Grasshopper

algorithmic modeling for Rhino

Grasshopper vs 3ds Max Design - Slate Material Editor

I'm currently teaching myself 3ds Max Design 2013 by going through the online tutorials and I was struck by how similar the Slate Material Editor is to the Rhino Grasshopper graphical interface.

 

The tutorials can be found here: http://docs.autodesk.com/MAXDES/15/ENU/3ds-Max-Design-Tutorials/ind...

 

The Slate Material Editor has some functionality that I wish Grasshopper had, namely the ability to drag a component onto the canvas and then drag a wire from the component to access a shortcut menu to other components. I tried doing this in my GH (build 0.8.0066) and it didn't work that way.

 

I suggest that this shortcut method would be a welcome improvement to the GH interface. What do you think? Thanks.

Views: 4031

Replies to This Discussion

I'm disappointed that this post has been viewed 84 times and no one has any opinion about it.

You don’t need to have 3ds Max Design installed to see what I’m talking about. The tutorials hyperlinked above walk you through it pretty well. I really think Grasshopper could benefit by adding similar shortcut menu functionality to its components.

Hopefully David Rutten has taken note of this discussion and will consider it for future builds of Grasshopper. Thanks.

Hi Mark,

I saw the post and went to the link you provided but couldn't find an obvious example of what you were talking about. Can you link me to a tutorial vid or screenshot of the feature you're looking for?

--

David Rutten

david@mcneel.com

Poprad, Slovakia

Sounds interesting.

How does the shortcut menu appear, how do you initiate it?

What is available on the short cut menu?

Can you give an example work flow of how you would want a feature like this in Grasshopper in terms of Grasshopper Components and actions?

I think the main issue with what you propose is the potentially large number of possible components that you would want to connect to.  Already there are many options in the max slate editor, in Grasshopper it could be many other components, and it might be overwhelming.

One possible solution would be to use the markov widget's intelligence of what YOU usually connect to.  As you use GH more and more, the system would learn which successful connections were made with the result of the component...does that make sense?

Here's two screen captures from the tutorials that illustrate what I'm talking about.

Screen Capture 1 shows the shortcut menu functionality

Screen Capture 2 shows an insert of a component over an existing wire functionality

Both of these features could really streamline the GH workflow by limiting the amount of times you have to search for a component in the existing interface and automating the connection of components.

 

Attachments:

As for me option n1 seems to be counter-intuitive. I love gh cause its really simple to learn user interface workflow, which is simple as.... notepad ?

Screenshot 2 provides something that was discussed here several times in different context...for 3ds max its possible to do it, cause there are mostly singular inputs & outputs. How would you see connecting wires when you have i.e. 2 inputs of integer type ? Which one to choose ?

I can certainly see how this sort of feature would improve workflow (which is why I wrote the Markov widget). There are however problems that I do not know how to solve (yet). Here are a few of them:

  1. When inserting a component into a wire, which inputs/outputs ought to be hooked up?Sometimes it's clear but at other times there is no obvious solution. Should I just guess and perform a potentially expensive solve iteration? Should I not hook up anything if it's ambiguous? Neither seems like a good approach.
  2. On big definitions wires could be quite dense or intersecting. How does one select a wire in a jumble?
  3. This empty wire popup menu obviously cannot contain all available components. Grasshopper ships with ~1000 standard components and there might be hundreds more if you've installed 3rd party plugins. Should I use Markov chain logic to populate the menu? Should I always display a set of standard 'most likely to be useful' components? Should I allow users to create their own list?

I'm not going to do anything about this soon, 0.9 is now being ultra-tested and I expect there will be a few months of serious bug fixing ahead once it hits the shelves. Also, further development of the Remote Control Panel is priority #1 after 0.9.

--

David Rutten

david@mcneel.com

Poprad, Slovakia

Hi Mateusz,

I understand your concerns. These shortcut methods aren’t intended to replace the current interface but rather to augment it by providing alternative ways to insert components into the canvas.

I think the shortcut menus could be kept simple by having them be context sensitive to the specific component you drag a wire from. The insertion of components over existing wires could be limited to the wires it will work on. For example, if there are too many inputs for the insertion to work, the insert icon won’t show up in that instance.

I also think Luis’ suggestion about the markov widget has interesting possibilities. Thanks.

With this kind of feature it should work for all components but sometimes even context sensitive is not a limiting factor.

Take The Preview Geometry Component with only two inputs Geometry and Shader (Colour/Material)

Here is the context sensitive options for Input S

However even when I limit it to context sensitive items the Geometry input has many more options all valid in one way or another. I stopped after the Surface Tab but there are plenty more valid choices of Native components and many more of plug-ins

Like Luis said maybe the markov widget could be employeed here to limit to user context items

Another option is to only include components which are actually visible on the canvas. It's a poor-man's filter, but may well work for 99% of cases.

--

David Rutten

david@mcneel.com

Poprad, Slovakia

Hi Danny,

Excellent point and example!

Maybe the shortcut menu option is limited to the simpler components that don't have so many options. For example, if you drag a wire from a component and a shortcut menu doesn't appear, you know that this component is too complex for that insertion method. You would then build the component connections in the traditional manner.

Over time you'd get a feel for which components you could use the shortcut methods with. Thanks.

For example, if you drag a wire from a component and a shortcut menu doesn't appear, you know that this component is too complex for that insertion method

Personally I think it should work for all or none.

 

RSS

About

Translate

Search

Photos

  • Add Photos
  • View All

Videos

  • Add Videos
  • View All

© 2024   Created by Scott Davidson.   Powered by

Badges  |  Report an Issue  |  Terms of Service