# GeometryNodes: how to pass attribute between domains?

I want to place packages at a specified x y grid position on a pcb board. For that I use 2 chained GeometryNodes Modifiers.

• The first Modifier (Board) creates a cirquid board with a grid of pads and
• the second Modifier (Placement) is responsible to instance and place the packages at the specified grid position

To describe the grid I pass attributes (nx, ny) down to the Placement modifier. Now I want to use ny to extract the corresponding pad index depending on x and y. That works well if I use an constant value input for ny but if I connect the incoming nx attributes seems always to be evolved to 0.

How can I use ny in the computations in the pad points context in the example?

I know, I could precompute x and y and append them to each of the instances in the first GeoNodes Modifier. But I'd like to solve the problem in the way described above to reduce redundancy and to learn more about Geometry Nodes.

Is it possible to pass a single value or do I always need a dummy Geometry for that (the mesh.vertex field below).

Many thanks for hints or an solution

although i read your description 3 times i am still not sure what you want. But i assume you want to use a value from the first GN modifier in the 2nd one. If that is right, you can use this in the 2nd GN:

Then in attribute you will have the value you want and can use that.

• Thanks,the value is already successfully passed to the second GN. As shown in the table in the screenshot above. My problem is, how to use this value in the instances domain, esp. as input for the multiplyAdd node. (see the red "replace by" arrow)
– leol
Feb 5, 2022 at 15:52
• i don't get it!? just plug from attribute output to multiply add input... Feb 5, 2022 at 15:56
• that does not work. (the red line with the "?"). It behaves as though it where 0.
– leol
Feb 5, 2022 at 15:59
• did you try to add the transfer attribute node? as i showed in my answer? Feb 5, 2022 at 16:00
• if this doesn't work...please provide blend file. Feb 5, 2022 at 16:00