10-06-2010, 09:16 AM
[quote name='genotypewriter' timestamp='1286325109' post='3516']
In science we don't use words like "silly" to get our points across. It just goes to show that you have little knowledge and exposure you have in the scientific method.
To put it in your sort of words... can't you read? I already have an AA-filterless camera and I have used Foveon cameras. You're the one who is speculating everything here without experience anything. Have you used any of these things you're talking so confidently about? No. lol
Manners aside, I guess you can't even read... if it's a visible only camera why do they use an ICF filter then as they describe in the first paragraph? lol
The only thing you explained is your uninformed guess, which is also scientifically unfounded. Demosaicing reduces spatial resolution. Don't you know how that works? Bayer puts 3 points in to 1 coloured pixel, Foveon keeps 3 separate coloured pixels. This is where I stop explaining because the other party is not willing to learn... at least read that dpreview article that was posted recently on the SD1. But I doubt you'd understand the theory.
GTW
[/quote]
I know apparently more about demosaicing that you do. Anyone who knows just a bit knows that all that you wrote is more or less incorrect.
First of all... "bayer" does not put 3 points into 1 coloured pixel. Depending on which ARW converter you use, demosaicing happens in different manners. And there are 2 different "values" being determined: The luminance of the pixel (mostly based on the luminance of the pixel itself, with checks in neighbouring pixels) and the hue of the pixel (which is interpolated from ALL neighbouring pixels, not "3").
As detail mostly is luminance based, it is just silly, nonsensical and untrue <img src='http://forum.photozone.de/public/style_emoticons/<#EMO_DIR#>/wink.gif' class='bbc_emoticon' alt='
' /> to think that bayer CFA sensors have a real resolution of a 3rd of the pixels or even a 4th.
Anyway, as I have shown already, the "softness" of bayer CFA sensors has to do mainly with the AA-filter used, and Sigma should use an AA-filter too (they just do not do that).
And it is you who has trouble to read (is that scientific?). They explain that the ICF and AA-filter are sandwiched together. They then say that they REPLACE the AA-ICF combination with an ICF only glass, with the same refractive index.
So what you said is just nonsense, the HR visible is not a "full spectrum camera".
There is nothing scientific about your assumptions, your arguments or you need to debunk.
And as I stated before, the SD1 article is marketing speak, not explanation and fact. It is totally written to try and exaggerate the sensor's "superiority", and does touch on how what they simplify is not really the truth, but then covers that up again.
They for instance start to state "Clever demosaicing reduces the gap.." but without going into that further they continue "..but you still have an anti-aliasing filter with Bayer."
NO word about that Sigma themselves really need an AA filter (they do suffer just as much from aliasing and moire, just only not from colour moire). No word about the clever interpolation techniques THEY have to do to try and get the colours right (and still, till this date, are unable to get colour right all the time) from the white, yellow and red layers a sigma foveon sensor has. No word about that the difficulty to get the colours right was the reason the earlier Sigma cameras only offered RAW and not JPEG (camera could not do the colour interpolation). No word about the inability to do video BECAUSE of the problem of the colour determination (too calculation intensive). And no word about the loss of light in the 3 layers.
So, basically, their article briefly mentions how you do get about the same resolution with bayer ( due to "clever demosaicing"), but how the choice to not use an AA-filter seems to give the Sigma an apparent resolution advantage. Even though the Sigma could well use a good AA-filter.
http://www.wfu.edu/~matthews/misc/DigPho...tifact.jpg
http://www.stv.ee/~donq/sigma/d/sigma-test.pdf
And no, I am not saying that there are no advantages to Sigma/foveon's approach. There are:
But there certainly are drawbacks too:
I am a fan of your lens "reviews", but your openness in this "debate" is lacking, in my opinion.
And nothing I have said in this thread is jut "my opinion", it is just widely known information about bayer CFA demosaicing, the workings and problems of foveon sensors and what AA-filters are for.
In science we don't use words like "silly" to get our points across. It just goes to show that you have little knowledge and exposure you have in the scientific method.
To put it in your sort of words... can't you read? I already have an AA-filterless camera and I have used Foveon cameras. You're the one who is speculating everything here without experience anything. Have you used any of these things you're talking so confidently about? No. lol
Manners aside, I guess you can't even read... if it's a visible only camera why do they use an ICF filter then as they describe in the first paragraph? lol
The only thing you explained is your uninformed guess, which is also scientifically unfounded. Demosaicing reduces spatial resolution. Don't you know how that works? Bayer puts 3 points in to 1 coloured pixel, Foveon keeps 3 separate coloured pixels. This is where I stop explaining because the other party is not willing to learn... at least read that dpreview article that was posted recently on the SD1. But I doubt you'd understand the theory.
GTW
[/quote]
I know apparently more about demosaicing that you do. Anyone who knows just a bit knows that all that you wrote is more or less incorrect.
First of all... "bayer" does not put 3 points into 1 coloured pixel. Depending on which ARW converter you use, demosaicing happens in different manners. And there are 2 different "values" being determined: The luminance of the pixel (mostly based on the luminance of the pixel itself, with checks in neighbouring pixels) and the hue of the pixel (which is interpolated from ALL neighbouring pixels, not "3").
As detail mostly is luminance based, it is just silly, nonsensical and untrue <img src='http://forum.photozone.de/public/style_emoticons/<#EMO_DIR#>/wink.gif' class='bbc_emoticon' alt='
' /> to think that bayer CFA sensors have a real resolution of a 3rd of the pixels or even a 4th.Anyway, as I have shown already, the "softness" of bayer CFA sensors has to do mainly with the AA-filter used, and Sigma should use an AA-filter too (they just do not do that).
And it is you who has trouble to read (is that scientific?). They explain that the ICF and AA-filter are sandwiched together. They then say that they REPLACE the AA-ICF combination with an ICF only glass, with the same refractive index.
So what you said is just nonsense, the HR visible is not a "full spectrum camera".
There is nothing scientific about your assumptions, your arguments or you need to debunk.
And as I stated before, the SD1 article is marketing speak, not explanation and fact. It is totally written to try and exaggerate the sensor's "superiority", and does touch on how what they simplify is not really the truth, but then covers that up again.
They for instance start to state "Clever demosaicing reduces the gap.." but without going into that further they continue "..but you still have an anti-aliasing filter with Bayer."
NO word about that Sigma themselves really need an AA filter (they do suffer just as much from aliasing and moire, just only not from colour moire). No word about the clever interpolation techniques THEY have to do to try and get the colours right (and still, till this date, are unable to get colour right all the time) from the white, yellow and red layers a sigma foveon sensor has. No word about that the difficulty to get the colours right was the reason the earlier Sigma cameras only offered RAW and not JPEG (camera could not do the colour interpolation). No word about the inability to do video BECAUSE of the problem of the colour determination (too calculation intensive). And no word about the loss of light in the 3 layers.
So, basically, their article briefly mentions how you do get about the same resolution with bayer ( due to "clever demosaicing"), but how the choice to not use an AA-filter seems to give the Sigma an apparent resolution advantage. Even though the Sigma could well use a good AA-filter.
http://www.wfu.edu/~matthews/misc/DigPho...tifact.jpg
http://www.stv.ee/~donq/sigma/d/sigma-test.pdf
And no, I am not saying that there are no advantages to Sigma/foveon's approach. There are:
- More precise colour per pixel in good light
- Even though there is no AA-filter, no colour moire patterns
But there certainly are drawbacks too:
- A lot of light "loss", so noisier... "no low light camera"
- Aliasing and moire problems due to the choice not to use an AA-filter
I am a fan of your lens "reviews", but your openness in this "debate" is lacking, in my opinion.
And nothing I have said in this thread is jut "my opinion", it is just widely known information about bayer CFA demosaicing, the workings and problems of foveon sensors and what AA-filters are for.