Velocity Reviews (http://www.velocityreviews.com/forums/index.php)
-   Digital Photography (http://www.velocityreviews.com/forums/f37-digital-photography.html)
-   -   Re: Circle of confusion (http://www.velocityreviews.com/forums/t752357-re-circle-of-confusion.html)

 Gordon Freeman 08-02-2011 09:31 PM

Re: Circle of confusion

Alfred Molon <alfred_molon@yahoo.com> wrote:

> It would seem that the circle of confusion (for the purpose of
> calculating DoF) only depends on the size of the sensor:
> http://en.wikipedia.org/wiki/Circle_..._confusion_dia
> meter_limit_based_on_d.2F1500
>
> with the formula d/1500.
>
> But doesn't the circle of confusion also depend on the size of the
> pixels of the sensor?

It's the other way round - there's no point in making the pixels smaller
than the detail the lens can resolve, and the circle of confusion figures
tell you what that size is, or more specifically the Airy disc calculation
tells you the limit for a given lens opening due to diffraction.

 David J Taylor 08-03-2011 07:50 AM

Re: Circle of confusion

"Alfred Molon" <alfred_molon@yahoo.com> wrote in message
news:MPG.28a30723cdf8b8fb98c58b@news.supernews.com ...
> In article <Xns9F35E52A79BCBC9A7@127.0.0.1>, Gordon Freeman says...
>> It's the other way round - there's no point in making the pixels
>> smaller
>> than the detail the lens can resolve, and the circle of confusion
>> figures
>> tell you what that size is, or more specifically the Airy disc
>> calculation
>> tells you the limit for a given lens opening due to diffraction.

>
> But the CoC does not depend on the lens.
> --
>
> Alfred Molon

Nor on the sensor. It's just another way of saying how many pixels you
need so that the eye cannot see those individual pixels, under particular
viewing conditions. In file days, that was most easily expressed as
either a nominal pixel size on the sensor, or as a fraction of the sensor
(film) height or width.

The bigger the print, or the closer you look, the smaller the CoC as a
fraction of the image size. Whether than CoC is greater or less than the
diffraction spread depends on the viewing conditions.

At least, that's the way I see it.

Cheers,
David

 David J Taylor 08-04-2011 04:58 AM

Re: Circle of confusion

"Alfred Molon" <alfred_molon@yahoo.com> wrote in message
news:MPG.28a3a2aaaaf233f998c58c@news.supernews.com ...
> In article <j1aukc\$fke\$1@dont-email.me>, David J Taylor says...
>> The bigger the print, or the closer you look, the smaller the CoC as a
>> fraction of the image size. Whether than CoC is greater or less than
>> the
>> diffraction spread depends on the viewing conditions.

>
> Clearly a camera with more MP can produce larger images, so the CoC will
> be smaller, i.e. if you define acceptable sharpness what the camera is
> able to resolve, the camera with more pixels (and same sensor area) will
> have a smaller CoC.
> --
>
> Alfred Molon

That's what I said - the CoC depends solely on the viewing conditions,
/not/ the camera.

For a give print size (and other viewing conditions), the CoC remains the
same, and independent of the camera. Whether the camera can achieve that
CoC is another issue.

Given a bigger print, viewed at the same distance, the angular resolution
of the eye is unchanged, so it's a smaller fraction of the angle subtended
by the image, and hence the CoC (expressed as a fraction of the image
height, or as a physical size on a given sensor size) will be less,

How the imaging system meets that CoC depends on both the point spread
function of the lens and of the sensor. You might use an RMS addition of
those two values to get the PSF of the system (in a simplistic way).

Cheers,
David

 David J Taylor 08-04-2011 07:39 AM

Re: Circle of confusion

"Alfred Molon" <alfred_molon@yahoo.com> wrote in message
news:MPG.28a45794a578241398c58f@news.supernews.com ...
> In article <j1d8sr\$pv5\$1@dont-email.me>, David J Taylor says...
>> That's what I said - the CoC depends solely on the viewing conditions,
>> /not/ the camera.

>
> And the viewing conditions depend on the camera, so the CoC depends on
> the camera...
> --
>
> Alfred Molon

I do see how you are thinking, Alfred, but I disagree with your statement.

If I'm looking at a certain size of print at a certain distance, I have no
knowledge of what camera took the image. The CoC is independent of the
camera. It's a different question to ask whether a particular camera and
lens can meet the CoC requirements for a particular image size and viewing
conditions.

Cheers,
David

 All times are GMT. The time now is 09:43 PM.