This new tool you propose has been requested before. The ability to edit more than one layer at a time to avoid constantly selecting other layers in a very complex file. My solution to get around this? Bones baby bones. Edit all the layers with bones. Right now that is the "easiest" way to "tweak" all the layers.
Synthsin is working on that layer select tool which is wicked cool and probably as close as you will get to this feature in the near future. I think this concept is so expansive that it will require a major "rewrite" of the application to achieve.
I'm not a programmer. I have some skills with Javascript and Actionscript for the web and Lua specific to ASP. I can still see how massive this undertaking would be. It would require a ton of new features not just a single new tool. This one "tool" would change how the whole application works.
In your wink presentation you indicate how the splines become "visible" for editing when the parent layer is selected. This alone is a major advancement. If just that were possible excluding "spearing" through sub group layers like switches and other bones or groups, that would be over the top freaking amazing.
Let's assume for the time being this tool would need to be created via the current lua scripting feature set. There are a ton of issues to deal with that I covered previously, like layer rotations etc. That is a huge obstacle from my point of view. Some people like to have hands on a separate bone layer that is bound to a hand bone in the skeleton bone layer. This hand bone layer "rotates" with the hand bone. To transfer that vector motion of the hand layers to the "tweak" tool and include the layer transformations is not an easy feat.
Another issue is how to "identify" layers with the lua script interface. Layers can only be identified accurately and specifically with the name of the layer. Layers can change order so the ID is useless. A layer's ID is base on its order in the layer palette.
I have dealt with this before and the main problem I could never get around was that the program allows for layers with the exact same name. Unless you make that a requirement for the user to follow on their own any script that requires targeting of the layers is very difficult. I tried to code in unique layer naming in a script and never quite pulled it off. This would need to be fixed in the application. At a minimum unique layer names on the same level or in the same group.
Synth came up with the cool idea of reading the actual file format. I'm using that for some scripts as well. It seems to work in some situations. Reading the file format is as close as you can get to having access to every single aspect of the whole document but it is extremely complex to do. I was hoping to come up with some reusable lua code that could be used to parse all the elements of the document for easy use. It's proving harder than I thought. I can't even get the key frame interpolations correctly.
That's another thing... key frame interpolations (smooth, linear, cycle, ease, etc). those values are not available for reading in the script interface. You can create a key frame with an interpolation value but you can't READ an existing keyframe and determine what interpolation is used. So transferring the motion to another global layer for editing is going to require reading the file format. But if you are CHANGING the file as you read it, changes are not available until the file is saved and read back in by the script.
Another issue, what about the sub layers point motion? Does the tweak tool only edit its "own" points? If you selected a sub vector layer would you not see the tweak layers point motion? Would you now have two sets of key frames? Point motion on the sub layer and added point motion on top in the "Tweak" layer? Yikes... my head starts to spin.
Another issue... yes another one... other tools. The display of this "tweak" tool layer. How would it work with other tools? You can't display splines on a bone or group layer. You can see the points on a sub layer only if "show construction curves" is set for that layer.
So... this tweak tool will require interface "drawing" of shapes. Another major crazy undertaking. Yikes. This would be similar to Fazek's Bezier curve tool experiment. It uses the script interface drawing features to "fake" a bezier curve in the drawing space. It isn't "real" it's only for display. So you would need to read in all the vectors of a sub layer and "redraw" those shapes on the group layer using the draw features of the script interface. Then... where does that vector layer "live"? How are those point movements moved around and stored?
I know programmers who do this stuff. I know Martin Hash and have talked to programmers who work on Animation Master. I know from what they tell me how freaking nightmarish it is to pull off any feature. This is a huge one. Any change like this could break a dozen other things in the application. It's huge.
I would suggest you work up a more detailed use case. Explaining and covering what actually happens to each layer with the tweak tool. Document it with screen grabs and how the interface looks and changes when selecting layers. Submit this in the feature request area.
------------
When I requested features for Animation Master Martin told me to create a "use case". A tutorial with screen grabs and steps for the feature as if it already existed. Imagine a help file or tutorial from a manual explaining how your tweak tool works as if it were a real thing. Just showing the splines visible on the group layer doesn't answer enough questions for the programmer/s. The programmer needs to know what is suppose to happen when this is clicked or that is clicked. All those issues regarding layer rotations or offset need to be specified.
Some of the above is actually part of a use case for your feature request. All those issues have to be dealt with by someone. If you, the requester of the feature doesn't answer those questions then the programmer has to figure them out. If it is complicated and hard to understand what is wanted then the programmer has do more work.
In anther topic Rhoel and I are discussing how to improve the stereo render feature so AS can be ready for the future of film making which is becoming more popular... 3D stereo. Lots of big budget new films are coming out in "3D". New 3D theaters are planned to get people out to the movies again. Mike Clifton pops in and asks "What format? Single images? Side by side? How do you want it to work?".
He can't read minds. He doesn't no how it should work. We must tell the programmers as much as we can EXACTLY how something will work.
------
Maybe some much smaller scaled back version of this might be done. It would have a lot of limitations, no sub group layers, no layer rotations in the sub vector layers. For instance this simplified version might be useful to AS users who do only point motion on multiple layers. They could move points from the top group but not any vectors inside switches or other bone layers, and also no layer rotations, so separate hand layers would be out.
I don't do a lot of point motion... very little in fact. A few tweaks here and there. I am a total bone freak
. If I never touch a point during animation I'm a happy camper. The current vector editing set up is good for that but it could use improvements.
This Tweak feature or tool is so massive to accomplish that the work to create it has to be compared to the benefits of the tool itself. If the benefit doesn't outweigh the effort I can't see myself getting excited about it.
For instance I use a custom tool set to group and colorize bones. It's a freaking nightmare to install but I love it. I can select groups of bones, turn them on and off in the interface. I use it all the time but it's so difficult to install I haven't made it available to other users. It requires editing every single tool and adding code to any new tools. It also creates a text file that has to go along with the AS file.
I LOVE discussions like this. I love thinking about how to create a new feature or some scripting experiment. Your idea GCharb is really out there. It's a HUGE one. Doesn't mean it's not a good idea. Just means it's going to be chore to pull off.
Keep thinking about it. As you become more experienced with AS you may start to see how this would work differently.
-vern
