What is the best "standard ISO" for the 5D MkII?

Panopeeper wrote in post #7703852
Correct.


Well, I posted above ISO 50 is simply a 1-stop overexposed ISO 100

"Overexposed" is not the correct term, for that depends on the metering and actual exposure; for me it is a "+1 EV autobias", but "overexposed" is simpler to communicate.

Don't you all mean ISO 50 is just an UNDERexposed version of ISO 100? How would it be an overexposed version if it's technically darker at ISO 50?

constrict wrote in post #7705429
Don't you all mean ISO 50 is just an UNDERexposed version of ISO 100?

No.

What is darker? Shooting with a lower true ISO requires higher exposure (at a given metering). Achieving higher exposure is easier in bright sunlight than in a cave.

100

clupica wrote in post #7705417
I assume that Canon knows their equipment best and I would think that a default of 400 is more than an idot mode

The following graphs show the reason:

the loss of dynamic range from ISO 100 to 200 is tiny; from ISO 200 to 400 it is only about 1/4 stop

This means, that the exposure can be quartered in exchange for 1/3 stop DR (i.e. going with 400 instead of 100); this is a good deal in many cases.

On the other hand, going from 400 to 800 costs almost 1/2 stop, that needs to be considered.

IMAGE IS A REDIRECT OR MISSING! MIME changed to 'text/html' | Content warning: script

I doubt it's anything so complicated as a dynamic range curve. It's probably simply because 400 is a good medium value--low enough to keep noise at a minimum and not risk blowing past max exposure capabilities while high enough to deal with a good range of light intensity (can keep up even if light dips a bit). 100 or 200 would be in danger of low shutter speeds, while 800 and 1600 would be too high for "typical" shooting situations.

What a load of button fluff you have all made of this, how many times whilst out shooting real life can you stay at 100?...... right, and looking at images how much difference can you see between a correctly exposed 100 and 400 image? Right again, pixel peep and you will find Elvis, PRINT the images and he will disapear...... hm something like arm chair morons and such like comes to mind.

wilky

Yohan Pamudji wrote in post #7705616
I doubt it's anything so complicated as a dynamic range curve. It's probably simply because 400 is a good medium value--low enough to keep noise at a minimum and not risk blowing past max exposure capabilities while high enough to deal with a good range of light intensity (can keep up even if light dips a bit). 100 or 200 would be in danger of low shutter speeds, while 800 and 1600 would be too high for "typical" shooting situations.

that really depends. if you are talking about landscapes with a CPL i agree. otherwise it just depends. if i'm shooting @ f2.8 with a 50mm lens iso 400 is way overkill.

ed rader

wilky95 wrote in post #7705821
What a load of button fluff you have all made of this, how many times whilst out shooting real life can you stay at 100?

Almost always (landscaping, often on tripod; when walking around, then using an IS lens).

Even the only 1/3 EV difference in dynamic range counts when the DR of the scenery is high.

I would not accuse you of being a moron, but a digitally illiterate loudmouth.

Yohan Pamudji wrote in post #7705616
I doubt it's anything so complicated as a dynamic range curve

If it is too complicated for you, then simply ignore it.

Honestly, do you understand this?

Too high in what respect? Are you scared by high numbers?

Panopeeper wrote in post #7703852
"Overexposed" is not the correct term, for that depends on the metering and actual exposure; for me it is a "+1 EV autobias", but "overexposed" is simpler to communicate.

What I meant is that from what you say, the resulting image will appear to be "overexposed" by one stop relative to what the photographer was expecting unless the processing software "knows" to make a -1EV correction when displaying it.

Which means that the processing software itself has to know to make an exception to its normal processing rules for ISO 50 shots.

Is that correct?

Panopeeper wrote in post #7705860
Almost always (landscaping, often on tripod; when walking around, then using an IS lens).

Even the only 1/3 EV difference in dynamic range counts when the DR of the scenery is high.

I would not accuse you of being a moron, but a digitally illiterate loudmouth.

Oh so your one of these "I can shoot at 1/15 hand held with my IS lens and have 100% sharp images ", get real and stop being sanctimonious, and yes landscaping we would be at 100 do I carry a tripod all day long NO.

I just show me an image exposed correct that 1/3 EV is noticeable, show me a portrait shot at 100% where 1/3 EV matters.

Illiterate.. probably... loudmouth... certainly it could after all save your life one day.

wilky

Panopeeper wrote in post #7705860
Even the only 1/3 EV difference in dynamic range counts when the DR of the scenery is high.

I expect the situations in which 1/3 EV would really matter are exceedingly rare, because for that to be true it would mean that you'd have to be unwilling to give up anything both in the brightest highlights and in the darkest shadows.

I could see wanting to preserve the highlights, but have found that it's rare that preserving the shadows makes much of a difference, because the darkest shadows are where people's eyes tend not to go when looking at the shot.

Honestly, if the dynamic range means that much and you're willing to use a tripod, then you can use HDR and eliminate these concerns entirely.

kcbrown wrote in post #7706130
Which means that the processing software itself has to know to make an exception to its normal processing rules for ISO 50 shots.

Is that correct?

Well, if the processing software intends to make up for that nonsense created by Canon, then yes.

In fact, DPP adjusts the "white level" border on the histogram; thus the 1 EV higher exposed shot appears as bright (or dark) as the "normal ISO 100" shot.

ACR carries out an automatic "exposure" adjustment (it is not visible on the slider), increasing the intensity by 0.4 EV of all 5D2 shots, except for ISO 50, when the adjustment is -0.6 EV, i.e. the brightness gets turned down by 1 EV compared to the "normal ISO 100".

Of course, all this does not mitigate the effect of factual overexposure: saturated pixels don't receive the correct values by virtue of such adjustments.

kcbrown wrote in post #7706578
Honestly, if the dynamic range means that much and you're willing to use a tripod, then you can use HDR and eliminate these concerns entirely.

I could. However, there is a middle way, not as good as HDR, but it helps utilizing the entire dynamic range of the camera: very exact ETTR using a neutral setting incl. WB, leading to histograms and clipping indication (flashing on the LCD), which closely reflect the raw data. Thus, with multiple exposures I can get the "ideal" shot. This does not solve all of my DR problems, but it does help in many cases; thus I do utilize that 1/3 stop as well.

(I go even further by using a magenta filter sometimes; however, there is no high-quality filter like polar or UV filters are available for digital cameras, thus I don't use it often.)

wilky95 wrote in post #7706242
Oh so your one of these "I can shoot at 1/15 hand held with my IS lens and have 100% sharp images ", get real and stop being sanctimonious, and yes landscaping we would be at 100 do I carry a tripod all day long NO

I really don't know what you are talking about.

I have a 40D; the loss from ISO 100 to 200 is negligable (perhaps 1/6 stop), so I am using mostly ISO 200. This is, because the native ISO of the 40D is about 160.

However, the step from ISO 200 to 400 costs 1/2 stop; that's too much, so I go over ISO 200 only if I really have to.

With ISO 200, the exposures on a sunny day are f/5.6, 1/500s, or even shorter, with focal length between 17mm and 50mm.