useful design's picture

Hey @Bodysoulspirit thanks for the composition. A cool approach. I doubt I would have thought of this way of doing it, especially since I didn't know of the Reinterpret 3D Object node! Very clever. I was thinking i'd have to recombine lists in some way to make a suite of four sided objects or triangles. Smokris' Reinterpret 3D Object node seems not to generate any ugly artefacts at the joins which is very nice.. Why is it that the X values appear to be the randomly moving values not the Y values that are the ones moving on screen? Is there a 90º rotation going on somewhere I can't see? Thanks again.

useful design's picture
@useful design commented on @monkey0815's Feature Request, “Polygon node

Point colour lists, like KinemeGL Triangle family patch could be cool too. 20 points (X,Y) 20 colors (R,G,B). Gradient mesh using point list and colour list data mapping.

useful design's picture

Why I need it:

View at full screen to see the joins, and check out 2013, ouch!

useful design's picture

See this comment for visual examples of how Adobe Illustrator does it.

useful design's picture

Thanks for your support and comments, @Bodysoulspirit.

I guess even if we could have a shape image which outputs an image as B&W (or layer I forget the difference as it's been a while!), that could then be used as a mask on anything, an image, a colour, a gradient, a movie. But that does add at least one or two more nodes and potential complications.

You know I'd like to be able to hand the shape node two curves and two lines and just have it determine the enclosed shape defined by those limits (think linear programming in maths) and render it. Would save me having to do maths to work out four intersection points and cull the unnecessary point data from the four limits paths.

God I would love to wake up and be knowing GLSL, conceptually this is piss easy in terms of what GLSL is capable of — but I just don't know how. And isn't Vuo for people like me who rather not code too much anyhow? ;-)

If this was a paid project wouldn't hesitate to reward Vuo with a node commission, but alas just for my graphing work for an NGO atm. Might grow into something I suppose.

faulty_a's picture

Thank you again :) And yes, one very last question. I honestly can't find an answer in the documentation. How to adjust the output ratio in VUO such that I will have a patch 1280x720 in VDMX? So far I managed to create a patch for VDMX using a protocol mode but my window size is 1024x768. I have tried a "Resize Image" node but once I connect to the Image Generator node I have 1024x768 window again.

Bodysoulspirit's picture

For the "Requested Frame" going into Curve, Math (time ports) you can simply use the "Time" published port from the protocol mode !
I guess in protocol mode the "Fire at Display Refresh" would work too.

There are some specifications yes to protocol modes. For mouse interaction for example, see feature request and comments below at "Pointer Interaction" protocol. I guess it's the same for keyboard inputs.

I have no other specific example in mind, ask here if you need more help ;)

Bodysoulspirit's picture

Alastair (@usefuldesign) I agree !

  • Totally would love an easy way to create a shape on the fly using a point list too ! (personally usually plane 2D rather than 3D too).
  • Be able to choose fill color or image.
  • Be able to outline a 3D shape AND layer (For layers, maybe a node like "Add outline to layer".

On the Polygon node request you say you don't see your points mentioned in your merged Make a Polygon Mesh with Fill node request. I guess since Jaymie said they're merged, I may guess your points are merged too ?!

Specific sub-parts of the line options maybe will be treated separately as part of : Render smooth joints when using the "Make Line (Strip) Mesh" nodes ?

Using Parabox's Get Mesh Values node along with Smokris Reinterpret 3D object I tried to make a graph like thing. But it is pretty hard since you have to retrieve specific vertices, then merging lists again etc.

PS : remember if you want faster implementation, you can always ask the team for a price estimation for a request and hire them :

faulty_a's picture

Thank you for your quick reply, Bodysoulspirit. Yes, I was doing it with some of the patches. However, there are many examples where the nodes of the Render Scene to Window node ( from the Requested Frame output or Showed Window) are plugged back into the composition (often into Curve, Math, or Mouse interaction). Examples from the VUO example set: Move Spinning Sphere, AddNoiseToClay patches and so on. I can just assume that some compositions cannot be simply adapted and must be totally restructured keeping in mind the requirements of the protocol mode.

Bodysoulspirit's picture

Hi @faultya and welcome ;)

Of course you can render layers and scenes to images for protocol modes.
Simply replace the "Render Layers/Scene to Window" with "Render Layers/Scene to IMAGE" nodes.

For example, a simple composition in default non protocol mode would be turned into protocol mode like this (joining them as a .zip too to download).



Vuo is more than nodes and cables, it's a community! Feel free to browse or add your voice.

Browse Discussions

Start a Discussion

Browse Feature Requests

Suggest a Feature

Browse the Composition Gallery

Share a Composition

How can I get notifications?

Learn more about the community

Learn more about Vuo

Vuo Announcements

Sign up for the Vuo announcements mailing list to get news and join the community. We post about once per month.