If layer can be animated, why not parent?

I’m working on an architectural walk-around. This is different than a walk-through because I’m evaluating how a building looks in its environment (not just the environment of the building), and in particular, how it may look from other buildings in the environment. I have been using the feature of putting complex geometries and lighting grids into different layers and then using the layer parameter to make them part of the render or not–very helpful for optimizing for speed. I also note that a layer is not a parameter that changes continuously–it’s a perfectly discrete parameter that cannot be interpolated.

I have now found a case where I wish I could change the parent relationship during the animation. I have an environmental context that contains a ground plane (actually, a topography), trees, existing structures, etc. The ground plane is the parent for all these existing structures. I also have a subject building that can be placed within this environment. Since I built the building first, everything is based around an origin of 0,0,0. For the walk-through inside the building, the target and the camera are parented to objects that allow for panning views and cycles around a point, and that’s all well and good. But when I walk around outside, I want the camera’s reference to be based on the existing buildings and the target to be based either on the subject building reference points or the existing building reference points. These reference points change in relation to one another if I move the ground plane with reference to the subject building.

If I could key the parent relationship, I could easily move my ground plane (the effect being to move the subject building relative to the environment) and have all my key frames still keyed to the appropriate points. As it is, I have to pick a given frame of reference, and when that frame of reference moves, I have to manually move those keyframes.

What do folks think about making the parent relationship animatable, like the layer parameter?

Aligorith coded an offset modifier which is essentially a “parent” modifier. Hopefully it will be in the next release. A patch was available but you may need to search Aligorith’s posts to find more info.

I’m not sure why you need the groundplane to be the Parent of what’s displayed on it. The Groundplane, buildings, trees etc could each have their own Origins and be keyframed to and from visible layers. You could also have duplicates that may or may not be translated relative to the original that you could then keyframe to and from layers. (In F7, Draw tab, Layers, you also have the ability to display the same object on multiple layers). The only possible problem I foresee is when trees or buildings need to change their elevations, but you could overcome that with duplicates.

%<

I think you misunderstand my intention. There are about 50 environmental objects that are fixed in relation to the ground. There are over 100 objects that are fixed in relation to the center of the building. I want to move the building relative to the ground. Everything in the building has to stay fixed relative to the building. Everything on the ground has to stay fixed relative to the ground. When I move the building relative to the ground (or vice-versa), everything within one of those two reference frames must stay positioned within that reference frame.

The fact that I can bounce objects from layer to layer is cool. But what I need to do is to make the camera and target object change their reference points, hence the need for a new animation feature.

I think you misunderstand my intention.

He he, highly likely. You can use more than one Camera with dedicated Target(s) though. You could also use Library Files; like use your existing file as the Library and Append or Link objects to seperate (other) file(s) for the final setup. Each of those files could have their own Camera (and maybe Lamps) setup.

After rereading I understand your need for a new feature but just trying to suggest a workaround for the time being.

%<

Hi ! As my english is far from perfect, I’m not sure that I have understood everything well, but I think that you could solve your problem using Empties and constraints for your cameras.
You can use as many empties as needed for your reference positions (targets), and Use Copy location constraints for your camera.
Applying Ipos to the influence factor of each target, you can displace your cameras from a reference point to an other along the animation, without using layers.
You can add a Track to constraints to keep the target in the camera field.
Philippe.

yes, track your camera to an empty. Animate your empty through time, flying it through the building, and then outside to rest on the other buildings or features, such as a fountain. At the same time, fly your camera behind the empty for the first part (inside the building), and then fly it outside to rest just on top of or in front of the other buildings; your new building then can be off center or wherever in your frame of reference. Your render will be as if from someone standing on the 5th floor of a building, looking out across the park at a a fountain, with the new building off to the left.

Think of the empty as a carrot on a stick (a very long stick sometimes), and your camera as the horse that will always look at it.

As far a physically moving the building, like having it fall from the sky, or adjusting its position on the lot, make sure that everything inside the building is the child of the building, so they move with it. Then when you animate the building moving from one place to another and the lot rearranging itself, nothing gets left behind. You will want landscape environmental objects to be children of either the landscaping that goes with the building, if you want to rearrange them as well, such as sidewalks, parking decks, fountains, so thatyou can animate going from Plan A to Plan B (building rotates to face East, fountain swings around to stay in front of building).