I am very new to Blender and find that one of the main issues people have is render time.
I have only buildt about 10 models and am now working with animation.
Question is, would it be faster, or is it posible to render layers using green screen techniques and then edit these layers using film editing tools? :spin:
I don’t think it’ll make a difference
check out this article
http://www.blenderguru.com/13-ways-to-reduce-render-times
I was looking at that site earlier tonight. Good tips.
The biggest eye-opener :eek: lesson I ever learned with Blender was … compositing. Really getting to know and to (try to) understand just what “nodes” can do.
That kind-of naturally leads you to a workflow in which you take a scene and break it down into little pieces that you know that you can deal with separately. You start maybe by working-up a rough version of the scene as a “preview render,” then you take the whole thing apart and start to build it up. You amass many directories of MultiLayer format files.
For example, in an animation where the camera isn’t moving, you need one frame for the entire background. When the character’s moving, you might treat the diffuse light, the specular light, and the shadows separately. Just like in a recording studio you capture the individual parts and then “mix them down” to make the final track. You therefore want each component to be isolated and pristine.“The truth, the whole truth, and nothing but the truth.”
You get somewhere, then you toss in other stuff just to see if it improves the stew… it’s a process of successive refinement. Sneaking-up on the goal until you get there. And, trying to be sure that everything that you commit compute-resources to, is something that you can actually keep and not re-do (even if you … oops … don’t actually wind up using it).“A lit-tle bit of this … a lit-tile bit of that … a pot … a pan … a broom … a hat…”
“Dou-ble, dou-ble toil and trouble, fire burn and caul-dron bubble …”
Another idea is: “forget reality.” Focus instead on what “reality” looks like, and maybe what the viewer will and won’t actually pay close attention to. If there’s a cheaper way to do it that’s almost good enough, that’s almost always good enough.
“Gee, I should use vector-blur or motion-blur here, but what if I just drop a blur in here? Do you think anyone would actually notice? Gee… guess not. Cool! Good enough! Next!!”
I think you missed the whole point.
Say you are creating a fishing scene at the lake. Rather than creating the entire scene, then having to render the entire scene. Would it not be easier to creat the animation using a green screen background, then using video editing tools such as Windows Movie Maker and simply add that to a pre-recorded Scene?
Actualy, would you not be capable of adding this animation to any pre-recorded scene, if it did have a green screen background?
You don’t have to use green-screen in this case, if you want to add a CG animation. The so-called “alpha channel” will conveniently do that for you.
Simply produce the animation without any background, without any “sky color” or anything like that, and be sure that you’re recording the output to a file format (ideally, MultiLayer…) that captures the Alpha channel. (For example: “.JPG” does not have alpha; “.PNG” does.)
Alpha, of course, represents transparency. It’ll be “1.0” where the image is solid; “0.0” where the image is not-present; somewhere in-between where it’s translucent. The alpha will be “anti-aliased” to provide a soft, non-“jaggy” edge. A perfect mask, automatically provided by the system with no need to chroma-key.
If you wanted to go the other way – to incorporate live footage – then that footage would need to have some kind of chroma-key information to allow you to produce an equivalent mask for it.
Blender’s got all the goodies you need to put it all together: nodes to pull the chroma-key masks; nodes to do alpha-over combining; and lots of stuff also in its Video Sequence Editor. You can certainly use Windows Movie-Maker as well. You’ve got choices. Good ones…
When designing the animation, you’ll need to devise a 3D space for the action to take place in, rigging that space carefully to coincide with the lake footage in a plausible way, suitably consistent with how you “know” a real, therefore three-dimensional real-world lake to be. If you want to get very fancy-pants about it 
you might even toss in some Z-depth driven defocusing and/or blurring nodes so that the animated characters seem to be more “distant” on a “foggy” lake when they’re farther away from the virtual camera. The sky’s the limit. (Or should I say, “the lake?”)
I think we’re on the same page. I am new to all this 3d modeling. The only experiance I have had, is from my nephew. He owns a photography studio and creates a lot of this stuff. But then, he has the equipment for it too. lol
I understand a bit about alpha channels through the work I have done in Gimp and Photoshop.
I am going to look for some tut’s on all you have mentioned.
Sure would be nice not to have to creat a green screen effect for all that I am doing.
I am working on a dragon now that is part of a scene I created years ago and would like to bring him to life.
I have pics posted in the thread “38 Year Old Dragon” Needless to say, I created him 38 years ago. lol
Just with all the detail in him, not to mention the rest of the scene that includes a castle and him fighting with Merlin, It would take a month to render out a 5 minute video. The dragon is about half done and using 1 subserf, the poly count is already 46,000.
Thanks for the help. 