Summer Start

by VonSchnauzer

submit your photo


Hall of Fame
View past winners from this year

Please participate in Meta
and help us grow.

Take the 2-minute tour ×
Photography Stack Exchange is a question and answer site for professional, enthusiast and amateur photographers. It's 100% free, no registration required.

Possible Duplicate:
Are there any situations in which it makes sense to raise the ISO in bright daylight?

I was looking at this image (from From Reuters blog http://blogs.reuters.com/fullfocus/2012/11/30/best-photos-of-the-year-2012/#a=3):

Is there a reason why the photographer choose to shoot at ISO 800, f2.8, 1/500? Couldn't he shoot at ISO 200 1/125 and get the same result with less noise? (I'm not saying the image has a lot noise, it doesn't have, just curious about the settings).

share|improve this question

marked as duplicate by dpollitt, John Cavan, Imre, mattdm, whuber Dec 5 '12 at 17:45

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.

    
I used my 1D mkIV as high as ISO6400 and it produced usable images though with a lot of noise and loss of dynamic range. I certainly wouldn't say the mkIV can "comfortably" shoot up to ISO 25,600 - you can use it in emergency but dynamic range and detail will be severely compromised. It would definitely have been preferable in terms of noise to shoot at 125s or 250s but the second half of your comment hits the nail on the head, it was a spur of the moment thing... –  Matt Grum Dec 3 '12 at 13:59
    
@MattGrum - I don't own the 1D MkIV, but dpreview noted in their review/final conclusion that 25,600 is "perfectly usable" and "Very good high ISO performance even at 25,600". See: dpreview.com/reviews/canoneos1dmarkIV/32 Even so - ISO 800 is again, 5 stops below that, so it's moot what exactly the max acceptable ISO is, 800 is well below it by almost any standard. –  dpollitt Dec 3 '12 at 14:04
    
Another thing to note - is that even Reuters gets some news wrong. If you browse through some of the images they selected for photos of the year and look at the equipment noted, I believe some errors exist in the equipment reported. Not that ISO 800 here is unbelievable, but I'd take any of the exact EXIF reported with a grain of salt. –  dpollitt Dec 3 '12 at 14:07
2  
running the numbers gives an EV of 9 for this shot, which means the light levels were pretty low, so this was very early morning, or late at night, so ISO800 isn't that far out. –  Matt Grum Dec 3 '12 at 14:20

3 Answers 3

up vote 8 down vote accepted

I used to think the same way, but then I realised how slow ~1/100s shutter really is. In my work as a machine vision engineer I am used to thinking of the shutter as milliseconds, rather than as fractions and for dynamic subjects (relating to its speed) general walking speed has to be faster than 10ms (1/100!), so with "fast" subjects you need only a few milliseconds (1/500 and faster). So if you have a walking speed subject with moving clothes and the legs moving faster than the subject itself and add camera hand shakes, you fast end up at needing over 1/200.

Here you see a fast 28mm 1.8 lens outdoor with a fast subject 1/125 low ISO vs 1/600 high ISO:

Fast subject

share|improve this answer

The standard answer most time this is asked is don't assume everything was done for a reason, the camera may have already been set to ISO 800 and the photographer might not have had time to change it.


However, this image looks like bright daylight and so ISO 800 seems way off, but in this case the image brightness is a bit misleading, 1/500s f/2.8 ISO 800 corresponds to an ISO 100 exposure value (EV) of 9 - about the same you get a night time sports venues! This was likely shot very early in the morning, or late at night.

Looking at the other settings we note:

  • 1.3x crop camera was used
  • focal length was 135mm

Applying the standard 1/focal-length rule for shutter speed and accounting for the crop factor would recommend shooting at 1/175. I prefer a 100% margin for safety when dealing with camera shake (resolution has grown massively since the rule was popularized) which would give you 1/350s, not far off what the photographer used.

So given the image was shot in fairly low light it looks like the settings were about right for those conditions (not taking into account image stabilisation if available).

share|improve this answer
    
hows the formula to translate setting into EV? –  Michael Nielsen Dec 3 '12 at 14:29
    
@Michael Nielsen Here's a table of values: en.wikipedia.org/wiki/… these relate to ISO100, to convert to a different ISO, subtract 1 for every stop above ISO100 –  Matt Grum Dec 3 '12 at 14:37
    
Pretty cool . I always wondered how people would start to figure out the exposure without the modern cameras that give a general indication and you had to go to your lab and develop the film to see how it went :) –  Michael Nielsen Dec 3 '12 at 15:16
    
Hmm, it doesnt add up. Im pretty sure I used ~F2.8 in the dog shots, and it was an overcast day, EV12, and I used ISO 200 vs ISO 800 to get ~1/125 and ~1/600. And the table suggest I could use ISO100 1/500s, ISO200 1/1000s, ISO800 1/4000s. Unless Danish weather is much darker than average. I'll check the exif at home. –  Michael Nielsen Dec 3 '12 at 15:38
    
Table 2 must be off.. my exposure is for EV9, which is supposed to be after sunset/ night sports, and it was daylight, cloudy, at noon. –  Michael Nielsen Dec 3 '12 at 19:14

Maybe to avoid some blur on the dog, the woman or her clothes due to wind for example. I guess any blur on the trees wouldn't be noticeable with this DoF.

The relatively high ISO allowed for a much faster shutter speed, which makes everything in the picture a bit more "frozen".

share|improve this answer

Not the answer you're looking for? Browse other questions tagged or ask your own question.