Framing geometry

[rurik]

Hello

I just created a network that has a geo node, a cam, a light and a render TOP.
I dive into the Geo node and bring in an FBX 6.1 file, comes in very nicely, properly orientated etc...
I'm trying to pan my camera so I can frame the object, which is a 3d building.

If I try to change the coordinates of the camera, I can see this happening clearly in the Cam node, but the light, the geo, and the render don't seem to be affected at all. Or it is the other way around, the cam node does not seem to be affected but the geo node does...

What is the recommended way of moving cameras and framing an object in Touch?

Thanks
Alex

[rurik]

Here is an example of what I'm trying to describe. As you can see I have a CAM that is not showing me anything. I then have a geo node that shows a certain angle on the Geo, and then the Light that is showing a different angle (I assume the light thumbnail is just showing what the light is seeing). However my render TOP is showing something different all together.

I just want to use my camera to frame the geometry and have the render TOP reflect that...

Thanks
Alex

[selina]

from the look of your network, the reason that the light & geo does not change when you change the transform of the camera is because the camera is the parent to the geo. disconnect the wire between them and the geo will no longer transform with the camera.

Selina

[rurik]

Ah, that brings up a great point...I've seen in tutorials the flow of holding down shift or control when adding operators...What is the difference between the two? And does the order of clicks mean the first clicked will be parent to the second clicked?

Thanks
Alex

[malcolm]

When holding shift you can add multiple the nodes that will be wired together, with the first node being at the left (or top of it's a COMP). Holding ctrl will also add multiple nodes in one go, but they will not be wired.
When not holding these you will just add one node and the window will disappear.

[rurik]

I see, thanks Malcom
So what is a visual way to indicate parenting in this case? How does one know what is parented to what?

Thanks
Alex

[rodberry]

wiring the geo to the camera seems like a logical thing to do but parenting works differently to the procedural flow in the actual operators in Touch.

the flow betwenn OPs is left to right and each processes whatever comes in the left side qnd feeds the result out the right side.

components however (although we can add procedural in and out connections to them) have connectors on the top and bottom signifying their place in a hierarchy of either geometry or location in a control panel.

so you are not feeding a signal from an object to a camera to make it render. If a geometry component is the child of a camera, it just means that whenever the camera moves, the geometry will move with it - in other words, the child inherits all the transforms of the parent. Usually , you might use a null to parent several geo components so you could move them around as a group. or, if you are making a control panel, you might use parenting to put panel components 'inside' parent containers to organise your interface better.

rod.