10

I'm a little curious on how different is the dynamic range in each kind of camera.

I know the DR of the human eyes is quit high. But, how different is the DR in SLR, DSLR and P&S cameras? Or is it almost the same?

Laura
  • 597
  • 2
  • 10
  • 25
tomm89
  • 3,231
  • 3
  • 29
  • 35
  • 3
    http://photo.stackexchange.com/questions/1050/how-many-stops-can-a-digital-camera-capture – Karel Oct 30 '10 at 11:49
  • As Karel pointed out, this is a duplicate of a previous question. Barring significant objection, I will either close this thread as a duplicate, or merge it with the thread linked by Karel, as the two are quite literally duplicates. – jrista Oct 30 '10 at 18:19
  • 1
    I don't think it's a duplicate. The other question asks about DR in general; this one asks specifically for a comparison between film SLR, DSLR, and digital P&S. – Reid Oct 30 '10 at 19:23
  • @Reid: I think the difference is minimal at best. I also think it would be nice to have a go-to thread for DR questions, rather than having them be answered individually each time the question crops up. Digital in relation to film, human eye, computer screen, etc. were all asked in the original question. That covers this one as well. – jrista Oct 30 '10 at 19:38
  • @Reid Both are comparing Film & the Human eye to digital though? – Rowland Shaw Oct 30 '10 at 19:42
  • Ok, although I search before posting, I couldn't find the other question. It's up to you (moderators) whether to merge or not. – tomm89 Oct 30 '10 at 20:45
  • I think they're close but not dupes. One is: compare and contrast these technologies. The other: what is DR in a digital context. I think merging this question into the other would destroy a significant subtlety. – Reid Oct 30 '10 at 21:03
  • @tomm89 Don't take it personally, and even if it were merged, your wording would be left so that the next person would still be able to find the question when they search the same way you did -- that's actually one of the benefits of how the system works. Regarding moderation, the software that the site is built upon works at its best when the wider community makes the decision -- as you gain higher reputations, you'll gain the ability to vote to mark things as duplicates yourself. It works best when moderators only step in to mediate in case of dispute. – Rowland Shaw Oct 31 '10 at 08:00

2 Answers2

9

Yes the larger your sensor and the more light you can capture the higher your dynamic range is likely to be. To put some numbers to it:

  • Phase One P65 (54mm sensor) 13 EVs
  • Canon 5D (35mm sensor) 11.1 EVs
  • Canon Powershot G9 (6mm sensor) 10.1 Evs

Source: http://www.dxomark.com/index.php/en/Camera-Sensor

Film is often quoted as 7EVs though it's not measured in the same way as the DXO-Mark data so it's not directly comparable.

Matt Grum
  • 118,892
  • 5
  • 274
  • 436
  • 1
    I am a bit wary of the accuracy of DXO-Mark. It says my Canon 450D has 10.8 EVs worth of dynamic range. I know for a fact that is most definitely NOT the case. With ETTR and some very meticulous manual adjustments to aperture, shutter, and ISO, I can MAYBE get 9 stops worth of real-world dynamic range, but normally I get about 7-8 stops. Sometimes I think these scientific measurements mislead potential buyers, as real-world performance in a photograph as observed by the human eye rarely approaches what can be scientifically measured with non-human algorithms and devices. – jrista Oct 30 '10 at 18:13
7

In general, the larger the photosite, the higher the dynamic range.

So given that DSLR's typically have larger sensors than P&S, they typically have higher dynamic ranges.

SLR's are slightly different, in that the film type dictates dynamic range, not the camera itself.

The human eye has about 24 f-stops of dynamic range, while the camera's range is around 5-9 stops.

Alan
  • 18,717
  • 1
  • 69
  • 97
  • +1, plus one! Finally! Someone has offered up a reasonable real-world estimate for camera dynamic range. The average DSLR has about 5-9 stops of real-world performance. :D – jrista Oct 30 '10 at 18:20
  • 1
    The human eye has that many f-stops only after factoring in adjusting the pupil size ("aperture") and other neurological "processing." (In any given scene it's not much more than 7 stops [very roughly], as you can confirm by considering how indistinguishable the details of a shaded area are when you're in the bright sunlight.) It's not a valid comparison unless you want to look at the entire gamut of exposure available in the camera, which is close to 24 stops (or greater, depending on how much noise you can tolerate). – whuber Oct 30 '10 at 21:57
  • 1
    @jrista I'm still a neophyte with digital processing and so don't pretend to speak with any authority, but I have been paying attention to the EV range found in my RAW files, of which I now have several thousand. The range depends strongly on the subject and the lighting. In some cases with uniform lighting, low-contrast subjects, and telephoto lenses it's only 4-5 stops, but in others--especially landscapes exposed not to blow the highlights in sky and clouds--there often is a range of just over 10 EV that can be brought out. (EOS 550.) – whuber Oct 30 '10 at 22:01