Possible to rotate new Relay-Node?

unhandled

(Lesan) #1

Hey Guys,

I just wanted to ask if someone knows a way to rotate the new relay-node. I really like this new feature, but I think it would be way more effective to tidy up the wires and control the flow, if one could rotate the nodes. Kind of like cable-management in the real world.

Especially if I want to connect a node with one that is on the left side of it, the wires quickly get chaotic, the new relay-node seems to almost be able to solve that problem and make grasshopper even more user-friendly.

Thanks in advance and have a great day!


(David Rutten) #2

That’s always a problem, which is why I’d recommend against it. It’s a pretty useful visual clue that objects to the right of some other object are never part of the source data.

Imagine [appeared=1] how difficult [a sentence=2] it would be to read [2] if some of the [later using=3] words [1] early on but were [index=4] referenced [3] some [4]. The solution here is not to make a more readable reference system, but to stop putting stuff in the wrong order :slight_smile:


(Lesan) #3

I get it, but in some cases these things are unavoidable, for example if one uses hoopsnake or any other script that is working with feedback loops.

Often I’d just like to see where the wires are going without having to concentrate too much on which wires intersect each other or go through nodes, while not having to seperate the nodes too much, so can can see a good portion of whats going on in the script.

By the way: Back when the relay-nodes didn’t exist I used data-nodes for changing the tangencys of the wires and tidying up a bit, with those I often grafted or flattened lists, which I could imagine would give the relay-node a bit more versatility.


(David Rutten) #4

Not being a hoopsnake or anemone expert myself, do they require that wires go back in order to work?

It would require a rather significant redesign of the GH1 code-base to allow for different wire-shapes, so at the earliest this will be GH2, but I’m still far from sold on this idea. It sounds like this would be a fix to a problem that shouldn’t exist in the first place. Perhaps you’re right that in the specific case of a cyclical network something like rotating relays make a lot of sense, my thoughts haven’t settled on that topic yet.


(Michael Pryor) #5

I think Anemone addresses Grasshopper’s left/right format well (Also it’s easier to understand if you imagine them as curly brackets in a c# script {loop code;} )


(Lesan) #6

Well, I’m not a hoopsnake expert as well, I just used it a couple of times, but the logic of a feedback-loop requires the outputs to go into the inputs, it’s a loop in the end, thus having a snake eating itself as a logo.

As I’m thinking more about the idea of rotating nodes, I see that it’s an option that can very well make a script less readable if used incorrectly. But at the same time, to some people it could benefit their script, I think it depends on the expertise of the user and understanding grasshoppers “syntax” whether it makes sense to use rotated nodes or not. If you guys someday (hopefully :smiley: ) will decide to do something like rotateable nodes, I’d probably add a checkbox to the grasshopper-options to activate the function just to add a bit of resistance to using them when starting out with grasshopper.

But I had no idea, it required to dig that deep into the basecode, oh well…


(Michael Pryor) #7

And that setup also makes it really easy to connect / understand nested loops as well. Plus it’s multi threaded (maybe hoopsnake is also, dunno it so well)


(Lesan) #8

seems like I’ll try out anemone then… that’s way easier on the eyes than hoopsnake. Thanks for showing these examples!


(Michael Pryor) #9

Ah and last you can run different loop actions in parallel by just adding more inputs / outputs.