10
$\begingroup$

I was wondering if changing the f-stop in cycles effects the brightness as in the real world like this:

enter image description here

$\endgroup$
12
$\begingroup$

No the F stop has no effect in the overall brightness of a scene like it does in real life. It is meant to affect focus and the depth of field primarily.

In real life a opening the iris by one f stop would mean that twice the light would enter the camera making things brighter. 2 stops would be 4 times the amount of light, and so on in a square progression.

That does not happen in blender.

Here's an example:

F stop 0.5: enter image description here

F stop 22:

enter image description here

Notice how the figure in the middle remains with no brightness change. In the real world the first image would have been much brighter by several magnitudes.

Here's a simulated illustration of what you would get if by every stop the intensity of the light would change: enter image description here

To control the overall exposure of a scene in cycles you have several options:

Using the Film->Eposure control.

enter image description here

enter image description here

Depending on what you are after you can also use Nodes on the compositor (Hue/saturation/Value, brightness/contrast/ gamma, or even tone mapping)

Or by messing with the color management's exposure control:

enter image description here

$\endgroup$
10
$\begingroup$

In some way... yes!

Cameras (real&virtual) stores the incoming light information in pixel. When you save your images to an output format you are basically saving a matrix of values.

There are file format that save lot of info (ex. HDR, RAW...) and files that store just what you'll need to see your image on a standard monitor.

One of the operation that may occour when choosing a lightweight format is bightness clamping. In order to just store the RGB values in the file matrix, pixel whose brightness is higher than 1, are just stored as full white (1,1,1)

enter image description here

If you examine the image above, you can make some considerations:

  • the first two picture of the first row are demonstrating that the overall brighness is not affected by the F-stop amount, the light is just "spreaded" in the first (because the background is blurry), while when the particles are more in focus, the light is more "concentrate".
  • The difference column demonstrate how the camera's F-Stop doesn't interfere with the sensor in a way to lighten up the foreground subject.
  • you can see that is almost no difference between the scene render with F-stop=1, Strenght of the background objects = 10 and the one with F-stop=1, Strenght=100. That's because brightness has been clamped as stated at the beginning. Even if we would put an Emission strenght of 10000.. we would always see a bunch of white pixels.
  • If you now examine the corrispectives with F-stop=0.1, the lower image result more bright. That's because the clamping has not occoured. As said in the first consideration light intensity has been spreaded by the bokeh effect, so in none of them pixel are white. So the value didn't need to be clamped.

Strenght=1000

In the image above, emission strenght is set to 1000, so in both pics values are clamped to white. As the bokeh effect spread the light, in a certain sense the image on the left can be said brighter.

$\endgroup$
  • 1
    $\begingroup$ This is the right answer. f-stop affects depth of field, and therefore changes what rays are recorded. Obviously this changes brightness, even if that's not its primary purpose. $\endgroup$ – imallett Sep 2 '15 at 2:28
  • 1
    $\begingroup$ @Carlo the puzzling thing to me is that the exposure on the monkey did not change by moving the f stop... in real life (and in the example image) all of the image's brightness would be affected... $\endgroup$ – cegaton Sep 20 '15 at 7:23
  • $\begingroup$ @cegaton Infact f-stop doesn't change the "quantity" of light of the scene. I would just suggest that if value clamping occurs, in some way the captured image may "looks brighter". In real life likely happens because most of light sources have an emission streght greater than 1 and clamping occours in cameras (if you are saving in jpg, png...formats). I don't think the image provided by the questioner is one of these cases, it is more probable that the exposure time has been increased to make the bokeh more visible. $\endgroup$ – Carlo Sep 20 '15 at 11:16
  • $\begingroup$ I doubt the exposure time wast changed, the result of opening the iris would be indeed what the image shows: a brighter image with a shallower depth of field and clearly defocused lights in the background. (In case you are interested that image most have been created with something like this lensbaby.com/usa/creative-aperture.php the first image was taken with the shape iris on, the secon one without, hence the exposure change) $\endgroup$ – cegaton Sep 20 '15 at 16:57
  • $\begingroup$ You are right @cegaton, I was misleaded by the caption in this page: calculus-geometry.hubpages.com/hub/…. I'll edit that part of the question. Thanks for pointing it out! I wrote this answer as a "second thought", it was not meant to be in opposite with yours which was obiuvsly the right one (Blender camera don't perfectly simulate the behaviour of a real one) $\endgroup$ – Carlo Sep 20 '15 at 18:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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