[boomklever]

Thanks for all the suggestions so far! Actually, on my PC it is easier to compare the pictures (I shot the whole range between min and max ISO). And the 160 ISO is definetely different from all others with respect to the highlight zone.
I noticed the same in another test, so I did a third test with D-range switched off completely. Now I see a much more consistent exposure between all shots.

I will try to reproduce the setup today, with and without D-range extended. If I can reproduce this effect it may be an indication that the D-range algorithm is not working as it should.
Btw, the K-5 is a great camera! It is my first DSLR (had a Sony F707 as my main digital camera for many years, still good, but anything above iso200 is unusable).

[Jasvox]

I think this thread deserves an award for containing the largest amount of incorrect, flawed and baseless guesses on a single page.

Fake ISO that results in a loss of dynamic range?!?
Lamp flicker at 1/30 shutter speed?!?
Limited CPU because of heat/size?!?

Are you people serious?!?!?!?

There's nothing magical about tiny exposure differences from one shot to the next when you're changing parameters!!! This is a complete non-issue!!!

edit: Obviously this is because of the magical gnomes that live inside the camera and operate the light meter. They require a daily feeding of Kraft Mac & Cheese to operate at peak efficiency. Otherwise these variances become more apparent. Just push a couple of pieces of Mac & Cheese pasta through the lens mount, squished up against the mirror. Then mount the lens and let the camera rest for 2 hours while the gnomes feast.

You get rep points for that one!! Awesome.

Jason

[wombat2go]

I think this thread deserves an award for containing the largest amount of incorrect, flawed and baseless guesses on a single page.

Fake ISO that results in a loss of dynamic range?!?
Lamp flicker at 1/30 shutter speed?!?
Limited CPU because of heat/size?!?

Are you people serious?!?!?!?

.

Hi, Hound,
Try for your self
Set 1/30th (even in the USA!) and Tungsten Balance and Manual
In a dark room turn on a tungsten lamp like OP did
Set Ap
Take 5 shots of the wall in quick succession.
The difference here was MUCH more than OP showed.
With AWB, the difference was about the same as op showed.
I tried cfl in AWB and could not see so much difference I will do some more tests..
My tests were with jpg***.

[Hound Tooth]

Hi, Hound,
Try for your self
Set 1/30th (even in the USA!) and Tungsten Balance and Manual
In a dark room turn on a tungsten lamp like OP did
Set Ap
Take 5 shots of the wall in quick succession.
The difference here was MUCH more than OP showed.
With AWB, the difference was about the same as op showed.
I tried cfl in AWB and could not see so much difference I will do some more tests..
My tests were with jpg***.

Holy Jeebus... it's a disease. It must be, and I think it's catching.

Remember: it has to be Kraft... because it's the cheesiest

[enoeske]

I think this thread deserves an award for containing the largest amount of incorrect, flawed and baseless guesses on a single page.

Fake ISO that results in a loss of dynamic range?!?
Lamp flicker at 1/30 shutter speed?!?
Limited CPU because of heat/size?!?

Are you people serious?!?!?!?

There's nothing magical about tiny exposure differences from one shot to the next when you're changing parameters!!! This is a complete non-issue!!!

edit: Obviously this is because of the magical gnomes that live inside the camera and operate the light meter. They require a daily feeding of Kraft Mac & Cheese to operate at peak efficiency. Otherwise these variances become more apparent. Just push a couple of pieces of Mac & Cheese pasta through the lens mount, squished up against the mirror. Then mount the lens and let the camera rest for 2 hours while the gnomes feast.

Are you really saying that DR doesn't change with different ISO values? I don't think you know as much as you think you think.

Check out this graph of the noise vs ISO graph for the canon 40d. There very much is a change in DR for ISO levels in between native stops. While the main 100, 200, 400... stops create a nice arc, the 1/3 stops bounce up and down.

This is because they aren't native stops and are created from either pushing or pulling the exposure from one of the native ISO stops around it. For instance, ISO 125 is an underexposed ISO 100 image pushed up to 125. This results in more noise and lower dynamic range. ISO 160, however, is actually an overexposed ISO 200 shot pulled back. This results in (essentially an ettr) a cleaner image than ISO 100.

[creampuff]

OP tested using an old M series lens. Can one assume the click stops on the aperture ring directly correspond to 1/3 stops/EV differences? Not likely I think, so basically the test is meaningless imo. Test with a modern lens where aperture is controlled from the camera body and I'm pretty certain the difference isn't so obvious.

[boomklever]

I tried to repeat the test today and one thing is clear for me: when I turn off D-range (both highlight and shadow) the 160 and 200 ISO shots are 'identical'. When I turn on D-range they are not (the 160 gets some more highlight overshoot).

Btw: I did the test with an A lens, the aperture stayed fixed at 3.5 for the lower ISO shots, I only changed ISO value and the camera calculated the new speed. So the lens is a constant factor here.

What strikes me most is that I get MORE overshoot in this particular situation than with D-range turned off. Of course it is only one situation, outdoors it can be different and maybe D-range is not intended to work good at low light conditions.
Interesting...

[Hound Tooth]

Are you really saying that DR doesn't change with different ISO values?

Of course it does!!! That's NORMAL! What I'm saying is that there's no magic conspiracy of camera defect at work just because the OP is seeing the tiniest bit of exposure difference between 2 shots taken at different settings. Go back and read properly, and save you giant charts for someone who cares (and hasn't already seen such charts a bazillion times).

edit: and by the way, your chart implies that ISO 160 would give GREATER dynamic range than ISO 200, which is exactly opposite what the OP is claiming.

[Wiker]

Of course it does!!! That's NORMAL! What I'm saying is that there's no magic conspiracy of camera defect at work just because the OP is seeing the tiniest bit of exposure difference between 2 shots taken at different settings. Go back and read properly, and save you giant charts for someone who cares (and hasn't already seen such charts a bazillion times).

Why such drama? No one is talking about conspiracy.
Not two different shots, but repeated test where the OP is seeing a systematic difference tied to ISO.
And, enoeske actually had saved that graph - for me. I haven’t seen it more than 10 times, and still find it interesting.

[Hound Tooth]

Why such drama? No one is talking about conspiracy.
Not two different shots, but repeated test where the OP is seeing a systematic difference tied to ISO.

Clearly you haven't been reading some of the ludicrous theories being spewed forth by participants of this thread to explain the simplest thing in the world of photography: 1/3 stop ISO does not always equal 1/3 stop aperture does not always equal 1/3 stop shutter speed. There's nothing wrong with exploring this fact and learning from it, but going on tangents about CPU defects, A/C flicker and the way light waves bounce off walls due to the current rotation of the earth is WAY off base.

(I made up that last one)

[boomklever]

Maybe this is all not that important, but the 'strange und unexpected behaviour' only occurs when comparing 160 and 200 ISO. Not with any other ISO. And with 'unexpected' I mean that when using D-range I would expect an increase in dynamic range (that is the purpose of D-range I assume).
I do find al suggestions about 1/3 stops, A/C flicker etc interesting, but I would expect that I would observe 'that behaviour' at any ISO setting, not only 160-200.

Maybe some software programming has to do with it? I do not fully understand why the ISO range is 'limited' to 160 when turning on D-range. But I do know that if still the original lowest value of 80 ISO is used somewhere in the processing then I understand why things went wrong in my 160-200 ISO shots.

It is not a big issue for me, I was just doubting whether or not to use D-range in critical situations (concert pictures at low light conditions with spot lights).

[sewebster]

You guys are funny. Anyway, if you go back and read my post and others, you will see that the camera settings are only approximately 1/3 stops. 160 iso is not 1/3 stop different than 200. 1/30 s is not 1/3 stop different than 1/25 s. So the two situations are not the same EV, therefore an exposure difference is definitely expected.

[Hound Tooth]

You guys are funny. Anyway, if you go back and read my post and others, you will see that the camera settings are only approximately 1/3 stops. 160 iso is not 1/3 stop different than 200. 1/30 s is not 1/3 stop different than 1/25 s. So the two situations are not the same EV, therefore an exposure difference is definitely expected.

...nope, that's far too reasonable. I'm still going with the gnome thing...

...stupid gnomes...

[simico]

...nope, that's far too reasonable. I'm still going with the gnome thing...

...stupid gnomes...

Are these the same gnomes that spit on your sensor? Or is that a different bunch of gnomes...?

[Wiker]

Clearly you haven't been reading some of the ludicrous theories being spewed forth by participants of this thread to explain the simplest thing in the world of photography: 1/3 stop ISO does not always equal 1/3 stop aperture does not always equal 1/3 stop shutter speed. There's nothing wrong with exploring this fact and learning from it, but going on tangents about CPU defects, A/C flicker and the way light waves bounce off walls due to the current rotation of the earth is WAY off base.

"There's nothing wrong with exploring this fact and learning from it..."
Exactly. So why the drama then?

The OP has observed something he didn’t expect. He’s the kind of guy that wants to understand why unexpected things happen, and in this process have done some (semi-?)systematic tests to figure out how the K-5 works.
And, that 1/3 stop ISO doesn’t 100% equal 1/3 stop speed does not seem like a fully satisfying explaining to why the difference occur between ISO 160 and 200 only, and only when D-rang is turned on.

A/C flick. Maybe it’s a stupid theory, I don’t know. But if someone does knows better, then explain it.

CPU defects ??? No one said anything about CPU defect. Someone did say that "The in-camera number crunching ability is presently limited by cpu heat/size and battery power and not by cost." It’s a general statement saying that how much CPU power can be put into a camera (any camera) is not limited by cost, but by other factors.


This is a public forum where the level of expertise and knowledge varies greatly. Might be a challenge for experts to bear with the novices, but it would be a much better forum if experts (or is it "experts" ?) could explain things properly, instead of just putting things off as ridiculous.