NLA

Support Basics & Interface
1

Does anyone have a good begginer NLA tutorial?

2

I don’t, but malefico (;)) has some:

http://www.argoslabs.com/~malefico/tutor/nla-en.html

http://download.blender.org/documentation/html/x6045.html

http://www.argoslabs.com/~malefico/tutor/walking.html

Good luck…

3

The way that I have learned the most about it, so far, is by futzing with it … a lot.

In spite of the peculiar name, the essential idea behind NLA is that it enables you to describe complicated motions of a figure in terms of a blended sequence of actions. You place these actions together on a timeline. Let me try to give a specific example.

Scenario:
A water-pump is churning away merrily in the background as a man walks into the frame, pauses in front of a switch, reaches up and turns the switch off. The pump glides to a stop as the man exits. The overhead ceiling fan does not.

We have several actions here, taking place over time:

  • The spin action, for the ceiling fan, executes continuously. It’s associated with the fan and “isn’t an ‘NLA action strip.’” (That might not make much sense now, but for completeness, a hierarchy of sorts does exist in the action system. Nevermind…) - The pump action, for the water-pump, is the first of many ‘action strips’ you’re working with. It’s already running when the sequence begins, and it “blends out” at the time the switch is thrown so that the machine returns to its resting pose. - The worker has many actions, like Walk, RaiseArm, StandStill, and PullSwitch. The Walk action blends-out as he comes to a stop in front of the machine, and blends-in again as he starts moving after throwing it. The character can be performing several actions at the same time in a smoothly-blended sequence. - Even the switch has TurnOff.

Each action consists of seperately-scripted movements for the various parts of the machines or armatures of the figure. These movements, defined in the Action Editor, are in turn made up of IPOs computed automatically by Blender as you establish “poses” for the actions.

So this gives us a three-tiered system for defining object-movements. The NLA is on the top level, describing movement in terms of actions. Actions are on the middle tier, describing the details of each action in terms of bone or object movements. IPOs are on the bottom tier as the actual graphs describing the change of variables (such as position) over time.

We humans routinely think in high-level terms, such as the Scenario given above. But for the computer, “it’s all ‘1’ and ‘0’ and nothing more.” So we have a system here that allows us to build from the bottom up, so that we can describe from the top down.