I Dare You!

by peter_budo

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.

Sometimes when I see a photo such as this one, I wonder why the photographer chose such a high ISO (640), despite using a tripod with a static subject.

As far as I know, a lower ISO means less noise, and vice versa. When shooting in low light with a tripod, it should be possible to just increase the shutter time. I believe the photo can be taken at a lower ISO without any major problems. So is there any other reason?

[EDIT]: For those who can not open the link, here is the photo: Image Equipment and settings: Canon EOS 5D, 24-105mm, 1/6sec at f/22, ISO 640, tripod, ND grad filter

share|improve this question
The other question to ask is why they're using f/22. Perhaps the answer is that you can still make great photos even if you don't do what is technically best. –  Philip Kendall Jul 23 '14 at 13:53
Such a high ISO and 640 do not belong in the same sentence on modern cameras. That's a mid range ISO bordering on low, that said, the question is still valid as to why not to use a slower shutter speed, even if noise is not a real issue at ISO 640. –  AJ Henderson Jul 23 '14 at 15:00
@PhilipKendall I guess because they read classical photography books and f22 on large format cameras is used quite often for landscapes. –  Andrew Jul 24 '14 at 7:31
I really think you're reading too much into the settings. This may be something as simple as that he had the camera set to ISO 640 for something else and just didn't change it. –  Mike Jul 24 '14 at 7:40
Include the photo inline please? The link is down. –  Phil Jul 24 '14 at 16:43

12 Answers 12

The mountain and the valley obviously are static -- even more from that distance. The clouds, however, move. If you chose a low ISO value, e.g., in the range of 50 to 100, the exposure time might be enough to get washy/faded/blurred clouds.

If I calculated it correctly, an ISO value of 100 with the other settings (exluding shutter speed) staying the same would result in a shutter speed of 2 seconds.

share|improve this answer
Plus, of course, using a tripod always helps with sharpness. –  Andy Blankertz Jul 23 '14 at 15:07
Yes, as anyone who has shot such scenes with these type of clouds can attest, they usually move rather quickly. Multi-shot HDR is out of the question unless you mask it and only use the sky from one of the multiple exposures. –  Michael Clark Jul 25 '14 at 3:00

I can't think of any technical reason for this to be the case. Even assuming that he used the lens at 105mm, if he was more than 387 ft from the closest object in the frame, he could have focused at 750 feet and had everything in focus even at f/5.6, so f/22 was completely unnecessary unless intending to get the shutter speed longer. A faster shutter speed or lower ISO could have easily been achieved. That said, my best guess is that the photographer may have been shooting in aperture priority mode and had the aperture cranked to max to help deal with the amount of light coming in from sunlight or possibly from a misunderstanding of how depth of field works.

This really demonstrates that taking decent photos is far more about the ability to recognize and take a good photo than it is about technical competency. It is helpful to have both, but you are far better off to be artistically skilled and technically bordering on illiterate than technically proficient but having no idea what makes a good artistic shot.

share|improve this answer
@Thomasrutter - huh? My point was that he could have used a lower aperture (ie, faster) and it wouldn't have mattered since he's shooting an landscape and it is well beyond the hyperfocal point. –  AJ Henderson Jul 25 '14 at 2:19
This is why DOF calculators are completely misleading. Depth of field depends on the distance at which you view the print/image. There's simply no way for you to know how large the photographer may have printed the image or where he intends it to be viewed from. The figures you quote assume a circle of confusion of 0.03mm, which is about five times the size of a 5D mkIII pixel! If you want to avoid a 5 pixel blur then f/4 wont cut it. In fact using the 5D pixel size as the CoC, according to the calculator to get everything from 147 feet to infinity in focus you need to stop down to... f/20! –  Matt Grum Jul 25 '14 at 11:44
@MattGrum - ok, so that points out that I should have double checked the DoF calculator I used online quickly (which was apparently crap, because I configured it so it should have given accurate numbers for a 5D.) It doesn't really change the main point though because the closest point in the image is still FAR more than 147 feet away. –  AJ Henderson Jul 25 '14 at 14:06
@MattGrum - I have updated the answer now to use a better DoF calculator that gives pixel accurate measures. I used the CoC for pixel accuracy on the original 5D which is the model listed as used. Thanks for pointing out the error. –  AJ Henderson Jul 25 '14 at 14:14

The photographer chose a slightly higher ISO to compensate for using f/22 aperture (small opening = less light). You may ask, so why not make the shutter speed longer? The shutter speed was set to 1/6th of a second, rather than say 10 seconds, so that the image is nice and sharp, it may have been windy and might have shaken the tripod even just a little bit.

Judging from the colors of the clouds it might also have been dawn/dusk. It may not look like it in the picture but that's exactly why the ISO was bumped up.

share|improve this answer
Good answer. I forgot about that it might have been windy and the tripod shaky. –  PattaFeuFeu Jul 24 '14 at 6:12
I'd have thought that 1/6 second is not going to help at all with a shaky tripod (whether vibration or wind). –  thomasrutter Jul 25 '14 at 2:20
It depends how sturdy the photographer's tripod is. 1/6th will avoid little shakes and vibrations compared to using say 10 seconds of shutter speed. –  Jeff B Jul 25 '14 at 5:36

I can't buy the clouds moving being an issue here. However, I can think of one reason that would be invisible: There are mobile objects near the photographer that sometimes get in the frame. He shot quickly to avoid them. He might have shot several, this is just the one that actually worked.

I do think it's more likely an oops, though.

share|improve this answer
like ufos maybe? –  Octopus Jul 24 '14 at 21:39
@Octopus I was thinking things like birds or insects. –  Loren Pechtel Jul 25 '14 at 1:45
Yeah, because if there were UFOs flying around there's NO WAY he would want them in his photo, right? –  Michael Clark Jul 25 '14 at 3:02

Disclaimer: this answer is highly dependent on equipment, firmware, etc. and may actually be conjecture.

Another possibility is that the digital camera is not optimized for the lowest ISO values. If you look at the internals of a camera some sensors do not support "50 ISO" and instead the camera firmware shoots at about 160 - 200 and pulls the exposure down with either amplifiers or software.

I remember hearing, anecdotally, that some earlier Canon sensors behaved this way and the lowest noise was achieved at ISO 160 - or multiples of 160. The in-between values are either amped up or pulled down in software providing less than perfect results. e.g.

100 is 160 pulled down 200 is 160 amped up 320 is truly captured at the sensor

Think of it as the opposite of the "HIGH" ISO modes that offer 25,600 ISO, etc. The high modes are actually shooting at 6400 and then amplifying the signal in a chip or software to give you the crazy values - but quality suffers.

share|improve this answer
There seems to be a huge amount of argument online about what the native ISOs of Canon cameras are and whether it actually makes any difference. Some people seem to be claiming that 100, 200, 400, ... are the native ISOs and you should use those; some that 160, 320, 640 are native and you should use those; some that 100, 200, 400 are native but you should use 160, 320, 640. What a mess. –  David Richerby Jul 23 '14 at 21:07
I completely agree - that's why the disclaimer at the top. –  infamouse Jul 24 '14 at 20:26
There's no controversy. Base ISO for all Canon EOS DSLRs for the entire digital era has been either ISO 200 or ISO 100. The way recent Canon EOS DSLRs shoot at the +1/3 stop and -1/3 stop (160, 320, 640, 1250, etc.) ISO settings means that when shooting at ISO 160 the sensor is set at ISO 200 and the exposure is pushed by 1/3 stop. When the RAW data is demosaiced the image is pulled 1/3 stop the other way. In effect you are trading a little less noise for a little less dynamic range, just as "shooting to the right" does. –  Michael Clark Jul 25 '14 at 3:06
With the +1/3 stop ISO settings (125, 250, 500, 1000, etc.) you are pulling exposure 1/3 stop when shooting and then pushing it 1/3 stop when demosaicing. This increases the noise as well as risks losing the highlights. –  Michael Clark Jul 25 '14 at 3:08

A very narrow aperture such as f22, which allows for an almost infinite Depth of Field, lets in very little light. ISO 640 may have been required to ensure the shutter speed stayed at a reasonable level, if the photographer didn't want to risk camera shake from wind or cloud blur.

share|improve this answer
But f/22 also causes diffraction effects, reducing sharpness. –  David Richerby Jul 26 '14 at 8:30
And those diffraction effects are almost universally overstated for general photography. There is a huge amount of absolute crap based on test targets that's floated around as Photographic Truth that has very little bearing on real photography. –  user28116 Aug 4 '14 at 13:45

F/22 suggests to me that the photographer first took some pictures with nearby objects in the foreground and then forgot (or didn't bother) to increase the aperture for this particular shot. The ISO setting of 640 is then quite appropriate, you can go lower with a longer exposue time, but as the other answerers point out that can cause problems if the clouds move visibly (they become unsharp). Also, even if the motion of the clouds is not visible directly, this can still cause fluctuations in the illuminaion of the landscape. The shadows that the clouds cast can move accross the landscape by a large amount in a few seconds causing parts of the image to get under or overexposed.

share|improve this answer

The only technical reason I can see in this particular case is that there might be some traffic on that road that you don't want in your shot. Thus the faster shutter speed was required to catch it between vehicles.

Not that I think this was actually the reason, since the timeframe for ISO 100 is still pretty short.

share|improve this answer
I don't think that really applies, because any car would be tiny in the photo, even at quite a large print size. Furthermore, I suspect that an exposure of 1/6s would actually make any traffic look worse. A car travelling at 30mph covers about 2m in 1/6s, which is going to give horrible blur; in a second, though, it would travel about 13m, i.e., about three times its own length. That might not even be visible, if the car was a dull colour and had no lights on. –  David Richerby Jul 23 '14 at 20:50

It could have been very windy. =0)

share|improve this answer
I don't think this should have been downvoted but it does need some expanding. I believe you mean that the tripod might not be enough support in the extreme conditions to provide a steady shot. Correct? –  infamouse Jul 23 '14 at 20:26
Or maybe that the wind would have moved the clouds too much and created a blurry shot. Given how vague this is, I think a downvote is completely justified. –  David Richerby Jul 23 '14 at 20:52
OK but why do you think the tripod could support the camera for a 1/6s exposure at ISO-640 but not a 1s exposure at ISO-100? Both of those exposure times offer plenty of time for the wind to move the tripod. –  David Richerby Jul 23 '14 at 21:50
@DavidRicherby a factor of 6 can make a difference of 6 times more. It looks like taken from the top of a mountain, so the likelihood of wind is high. And even more if there are sudden unpredictable gusts. –  Davidmh Jul 24 '14 at 22:21
@robjcrowe The point isn't necessarily to give information just to the person asking the question, but also to anyone else (potentially less technical) who finds this question in the future. –  Philip Kendall Jul 25 '14 at 7:41

One (remote) possibility, based on a different camera...

My Fujifilm X-Pro1 has a "dynamic range" setting that lets me choose between 100%, 200%, and 400%. Higher settings are supposed to extract more shadow and highlight detail (almost like a simulated HDR) in the out of camera JPEgs.

BUT the 200 and 400 settings are only available at higher than minimum ISO settings. DR 200 becomes available at ISO 400, and DR 400 becomes available somewhere higher than that.

Maybe this guy's camera has a similar feature (or some other feature) that's only enabled at higher ISOs?

share|improve this answer
I'm not aware of any such feature on the Canon 5D. Most people who own a camera at that end of the scale are working with RAW files and don't care what's in the JPEG. –  David Richerby Jul 24 '14 at 7:21
@DavidRicherby Highlight Tone Priority is almost exactly the equivalent of the DR settings on the Fuji - but that only bumps the minimum ISO to 200. In general, I'd agree with you about RAW, but this photo was taken at f/22 :-) –  Philip Kendall Jul 25 '14 at 7:40

Perhaps the photographer believes that ISO 640 can be LESS noisy than ISO 100.

I'm not necessarily supporting that viewpoint, just stating that there is one.

share|improve this answer
What does this add to the answer from @infamouse, except for another link? –  David Richerby Jul 24 '14 at 22:30

1) Be simple. (s)he forgot to switch to the lower ISO. ). For example, (s)he wanted to catch a bird before.

2) But I heard one story from an older photographer (?) about the long exposure (sensor reheating during the long exposure): sometimes it's better using short high ISO shot vs. long exposure shot to avoid noise.

The best solution to make two shorts and to compare results... :)

So, I think - 1)

share|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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