Christian wrote:
> I'm with Godders on this one... Maybe something to do with the type of
> window. For example, argon gas between the panes or a UV (or some other
> insulating filter) on the glass could cause weird colors that may
> otherwise not be noticed by your eyes. Better to test outdoors in
> "real" natural light.
>
Good idea, but I just went outside and reproduced the colors outside.
I went outside and tested the k10d with a fa50f1.4. I shot a brickwall,
natural tone jpegs. This was to correct some potential problems as
posed by the list yesterday, that is, problems caused by light going
through window material and shooting of a white object.
First I tested at different speeds, in groups of three, normal, 1 stop
under, 2 stops under. Let the lens open up to f4 or f4.5 and varied the
shutter speed.
Pic's 1-3: 1/8 sec, f4, 100 speed. normal exposure.
Underexposed by slowing the shutter by 1 and 2 stops. No color
shifts.
Pic's 4-6: Repeated of above, except start at 1/15s, f4, 400
speed. No shifting for any of the 3.
Pic's 7-9: Similar test at 800 speed. No problems.
Pic's 10-12: Repeat for 1/30s, f4.5, 1600 speed. Looks okay.
Then I decided to try things that were near what I did yesterday. That
is, high sensitivity, long exposure (greater than 1/4s) and small f
stops (f22).
Pic's 13-15: Repeat situation above starting at .5s, f22, 1600
speed. There it is again, green and magenta. Specially so in
the first (normal exposure) and the 1 stop under exposure.
Pic's 16-18: 1 sec f22, 800 speed. Same problem as above for
first two pictures especially, but not as strongly as the one's
at 1600.
Then I decided to see if long exposure and small f stops will produce
the color shifting at other sensitivities.
Pic's 19: 2 sec, f22, 400 speed. Still there, but not
strongly.
Pic 20: 4 sec, f22, still there, although getting to the point
where it is difficult to tell.
Pic 21: 2 sec, f22, 200 speed. Still there. Very Faint.
So, my conclusions for MY CAMERA (do not know about other k10d's):
1. Color shift will be produced at any sensitivity, during
underexposure or normal exposure, if both the shutter speed is long and
the f stop is small.
2. At conditions as described in number 1, color shifting is maximized
at higher sensitivity and a stop of underexposure. Normal exposures at
lower sensitivities minimize the effect.
3. Camera is capable of images without any color shifting at all light
sensitivities, even with underexposure if conditions in number 1 are not
met.
I am beginning to think it is just a limit of the technology. I would
bet that others could reproduce this result. On the other hand, I am
surprised that someone else has not bumped into this problem. I am
convinced that it is not the light source or lens. Could just be the
unusual conditions or just my camera. Cannot say unless some others are
willing to give it a go.
Godfrey (or anyone else), if you are willing, could you try a long
exposure, small f stop at a high sensitivity?
I posted 6 of these on flickr. 10,11,12 look fine (shot as described
above), 13,14,15 show the shifting (shot as described above). If
someone would like me to post any others, I could do that.
http://www.flickr.com/photos/[EMAIL
PROTECTED]/?savedsettings=375990192#photo375990192
Contacted pentaxusa - they could not access the flickr pictures.
Suggested I send the camera back in. Ick.
I am still interested in any speculation and possible attempts of others
to duplicate this.
Thanks for your time and patience,
Dave
x
--
PDML Pentax-Discuss Mail List
[email protected]
http://pdml.net/mailman/listinfo/pdml_pdml.net
