that your camera is taking a picture of a very small blue object. Just big enough give an image on the sensor one pixel in size. Now move the camera a little bit so that this image falls onto a red sensitive element. The sensor will not see it.

Now put in a birefringent filter just thick enough so that the double image has spacing equal to the distance between the sensor elements (ignore for the moment how the colour sensitive elements are arranged, i.e. imagine the Bayer sensor turned by 45 degrees). You'll get an image of your blue point falling not only on the blind red sensitive element but also on an adjacent blue sensitive element. Bingo. Your Bayer sensor will now see it.

You'll also need another birefringent filter at 90 degrees to make this work in the other (perpendicular) direction. So the image of your blue point actually becomes four blue point images, each as sharp as the original but offset from it.

**RobA_Oz** · 09-29-2012, 07:36 PM

Originally posted by kh1234567890

Aaargh !

At a risk of a considerable oversimplification (it is much easier to explain this with some overhead transparency and coloured felt tip pens) :

Imagine that your camera is taking a picture of a very small blue object. Just big enough give an image on the sensor one pixel in size. Now move the camera a little bit so that this image falls onto a red sensitive element. The sensor will not see it.

Now put in a birefringent filter just thick enough so that the double image has spacing equal to the distance between the sensor elements (ignore for the moment how the colour sensitive elements are arranged, i.e. imagine the Bayer sensor turned by 45 degrees). You'll get an image of your blue point falling not only on the blind red sensitive element but also on an adjacent blue sensitive element. Bingo. Your Bayer sensor will now see it.

You'll also need another birefringent filter at 90 degrees to make this work in the other (perpendicular) direction. So the image of your blue point actually becomes four blue point images, each as sharp as the original but offset from it.

Thanks. I understood that from your previous posts. My point (really half a question, so I probably should have used a question mark) was that the halving in the number of lumens falling on a given element will introduce a great possibility of noise being generated. That's a different matter from the colour-blindness arising from the Bayer mask, or have I missed something fundamental?

**RobA_Oz** · 09-30-2012, 01:12 PM

At the risk of answering my own question, I probably did miss something. Of course, the light falling on adjacent elements will be likewise split, so there will be no change in the overall luminous intensity on the sensor surface. The only losses will be the light falling between the sensor elements, which will be the same regardless of the number of times the image is split, except around the edges of the sensor.

newarts · 09-30-2012, 03:23 PM

Originally posted by newarts

Following up with Falk's work LumoLabs -- Nikon D800 AA filter vs. D800E -- Whitepaper it is clear that the net effect of the birefringent AA filter is to broaden an edge profile a bit. I don't yet understand his work well enough to compare his result to the corresponding effective ESF increase.

It seems to me a 100% strength birefringent AA filter atop a Bayer sensor array must add at least one 50% bright pixel at the edge of a step function illumination pattern.

That's the result I get thinking it thru & if I simulate it numerically by averaging four adjacent pixels. I think maybe this edge broadening should add according to the Kodak rule with p=2.

Dave

Blue · 09-30-2012, 04:02 PM

Out of curiosity, how does everyone think the K-5 II/s will stack up against the Canon 60Da for astrophotography?

rawr · 09-30-2012, 04:24 PM

Originally posted by Blue

Out of curiosity, how does everyone think the K-5 II/s will stack up against the Canon 60Da for astrophotography?

The 60Da seems a little bit more specialised than the K-5 IIs:

Quote:

this camera incorporates a modified infra-red blocking filter in the low-pass filter allowing great sensitivity to hydrogen-alpha (H-α) wavelengths of light - around 3 times greater than with the 60D. This enables clearer pictures of astronomical phenomena such as red-coloured Emission Nebulae.

[from the 60Da blurb at a local camera site]

09-29-2012, 04:20 PM	#18
kh1234567890 Veteran Member Join Date: Sep 2010 Location: Manchester, UK Photos: Gallery Posts: 2,653	Originally posted by RobA_Oz Doubling the image will have the concomitant effect of halving the light that falls on each sensor site, ignoring losses. On the surface of it, this means that removing such a filter may improve low-light performance, for a given sensor. Aaargh ! At a risk of a considerable oversimplification (it is much easier to explain this with some overhead transparency and coloured felt tip pens) : Imagine that your camera is taking a picture of a very small blue object. Just big enough give an image on the sensor one pixel in size. Now move the camera a little bit so that this image falls onto a red sensitive element. The sensor will not see it. Now put in a birefringent filter just thick enough so that the double image has spacing equal to the distance between the sensor elements (ignore for the moment how the colour sensitive elements are arranged, i.e. imagine the Bayer sensor turned by 45 degrees). You'll get an image of your blue point falling not only on the blind red sensitive element but also on an adjacent blue sensitive element. Bingo. Your Bayer sensor will now see it. You'll also need another birefringent filter at 90 degrees to make this work in the other (perpendicular) direction. So the image of your blue point actually becomes four blue point images, each as sharp as the original but offset from it. Last edited by kh1234567890; 09-29-2012 at 04:32 PM.

09-30-2012, 03:23 PM	#21
newarts Inactive Account Join Date: Dec 2008 Location: Ames, Iowa, USA Photos: Albums Posts: 2,965	Originally posted by newarts Following up with Falk's work LumoLabs -- Nikon D800 AA filter vs. D800E -- Whitepaper it is clear that the net effect of the birefringent AA filter is to broaden an edge profile a bit. I don't yet understand his work well enough to compare his result to the corresponding effective ESF increase. It seems to me a 100% strength birefringent AA filter atop a Bayer sensor array must add at least one 50% bright pixel at the edge of a step function illumination pattern. That's the result I get thinking it thru & if I simulate it numerically by averaging four adjacent pixels. I think maybe this edge broadening should add according to the Kodak rule with p=2. Dave Last edited by newarts; 09-30-2012 at 03:53 PM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Donation/information request - my school is running out of cameras!	Steve Beswick	Film SLRs and Compact Film Cameras	45	03-20-2014 08:43 PM
Running thread of consolidated Tea Party-related stupidity	deadwolfbones	General Talk	289	11-04-2011 08:17 PM
Running Thread of Catharsis	shooz	General Talk	54	05-24-2011 03:30 PM

09-29-2012, 03:34 PM	#16
RobA_Oz Site Supporter Join Date: Nov 2008 Location: Tasmania, Australia Photos: Gallery Posts: 8,197	Originally posted by kh1234567890 A birefringent filter does not blur the image, it doubles it. Birefringence Demonstrated - YouTube Doubling the image will have the concomitant effect of halving the light that falls on each sensor site, ignoring losses. On the surface of it, this means that removing such a filter may improve low-light performance, for a given sensor.

09-29-2012, 03:55 PM	#17
newarts Inactive Account Join Date: Dec 2008 Location: Ames, Iowa, USA Photos: Albums Posts: 2,965	Originally posted by kh1234567890 A birefringent filter does not blur the image, it doubles it. Birefringence Demonstrated - YouTube Thanks, I was aware of that from Wikipedia Anti-aliasing filter - Wikipedia, the free encyclopedia. Following up with Falk's work LumoLabs -- Nikon D800 AA filter vs. D800E -- Whitepaper it is clear that the net effect of the birefringent AA filter is to broaden an edge profile a bit. I don't yet understand his work well enough to compare his result to the corresponding effective ESF increase. Dave

09-29-2012, 07:36 PM	#19
RobA_Oz Site Supporter Join Date: Nov 2008 Location: Tasmania, Australia Photos: Gallery Posts: 8,197	Originally posted by kh1234567890 Aaargh ! At a risk of a considerable oversimplification (it is much easier to explain this with some overhead transparency and coloured felt tip pens) : Imagine that your camera is taking a picture of a very small blue object. Just big enough give an image on the sensor one pixel in size. Now move the camera a little bit so that this image falls onto a red sensitive element. The sensor will not see it. Now put in a birefringent filter just thick enough so that the double image has spacing equal to the distance between the sensor elements (ignore for the moment how the colour sensitive elements are arranged, i.e. imagine the Bayer sensor turned by 45 degrees). You'll get an image of your blue point falling not only on the blind red sensitive element but also on an adjacent blue sensitive element. Bingo. Your Bayer sensor will now see it. You'll also need another birefringent filter at 90 degrees to make this work in the other (perpendicular) direction. So the image of your blue point actually becomes four blue point images, each as sharp as the original but offset from it. Thanks. I understood that from your previous posts. My point (really half a question, so I probably should have used a question mark) was that the halving in the number of lumens falling on a given element will introduce a great possibility of noise being generated. That's a different matter from the colour-blindness arising from the Bayer mask, or have I missed something fundamental?

09-30-2012, 01:12 PM	#20
RobA_Oz Site Supporter Join Date: Nov 2008 Location: Tasmania, Australia Photos: Gallery Posts: 8,197	At the risk of answering my own question, I probably did miss something. Of course, the light falling on adjacent elements will be likewise split, so there will be no change in the overall luminous intensity on the sensor surface. The only losses will be the light falling between the sensor elements, which will be the same regardless of the number of times the image is split, except around the edges of the sensor.

09-30-2012, 04:02 PM	#22
Blue Moderator Site Supporter Join Date: Jun 2008 Location: Florida Hill Country Photos: Gallery \| Albums Posts: 17,377	Out of curiosity, how does everyone think the K-5 II/s will stack up against the Canon 60Da for astrophotography?

09-30-2012, 04:24 PM	#23
rawr Veteran Member Join Date: Jun 2009 Posts: 11,913	Originally posted by Blue Out of curiosity, how does everyone think the K-5 II/s will stack up against the Canon 60Da for astrophotography? The 60Da seems a little bit more specialised than the K-5 IIs: Quote: this camera incorporates a modified infra-red blocking filter in the low-pass filter allowing great sensitivity to hydrogen-alpha (H-α) wavelengths of light - around 3 times greater than with the 60D. This enables clearer pictures of astronomical phenomena such as red-coloured Emission Nebulae. [from the 60Da blurb at a local camera site]