I think you're chasing a ghost. I had a look at the tables you linked
below and did some sample calculations for 50mm at f/8 using a
conventional DoF calculator. Those calculations don't agree with the
Zeiss cine lens table for 50mm but they're not far off.
The first thing I noticed (which is typical for cine lenses) is that the
aperture stops listed are "T-stops" instead of "F-stops". For those not
familiar with T-stops a T-stop is based on actual light transmission
through the lens as opposed to an F-stop which is based on physical
aperture. The difference between an F-stop and a T-stop is that the
T-stop corrects for actual light lost to absorption and reflection in
addition to that from simply changing the aperture. It's a small amount
but critical to cinemaphotographers.
I next asked myself how much light might be lost in passing through the
glass and how much might the lens have to be opened from a marked F-stop
in order to compensate. I thought maybe 1/3 stop might be a reasonable
guess as to a maximum. Opening up from f/8 to f/7.1 is 1/3 stop. What
I discovered is that using a conventional calculator and substituting
f/7.2 or f/7.3 (more like 1/4 stop) gives very close agreement with the
Zeiss 50mm table.
My conclusion is that the differences between the Zeiss table and a thin
lens calculation is likely mostly due to the difference between a T-stop
and an F-stop. While the real geometry of the lens may lead to
different values than the thin lens equation, the differences due solely
to lens geometry are probably very small and not necessary to worry about.
Finally, the only time I really pay much attention to DoF is when doing
hyperfocal shooting of landscapes. Usually the near focusing distance
has to be estimated. If your estimate is off such that you are focusing
too close you risk not getting infinity in focus. Therefore I always
focus slightly past my estimated distance or else use an aperture
slightly smaller than my table calls for.
Dr. Focus
On 11/28/2010 8:51 PM, usher99@xxxxxxx wrote:
> I asked Zeiss to comment on the possibility of lens specific dof tables
> for some of their lenses especially given the white
> paper by Nasse stating that the thin lens approximations for dof
> "should not be taken too seriously." Dr. Nasse himself responded.
> They do provide full wave calculated dof tables for cine lenses--see
> below---- but not otherwise, for reasons that are not clear to me. He
> stated that the calculated tables
> would apply to the center area if any field curvature were present and
> subtle errors in focusing or manufacturing flange to sensor/film
> distances could easily
> overwhelm the effects. He by no means denied that lens design per se
> can affect the dof with all other factors being equal.
>
>
> I did a few checks on the lenses below and there are some differences
> with the usual dof formulas. I could not easily extract the pdf data
> into excel
> where it would be easier to try and discern what factors lead to the
> largest differences from the usual dof calculators.
>
> Trying to Escape the Black Hole that is the CoC for dof, Mike
>
> http://tinyurl.com/28qqhqq
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|