About light... again
NOTE: let me know if there are errors; I'm not so expert in English nor in photography
A new group for this: Ipernity Academy.
In a previous article I said photography is about painting with light.
I started talking about the various parameters we usually can control in a P&S camera. Now it's time to talk about the other parameters you usually can control on a DSLR camera.
Given all the other settings, in manual mode we must play with the combination of ISO sensitivity, aperture and shutter speed.
I try to explain these three parameters with an example.
Imagine our sensor as a discotheque. Outside there are people waiting for their occasion to enter and there are guardians controlling that "personae non gratae" don't come in. Imagine all the people moving at the same speed, in random directions, farther or nearer to the entry door. The guardians open the door for a certain time period and start monitoring for undesired people. Our door is the shutter: the more you open it (i.e. the smaller the F-number), more people can enter. The time the door is open is our shutter speed: the longer you open it (i.e. the smaller your shutter speed), more people can enter. The attention of the guardians is our ISO sensitivity: less they are careful (i.e. the higher is the ISO number), more undesired people will enter (EDIT: and, clearly, people will enter faster). The people are the photons of light, the undesired people are noise, the quantity of people in the disco is exposure.
If your combination of shutter speed, shutter sperture and ISO sensitivity let too many photons hit the sensor, you'll have an overexposed shot; if your combination of shutter speed, shutter sperture and ISO sensitivity let too few photons hit the sensor, you'll have an underexposed shot.
You can obtain the right exposure with many combinations of the three parameters: if you double one of the three, you must divide by two one of the others (or their product).
But you must consider that those combinations give you the same exposure but not the same shot.
If you open very wide the door (high aperture, low F-number), only people reletively near to the door will be able to enter; so you will have a little focused zone with a blurred background; if you use a little aperture and open for more time, people farther from the door will have their occasion to enter (do you remember? all the people move at the same speed in random directions!): you will have a deeper focused zone. The depth of the focused zone is called depth of field (or DOF): so the smaller the aperture (high F-number), the deeper the focused zone.
Last information about focus: when you focus on a subject, the focused zone is divided one third in front of it and two thirds behind it.
About noise: when you use an high sensitivity film (or sensor setting), you can take shot with less light; the shot you'll obtain will have more noise; this noise will appear as grain in b&w shots and as colour imperfections in colour shots. There are software filters that try to correct this, but remember the old GIGO rule: garbage in, garbage out --> the mean quality of what you obtain out of the filter will be the same of what you trough in --> some shots would be corrected and some shots will look worse. Tipically those filters use some kind of interpolation, giving good results on smooth situations and bad results where there are great variation of light or colour. I found numerous sources that invite us to beware of the long exposure noise corrections integrated in some DSLR cameras. Stop with the bad news. Go for the good news: noise is a statistical entity. So, if you take different shots of the same scene, the noisy pixel will be distributed randomly. How can we use this information? If we take a shot of a dark static scene, take more than one shot and combine them as layers in GIMP or Photoshop, using the Addition mode: noise will be less evident.