Steve & Alicia Goss wrote:
>
> Hi, Y'all-
>
Hi Steve-
> Here are two pictures from my vacation. Both are from the same OM4Ti camera,
> and both on Kodachrome 64, from the same roll. Both were scanned by Dwayne's
> photo when they were processed.
> I don't remember which lens I was using for 01-025, but 01-039 was done with
> a 24mm f2.8 at f16 and a circular polarizer, to slow down the shutter as
> much as possible.
> 01-025 was hand held, 01-039 was done on a tripod, with the self timer.
> No flash on 01-025, but I was hand holding a pair of flashes on 01-039. They
> were pointed toward the building to try to lighten it up a bit. The building
> looked quite dark compared to the rest of the scene.
>
> The main question is what went wrong on 01-039? 01-025 has a sharp scan, but
> 01-039 just looks bad. 01-039 does look fairly sharp when it is projected.
>
First, 039 doesn't look particularly unsharp overall to me. The rocks in
the foreground are a bit soft, but that could easily be DOF. F16 or not,
that's a pretty deep subject. Assuming it is soft, there are only a
limited number of possible causes.
- Lens. Gary's tests show the 24/2.8 to hold up well at f16. Assuming
yours is in as good shape as the ones he tested, the lens seems not to
be the problem.
- Filter. Gary proved that some filters that look good aren't. Even if
yours is fine in that sense, a polarizer introduces 4 additional air to
glass surfaces. 'Twere me, I'd do a test with that lens with and without
filter.
- Motion blurring. I know it was on a tripod, but motion blurring is
still possible if the tripod isn't sturdy and "dead" enough, if there is
any breeze, even if the falls cause vibration in the ground. I've also
on occasion found motion blurring in foliage from light, unobvious
breezes. Falls do cause their own wind.
- Scan. Just because it's Dwayne's doesn't mean it's good. There are a
number of other problems that seem likely to be from the scan.
> Independence CO, a ghost town preserved by people from Aspen, CO.
> http://www.speakeasy.net/~stevegoss/01-025.jpg
> <http://www.speakeasy.net/%7Estevegoss/01-025.jpg> [1]
>
This scan compresses a lot of highlight and shadow detail into
invisibility. Although there are some lost highlights which are probably
from the slide, there is more cloud detail than might be apparent. Is
the color true to the slide? It doesn't look very "Kodachromy" to me. WB
based on the clouds as neutral make it look more like what I expect.
<http://www.moosemystic.net/Gallery/Others/Goss/Cabin.htm>
> The "Crystal Mill", actually a power plant for the mill, which was just to
> the right.
> http://www.speakeasy.net/~stevegoss/01-039.jpg
> <http://www.speakeasy.net/%7Estevegoss/01-039.jpg> [2]
>
This scan seems to me to be quite poor:
- Where did all that purple in the shadows come from? As with the first
scan, I suspect that auto WB was used in the scans - and worked poorly
for this shot. Auto WB has to assume that the average of the image is
neutral gray. When that isn't the case, it introduces unnatural color
casts. I've assumed the rocks in shadow are gray. Again, I can't know
how it compares to the slide, but it certainly looks more like
Kodachrome to me.
- There is a lot of grain/noise in the shadows. How much of it is from
the film and how much from inadequate DMax in the scanning process , I
don't know. I wonder is it is contributing to the sense of unsharpness,
not only in the obvious building shadows, but in the rocks in the
foreground.
- As with the first scan, highlights and shadow are over compressed.
Most of the sky is probably gone on film, but spreading the highlights
down does a lot for the things silhouetted against the sky and brings a
hint of cloud detail. There is a fair amount of shadow detail to be
coaxed out. Noise needs to be addressed first, though.
- As with any scan, sharpening is required for best results. It can't be
done here before noise reduction without unfortunate results. Looking at
it again, I should probably back it off a bit in the foreground rocks.
<http://www.moosemystic.net/Gallery/Others/Goss/Mill.htm>
Moose
==============================================
List usage info: http://www.zuikoholic.com
List nannies: olympusadmin@xxxxxxxxxx
==============================================
|
