what the hell is compositing

hi guys,its been two years I’m working with blender and since that time i have seen lots of tutorials about compositing and lots other about modeling,camera tracking,particles… and all of those arrive at the same point and is that my nightmare “COMPOSITING”. yes nightmare,actually i hate that because i don’t understand it i don’t really comprehend compositing however i know much about that and i did some jobs doing compositing.for example when i started working with blender i made this ball which is awful.



and now i have done this one recently,the same but completely different.


I see some blender experts doing very very complicated compositing with nodes i just follow them connecting some ropes to some materials very fast very quick the don’t say why we should connect these two together they just connect and connect and connect.and now i’m tired of telling my self “o yeh i can do compositing very easily”:yes:,but the fact of the matter is that i know nothing about that:spin: actually i derive everything like a parrot:(.someone helps me with that please,I’d be thankful.

You have to give people more than “teach me how to composite” because that’s not going to happen. Too wide topic. It’s possible to give suggestions and tell how things work based on what you’re trying to do but you have to try first.

What is compositing? From Wikipedia:

Compositing is the combining of visual elements from separate sources into single images, often to create the illusion that all those elements are parts of the same scene.

That’s what it is. But because rendering creates a combination of everything in the scene and we often want to make changes to just some aspects of it in post-processing, the first step is to prepare for the changes we’re going to make after rendering. Separating and breaking things into components gives easy access to data which can be then used to make changes and compile (composite) final image back together without rendering again.

This post has a picture of basic separation of scene elements. The reason to do that is to have easy access to different parts of the image and make changes without rendering again. Want to change the sky? Put another image of a sky in the background. How about having less saturated red on the jacket? That’s easy because the character is separated and changing saturation on the red channel doesn’t affect the whole image.

Of course compositor can do much more but that is the usual thing to do when just making changes to your scene after rendering.

So maybe do that. Separate background from the foreground (cube, Suzanne, doesn’t matter) and then change the background in some way in the compositor and put out the final image. If you don’t follow tutorials and get stuck, that’s one way of knowing what spesific question to ask here.

I found this introduction tutorial wich explain very well what compositor is for and what compositing is:

Compositing is when you save your organic material from the food you don’t eat and put it in a pile in the yard. Put grass clippings on top of it and you can use it as fertilizer in the garden.:wink:

i can’t understand !! you blender expert are making me more and more confused :mad:

No! That’s composting! :rolleyes:

@rasolrpd sent

http://digitalcomposting.com/

Compositing is a way to combine data from different sources. These might be different render layers, different passes, or even a render and a photograph. For instance, when a cg shot is combined with camera footage in a visual effects shot, that is done by compositing.

People also use the compositor to tweak their renders. If you render out an image with lots of passes, you can individually change these passes, for instance by color correcting them in some way and then combine them. This allows you to change the look of a render without actually having to rerender it!

With the right passes you can even re-light a scene (to some degree).

In a nutshell, “compositing” breaks down the process of creating an image so that it is “a process.” Instead of mashing the Big Red Render Button and waiting umpteen hours to see that (“if”) the computer has magically and all-at-once produced just what you were hoping for, you break-down the process into smaller steps that do not have to be entirely repeated from scratch when (not “if”) something needs adjustment.

For instance: you have a shot of two kids walking down the street with their dog. Say the camera doesn’t move. Instead of rendering all of those frames, and all features of each frame, all-at-once such that if anything is wrong you’d have to start over, you instead break down the shot into small pieces. For instance, the background only needs to be rendered once. The three models can be rendered separately. The light that falls upon them – the shadows that they cast – can be dealt with separately. Also, many of the things that you might like to do to “sweeten” the shot can be expressed in two-dimensional terms, possibly with reference to Z-Depth information, and those things if done that way are considerably faster and cheaper.

So, that’s what you seek to do. And “all those nodes and pipes” are simply a way of describing a “data flow pipeline” of sorts by which the various pieces of data will finally be brought together, and tweaked, to produce the final shot.

Ansel Adams, a very famous photographer, once said that “an image is captured in the camera, but it is made in the darkroom.” The same reasoning applies here, too.