10-01-2010, 10:54 AM
(This post was last modified: 10-01-2010, 10:57 AM by Brightcolours.)
[quote name='Lomskij' timestamp='1285921478' post='3418']
I'm equally surprised by this claim - interpolated image always will be softer, AA filter or not. I guess the only situation where Bayer would be equally sharp is if someone created pure B/W sensor.
[/quote]
It is not just a claim. Sharpness does not come from the colour information, but from luminance information. The luminance information is based on the information per pixel, mostly.
So, yes, a Bayer sensor without AA-filter gets about the same resolution as a foveon sensor without AA-filter.
An AA-filter is not something particular to a bayer CFA sensor. It is used to blur the image to such an extent that there are no details/edges that span less than 1 pixel, to avoid... aliasing. Aliasing happens just as much with a foveon sensor as with a Bayer CFA sensor. Aliasing is also known as "jaggies". Lines in images start to look like jagged lines, where you can see the individual square pixels, making false or wrong detail.
Aliasing can also create patterns, which are known as moire patterns. In the luminance area moire patterns show up equally with both bayer CFA and foveon sensors. The only thing that foveon sensors do not do is creating moire patterns in the croma space.
But AA-filters are not just there to avoid moire patterns with colours... yet that is the reason Sigma gives for not "needing" AA-filters. That is nonsense of course, foveon sensors without AA-filter have heavy aliasing (just not colour moire effects).
For those who like the aliased output, a the Sigma approach of course is fine. But it is incorrect to think AA-filters are only there to avoid colour moire effects.To me, aliasing in photos is just as unwanted as aliasing in computer graphics.
http://www.wfu.edu/~matthews/misc/DigPhotog/alias/
I'm equally surprised by this claim - interpolated image always will be softer, AA filter or not. I guess the only situation where Bayer would be equally sharp is if someone created pure B/W sensor.
[/quote]
It is not just a claim. Sharpness does not come from the colour information, but from luminance information. The luminance information is based on the information per pixel, mostly.
So, yes, a Bayer sensor without AA-filter gets about the same resolution as a foveon sensor without AA-filter.
An AA-filter is not something particular to a bayer CFA sensor. It is used to blur the image to such an extent that there are no details/edges that span less than 1 pixel, to avoid... aliasing. Aliasing happens just as much with a foveon sensor as with a Bayer CFA sensor. Aliasing is also known as "jaggies". Lines in images start to look like jagged lines, where you can see the individual square pixels, making false or wrong detail.
Aliasing can also create patterns, which are known as moire patterns. In the luminance area moire patterns show up equally with both bayer CFA and foveon sensors. The only thing that foveon sensors do not do is creating moire patterns in the croma space.
But AA-filters are not just there to avoid moire patterns with colours... yet that is the reason Sigma gives for not "needing" AA-filters. That is nonsense of course, foveon sensors without AA-filter have heavy aliasing (just not colour moire effects).
For those who like the aliased output, a the Sigma approach of course is fine. But it is incorrect to think AA-filters are only there to avoid colour moire effects.To me, aliasing in photos is just as unwanted as aliasing in computer graphics.
http://www.wfu.edu/~matthews/misc/DigPhotog/alias/