[LRflex] Re: Interesting info for digitographers
- From: Douglas Sharp <douglas.sharp@xxxxxx>
- To: leicareflex@xxxxxxxxxxxxx
- Date: Tue, 08 Nov 2005 23:36:24 +0100
Andy,
If someone can tell me what size the sensor is, it's diagonal, and what
the distance is from centres of adjacent pixels, I'll take a crack at
working it out with the method they used in the magazine.
The interesting part about the info is -it has absolutely nothing to do
with the angle of the light hitting the sensor, it's the circle around a
point of light (Airy Disk, which I believe also has something to do with
resolution for astronomy - seems to be lurking somewhere in the back of
my mind). the smaller the aperture the larger the diffuse circle of
light (with respect to the size of the aperture) around the point(ray)
of light it's letting through. At a certain aperture this exceeds the
distance between two adjacent pixel centres and can't be resolved as a
point any more. Then the article gets into wave and particle theory as
explanations, which is probably unnecessary 'cos it all ends up as a
reasonably simple mathematical formula. At the end of the day it's not
the number of pixels that define the resolution for very small apertures
but the size of a pixel. looks as if the next development should be an
automatic algorithm for cross coupling groups of 4 adjacent pixels as
soon as the aperture exceeds f11 (or the limit defined by the sensor
geometry). I suspect that the 3 "layer" Foveon chip in the Sigma does
something along these lines.
At least I hope that's right, I'm a Geophysicist not a Physicist.
If I have time tomorrow I'll translate and post the article, and the
formula for working it out.
Douglas
Andy Wagner wrote:
>Interesting comment!
>Do we know how the DMR stacks up in this comparison?
>I remember reading somewhere that the problem Leica was having with
>introducing a digital back was because the angle of light hitting the
>perimeter pixals was causing havoc with sharpness. Also pixels sensitivity to
>light and the way they interprate colors varys with the same angle.
>
>LONG LIVE THE SL's :-))
>Douglas Sharp <douglas.sharp@xxxxxx> wrote:
>In the latest issue of ColorFoto (Germany) there's an interesting
>article about the limitations of digital cameras at certain lens apertures.
>It's all to do with refraction and pixel sizes, Airy disk size or
>Fraunhofer circles of diffraction - the main article is pretty much
>physics/optical theory but the list of cameras and limiting apertures is
>interesting.
>
>What it boils down to is - it's not worth stopping down to get a sharper
>image any further than the apertures shown below, it just won't increase
>sharpness, in fact it will get unsharp.
>(NOT DOF, that, of course, will increase)
>
>Camera/aperture
>Canon EOS 1Ds Mk II 11.8
>Canon EOS 1D Mk II 13.4
>Nikon D70 12.9
>Fujifilm S3 Pro 12.5
>Canon EOS 20 D 10.6
>Sigma SD10 15.0
>4/3 cameras 9.0
>
>And it gets worse for compacts.
>
>It finally explains why my macro shots with f16 were not as pin-sharp as
>those with f8.
>Douglas
>------
>Unsubscribe or change to/from Digest Mode at:
>http://www3.telus.net/~telyt/lrflex.htm
>Archives are at:
>www.freelists.org/archives/leicareflex/
>
>
>------ >--
>Regards
>YXAndy
>
>
>------
>Unsubscribe or change to/from Digest Mode at:
> http://www3.telus.net/~telyt/lrflex.htm
>Archives are at:
> www.freelists.org/archives/leicareflex/
>
>
>
>
------
Unsubscribe or change to/from Digest Mode at:
http://www3.telus.net/~telyt/lrflex.htm
Archives are at:
www.freelists.org/archives/leicareflex/
Other related posts:
