Page 1 of 1
D200 odd image results
Posted:
Sun Oct 08, 2006 8:03 pm
by cawdor
Hi all,
I got to play around with the D200 this weekend and some of the images have turned out overexposed and the colours were "off".
The settings were the same as in my D70, jpg optimisation is set to "normal" in both. ISO is NOT on auto. Matrix metering, spot focus. The first sample shot was taken with my D70 last weekend, the second shot is by the D200 yesterday. Same weather, same location, same breed of dog
It's not a one off shot, the entire series of shots all look similar to this one.
Any input is appreciated!
D70 image:
D200 image:
Posted:
Sun Oct 08, 2006 8:13 pm
by PiroStitch
Were the shutter speeds and aperture settings the same as well? My guess is that you're comparing an apple with a banana (the d200 being the banana as they're more expensive nowadays
)
If you're leaving it to the camera to decide the exposure, there are a number of things you have to factor in and as far as I know, the D200 and D70 have different algorithms. Also are you sure the exposure compensation wasn't accidentally adjusted at the time as well?
Posted:
Sun Oct 08, 2006 10:15 pm
by cawdor
Shutter speeds were the same (1/1000), aperture was 6.3 on the D70 and 5.0 on the D200.
Exposure compensation isn't mapped to the front command dial on the D200, so I couldn't have adjusted it by mistake.
Posted:
Mon Oct 09, 2006 12:26 am
by Yi-P
The two cameras have different WB algorithms as well, you may want to check if the WB on the cameras.
Also, f6.3 and f5.0 is about 1/3 stop difference.
Posted:
Mon Oct 09, 2006 9:24 am
by gstark
Tim,
The embedded EXIF doesn't give us the ISO for the D70 image, b ut given the similar exposure settings in place, it's probably somewhat safe to suggest that it was set to200, which is the stated value on the D200 image.
Your in-camera contrast settings were different - the D70 was set to hard - and that may have an effect on images.
Colour space was the same, but, you used auto white balance.
While Yi-P has correctly pointed out that the cameras have different WB algorithms, the fact that auto wb was in use means that any comparisons are, really, pointless. You simply don't really know what the camera was doing at the time.
Using auto wb can bring differences such as these to similar images shot with same camera, and thus the results that you're asking about here are not too surprising to me.
This sort of test can only be meaningful when you take control of the various shooting parameters in place, and thus you need to move away from the various PHD settings and deliberately set up the values so that you are in control.
Take both cameras to the park next time the dog takes you for a walk, set your wb in each camera to the same setting, set the same ISO, take a manual reading of the exposure and set both cameras to the same shutter speed and aperture values, check that the same colour space is in place on both cameras, and check that your image settings (contrast, sharpening, etc) are the same and that there are no custom curves in use, and then start shooting.
In raw.
Using the same lens on both bodies.
Let's see what those image comparisons look like.
Posted:
Mon Oct 09, 2006 9:52 am
by Justin
Posted:
Mon Oct 09, 2006 10:26 am
by Greg B
Tim
Just a general comment. I was surprised at how different the D200 is compared to the D70. I knew there would be differences based on the specs etc, but I reckon it is a significantly different beast (all good I might add). There are a lot of variables, and a lot of choices.
I must say the second shot is very crisp.
cheers
Posted:
Mon Oct 09, 2006 10:39 am
by Justin
Thanks to Gary's investigation we know there was high-contrast in the D70, you can see this from the detail in the dog's coat - #2 appears to have a lot more detail. Also as there are no blown highlights in either that I can see (although the paws may be high in #1), all of the information is there.
I suspect the histogram for #3 was further to the right (work pc so cant tell right now)
I read somewhere that the digital camera has five bands of recording, dark to light and that 50% of the image information is recorded in the top (lightest) band (20%) and it halves at each band. Thus the 'shoot to the right' theory.
e.g.
band 1: 8192
band 2: 4096
Band 3: 2048
band 4: 1024
band 5: 512 (shadow)
Does anyone else know this theory better than I?
Posted:
Mon Oct 09, 2006 11:40 am
by cawdor
Thanks for your replies guys, very helpful. I haven't really done anything with WB, I always have it on auto - maybe it's time to set it manually
Posted:
Mon Oct 09, 2006 1:52 pm
by obzelite
the sun wasnt coming thru the smoke from the calm burnoff was it?
Dont know about the canine but our place had a smoke haze most of the day.
Posted:
Mon Oct 09, 2006 8:32 pm
by cawdor
No it wasn't the smoke, the sky was clear. I think it may have been the auto WB that gave the overexposed result. I'll have a play around with setting it manually and see if the results are different.