a few poorly exposed images from this flash combination, I've attributed
this
more to not understanding the exposure conditions than to any fault with
the
equipment. Overexposure for print film by 1-2 stops rarely is an issue;
you
should be able to get good prints. The real problem is underexposure.
This is often due to the flash pooping out at far away distances. You can
lose
2-4 stops of light when you use bounce flash off a ceiling (sorry don't
know about
Lumiquest).
IMO, I don't know many metering systems that have taken away the need to at
least
examine exposure conditions. For my P645, the TTL metering measures light
off-the-film
plane with a separate sensor, and the luminosity is then controlled. It
takes an 'average'
reading of the scene. If everyone is wearing white clothes on a white
floor and ceiling,
the image is going to be underexposed. I'm surprised the P645N manual
doesn't
at least give a basic explanation on how your TTL metering works. I'll be
really
surprised if autofocus operation is necessary to use TTL metering for the
P645N!
If there is darkness surrounding the subject, I fill the frame with the
subject.
If the subject is very light, than I compensate by lowering the ISO
setting. (Yeah
this kind of cheating, but it works for TTL). Unless the subject is
pitched black,
for print film a little overexposure is acceptable.
Another important thing is to make sure that the flash can reach the
subject with
the given F-stop. IMO, Metz flashes settings are optimistic; I normally
open up at
least another stop as compared to the recommended Metz scales.
For me TTL does not address every conceivable flash situation. However,
understanding how it works and with practice, things should get better.
Some
of my best TTL images were with transparancy film (both E100SW and Velvia)
during parties and events. This was after I've taken all my print shots
and I
wanted to practice exposure analysis. After all, I'm just an amatuer who
wants
to enjoy the hobby.
Hope this helps.
G.
leigh dawson <jd...@swbell.net> wrote in article
<3528DA1D...@swbell.net>...