SEMs Alway Have Horrible Offsets?

Sorry, there seem to be details missing in your question. 

A SEM is a virtual object under which you can nest other objects in the A3d hierarchy. These nested objects will then move relative to the position you give the SEM.

The distance the other objects have to the SEM when nesting will become an offset. That means the position values the child objects show in the inspector are the distance of the child to the parent. This offset stays constant, for it to change you would have to modify it in the inspector.

I’m talking about using it as a focus palette. Not a stage object like a curtain. I expect no offset, or at least the ability to remove the offset, when using SEM as a FP, a supported feature (Track to SEM). If I set the FP to track to the SEM, it should track to the SEM, not a random point in space. Relative changes work, but the point was for the light to track the SEM, not to experience the change of the SEM. I’ve tried pointing it at the SEM manually and then recording FP, but it jumps to a random point in space no matter what I do. And it keeps messing with the fixture’s other 3 xyz-focus parameters, which make no sense because that’s not a real world parameter. I just want it to point at the SEM when it is in the SEM fp!

When you look at the content of the focus palette that has the Track SEM property, does it look like this?

The three white 0's are the offset. This can be used e.g. when you want the light to point at the actor's head, but the SEM is located at his feet.

If you already have the three 0's, please send the showfile along with instruction to reproduce to eos(dot)moderator(at)etcconnect(dot)com.

I just deleted last reply, I monkeyed around with it some more and found the screen you had. Took me 10 minutes to decide if your screenshot was blue or black, then I compared it to my own screenshot and figured out it was definitely blue. Yes, it looks exactly like that.

OK, what I've discovered is that if I place the SEM at location (0, 0, 0) and THEN do the focus palette stuff, then it works just fine. But if the SEM starts at anywhere but (0, 0, 0), then the offset from zero at that point in time is implicitly baked into the focus palette for all eternity. Because the operation is purely relative. 

That's a bug in my opinion. If it's not a bug in ETC's opinion, perhaps document the decision? (Note: Such phantom offsets are common bugs.)

The expected behavior is that the light points at the SEM. The expected behavior is not that it simply inherits the SEM's change. Inheriting change is fine for the set piece stuff, but is antithetical to the whole concept of focus palette. You don't set a light into a SEM focus palette because you want it to point at not the SEM.

I just deleted last reply, I monkeyed around with it some more and found the screen you had. Took me 10 minutes to decide if your screenshot was blue or black, then I compared it to my own screenshot and figured out it was definitely blue. Yes, it looks exactly like that.

OK, what I've discovered is that if I place the SEM at location (0, 0, 0) and THEN do the focus palette stuff, then it works just fine. But if the SEM starts at anywhere but (0, 0, 0), then the offset from zero at that point in time is implicitly baked into the focus palette for all eternity. Because the operation is purely relative. 

That's a bug in my opinion. If it's not a bug in ETC's opinion, perhaps document the decision? (Note: Such phantom offsets are common bugs.)

The expected behavior is that the light points at the SEM. The expected behavior is not that it simply inherits the SEM's change. Inheriting change is fine for the set piece stuff, but is antithetical to the whole concept of focus palette. You don't set a light into a SEM focus palette because you want it to point at not the SEM.