usher99@xxxxxxx wrote:
> Sorry for all the abbreviations and acronyms--trying to save a few key
> strokes at work, doing more than I should at one time.
> FM with the link provided obviously is the fredmiranda forum.
> I did relocate another example of similar testing for the rightfully
> skeptical Chuck:
>
> http://www.northlight-images.co.uk/article_pages/cameras/canon_1ds3_noise.html
>
Quite interesting, Mike, thanks. I'm often amazed at how just following
one's instincts works out. I almost always use full stops in iso
settings. About the only other ones I can recall using on rare occasion
are 640 and 1250, both of which look good on these tests. One less thing
to worry about, full iso stops only. I like simplicity.
> I wonder how other manufacturers handle the fractional stop ISO adjustment.
My guess would be that pretty much everybody originally designed their
systems around full stops. there's almost no practical reason to need
more steps, as the other exposure variables adjust by small steps. Then
someone put in finer graduations as a feature, so everybody was obliged
to follow.
For a system where gain in set in an amplifier after the data is off the
sensor, it might be easy and reasonable in expense to change the
amplification step. For a system where any of the iso selection is
happening on the sensor chip itself, such a change could be very
expensive. Beyond that, unless or until they go to two level chip
design, every extra element on the chip takes up space otherwise used
for sensing light. I don't think anybody is using a two level design in
production yet. In theory, it would produce a big bump in performance
for CMOS.
If I recall correctly, the reason CMOS isn't being used on the tiny
chips in P&S cameras is that it has poorer sensor coverage of the chip
face than CCD, which outweighs any other factors at those sizes.
Moose
==============================================
List usage info: http://www.zuikoholic.com
List nannies: olympusadmin@xxxxxxxxxx
==============================================
|