Re: comments on Anti Aliasing filterless design ?

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



David:

On 2013-03-04 08:16, PhotoRoy6@xxxxxxx wrote:
Nikon put out the D800 E and Pentax put out the K-5 IIs. Pentax say
"Anti Aliasing filterless design achieves superior resolution for rich,
detailed imagery at supreme resolution"
If this is so good why do previous digital cameras come out _with_ anti
aliasing filter and why can one get away _without_ the filter now?

Resolution increases are a lot of it. It's moving the frequency ranges that give moire away from common subjects somewhat. And some of it is an increasingly sophisticated market.


What David said..

and partly referring to my gripe from a time back when I suggested the advertised 'resolution' of a camera had little to do with resolving power limits and more to do with the image algorithms in-camera overcoming many of ther inherent flaws of a linear sensor matrix.

Also as Randy says, moire can and will still be an issue - even with higher res cameras because of the nyquist limit and the way an image is formed.

http://en.wikipedia.org/wiki/Nyquist_frequency (of course this article presumes a linear sample and digi sensors are not, they have an x and a y axis, so deviating from the x or y axis will mean the sample must be taken at *greater* than 2x the sensor frequency to reproduce it accurately. In this case 3x is preferable, ie, the theoretical limit of a 24 Mp camera is 8 Mp of true uninterpolated resolved data )

New higher resolutions means higher sampling rates. Couple this with the much clever(er) algorithms than we had in the past that can actually spot linear patterns, edge effects, other patterns and such and then compensate for them and and the need for dumb AA filter is lessened considerably, however they do not get everything and faults will still occur.

I must add that this did occur even with film but to a much lesser extent. I had a photo shot on techpan enlarged to 12x16 which showed very high fine detail receding and converging lines. I happened one day to make a colour photo where this B&W print was in the background and the resulting image showed a wild array of colours where the points converged as a consequence of diffraction and polarization. It was not visible to the naked eye.. it was purely the mechanics of the print, camera lens and film

k






[Index of Archives] [Share Photos] [Epson Inkjet] [Scanner List] [Gimp Users] [Gimp for Windows]

  Powered by Linux