That may actually sound fair.
Digital image sensors as opposite to analoque media is like discrete
against continuous spectrum, yielding somewhat different approaches for
charcteristics evaluation.
I would expect discrete nature of image sensor will influence the
requirements for a physically sharp appearance. I would say that
probably it would be better estimated in pixel dimension units, i.e. to
warrant a technically maximum sharpness one would require to keep image
stable within single pixel, however not sure how that would apply for
a percieved sharpness.
Well, I think we are exceeding here the scope of this forum, so I would
be glad to continue our discussion off-line to save on EOS traffic.
Regards, Alex
--- Harman Bajwa <[EMAIL PROTECTED]> wrote:
> --- Alex Z <[EMAIL PROTECTED]> wrote:
> > In fact I did my own estimation in similar way
> > assuming 1/focal lentgh
> > and acceptable circle of confision for 35mm are tied
> > to each other.
> > however I assumed circle of confusion to be 0.03mm
> > which is apparently
> > the common wisdom for 35mm (at least in film era).
>
> .02: I think the C-of-C for the common DSLR sensors is
> smaller (.019). Read something to that effect in the
> ciurrent issue of EOS Mag. According to what was
> written there it appears that even minute vibrations
> are more amplified in digital due to other factors
> (Something to do with the fact that the individual
> sensors becoming smaller and smaller when the MPs keep
> going up and up). So apparently in high end digital,
> there may be other factors than C-of-C in introducing
> blur into an image via vibrations vis-a-vis 35mm.
>
> - Harman
__________________________________
Yahoo! Messenger
Show us what our next emoticon should look like. Join the fun.
http://www.advision.webevents.yahoo.com/emoticontest
*
****
*******
***********************************************************
* For list instructions, including unsubscribe, see:
* http://www.a1.nl/phomepag/markerink/eos_list.htm
***********************************************************
