Jump to content

Is a dissolve effect (or lap dissolve, or crossfade) different when done in camera versus when done in post?


Recommended Posts

I grew up shooting a lot of 8mm and 16mm and I recall using several cameras that allowed you to dissolve shots in the camera.
I've been wondering for a while now if a dissolve between two shots will look identical when done in Camera and when done in Post.

How does the "dissolve" option work when done in camera (in my humble experience)?
There are different ways of doing it. I've used cameras that did it in an automated way and cameras that would do it manually.
Simply put, in automated cameras, the camera will shoot Shot A and when the user switches the "Dissolve" button on, the camera's iris will progressively close for, say, 3 seconds, then rewind the film 3 seconds back.
Then the user can shoot Shot B (and I'm assuming with fully open iris, or anyways automatic exposure), thus overlapping the last 3 seconds of Shot A with Shot B - since Shot A was terminated by closing the iris, becoming darker, the image captured for Shot B will become brighter.

When done in-camera, the last 3 seconds of Shot A are going to be exposed again with Shot B at full brightness. This produces a double-exposure effect (easily replicable in film photography).
That is to say, that if during the 3 last seconds of Shot A there are areas of the shot that are very dark, these will allow Shot B to be visible in these areas very early as Shot A dissolves.
Another way to do this is by closing the aperture and rewinding, and then opening the aperture for Shot B.
I've used the first method more often because it is automated and was never able to replicate the feel of these dissolve transitions in digital.

In digital, a cross-dissolve (for example the default ones in Premiere Pro, but I'm assuming it's the same in other programs) will fade the images of Shot A from 100% opacity to 0%, and will bring Shot B to visibility the same way but in reverse. There are no possible overlapping effects when applying the dissolve, the overlap will be perfectly even even in darker areas of Shot A.

I wonder if anyone could clear this up for me. Maybe the cameras with automatic Dissolve features would open the iris of Shot B from 0 to 100 but in many cases I would shut down the camera and shoot Shot B in different circumstances so maybe the camera wouldn't "remember" to dissolve "in" for shot B. But whenever I see cross dissolves that I know have been edited digitally, they never have the feeling of a double-exposure image, where dark areas of the image show the second image more clearly (in-camera dissolve resembling more of a blending mode...)

Link to comment
Share on other sites

  • Premium Member

Interesting question -- I wonder if a test chart could show this, maybe a field that transitions from black to white with grey in the middle shot for the A side and flipped so that it goes from white to black on the B-side, then dissolved together you could look at the halfway mark and compare a digital dissolve to an in-camera one...

Link to comment
Share on other sites

  • Premium Member

The ramp in exposure on both scenes of the photochemical dissolve will likely be fairly nonlinear, unless an awful lot of work has been done to match the iris ramp rate to the photochemical behaviour of the film/developer/exposure combination.

Issues of gamma encoding actually make the digital situation quite complicated, too; it's not just a case of adding the numbers and dividing by two to get a 50% dissolve. Well, OK, a lot of modern software works in linearised 32-bit float internally which would actually make that the case, but even that's quite dependent on the input and output lookup tables involved, the behaviour of the camera, and so on.

If you're talking about a dissolve that has ever been through a photochemical process and then been scanned, it becomes even more involved. The effect is most visible on (poor) transfers made from prints, but sometimes it's very obvious that things like titles and specular highlights appear first during fades up and disappear last during fades down, in comparison to the rest of the scene. In that situation the peak whites are outside the dynamic range of the video image; sometimes there's even optical flare around them. As the film element fades up and down, the brightness of the image area in question can decrease quite a bit (potentially reducing the flare) before it stops looking solid white.

In short: it's complicated. I sometimes find myself using curves or levels controls and additively superimposing the images in order to have more control, and create a more organic, photochemical-looking dissolve. It can be done.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...