Okay. Thanks a lot. I took the camera
outside again and did some more tests in
the snow. I asked about TIFF files
because I've been using TIFF for
photomicrography. We need big files for
big prints and the results seem good.
This is what I found. The RAW and TIFF
snow scenes were all very good without
exception. The histograms showed that
the five stops the camera is capable of
were to be found between 1 and about
160. Spreading them using levels gave a
close match to what it actually looks
like outside. I picked white from the
lightest area I could find and black
from the deepest shadow.* The result is
snow that has no blown out highlights
(which detract from the otherwise
wonderful picture of terns I saw the
other day). This is something I've not
been able to handle well with film. I
always ended with some blown out whites
in winter.
The RAW converter is a plug-in for
Photoshop 7 that I found on the web some
time back and saved for the day I'd get
an *ist D. I'm working with a rather
superficial idea of what needs attention
in digital and I need a lot of practice.
In other words I don't know what the
hell I'm doing. I could have set the
white level on snow I suppose, but left
it at automatic. I didn't want to make
too many changes at once. Oh yes ... I
noticed something else which would
confirm what you say above. The output
from the converted RAW files in
Photoshop are sharper than the TIFF
images. Also the Pentax Browser software
is better than Windows for uploading
images from the cards.
* This also reduces the very slight blue
cast. Using auto color makes them a tiny
bit too red. But the result is almost
the same and by this stage highly
subjective anyway. Manually altering the
color with the sliders can give a result
indistinguishable from the levels
dropper manipulation. In 'snow only'
pictures the camera tries to put
everything in at about 109 -- a
beautiful bell curve. I guess that's
where the *ist D thinks Neutral Gray
(18%) lies and that would be about right
I suppose -- with the white level set to
auto.
When we have more fresh snow I'll try
setting the white level on a clear flat
area and see what happens. I don't have
a Kodak Gray Card to check the above
theory.
Don
Jan van Wijk wrote:
Hi Don,
On Sat, 03 Dec 2005 09:42:54 +0200, Don Williams wrote:
I need more help and the quickest way to
get it is to ask PDML.
1. Is the TIFF file output from the *ist
D a real RAW file -- but with TIFF format?
No, it is NOT raw anymore, the conversion from bayer-pattern
to full RGB info for each pixel has already been done ...
And the TIFF has just 8 bits per channel I think,
so you have the same potential problems as
with JPG regarding lattitude ...
2. Does the 'in camera' sharpening work
with TIFF? I suppose it does not with RAW?
I would think so, but never use TIFF, don't know for sure ...
The camera does not not add it in RAW.
However, most RAW-converting algorithms include some
sharpening by design, usually configurable ...
I would say that if the contrast ratio in your images (microscope ?)
is 5 stops or less, and the lighning is constant/predictable,
you will see very little difference between RAW and TIFF
(or even maximum quality JPG)
There may be very subtle quality differences since the RAW converter
software may be better than in-camera, but the big advantage with
RAW is with larger lattitudes and/or non-optimum exposure.
Regards, JvW
------------------------------------------------------------------
Jan van Wijk; http://www.dfsee.com/gallery
--
Dr E D F Williams
_______________________________
http://personal.inet.fi/cool/don.williams
See feature: The Cement Company from Hell
Updated: Print Gallery -- 16 11 2005
