![]() Break down the problem into pseudo-code first, and gradually boil it down to snippets of code you can try to accomplish or research. Myself I wouldn't dare call myself even an intermediate, but I've gotten by over the years with some Java in after effects or python in Maya, often to automate labourous manual tasks, the result of this was that vex was very familiar the first time I looked into it.Īnd as someone else pointed out, I think the best way is to figure out a task you need vex for, rather than "trying to learn vex" - for the sake of knowing it. Some have touched upon this but just wanna underline that if you haven't done any coding before (which I assume) then you're not just learning vex, you're learning the basics of programming. ![]() It's important to keep things modular and simple to give you power, flexibility and abstraction, which are the things Houdini excels in. With both of these, be sure to press the little button with the sliders on the right side of the wrangle code box, that will create spare parameters automatically based on your code. You could rotate it, transate it, or even apply a noise displacement on it, and the wrangle, in a single line, will blend it with your original geo using a simple 0-1 attribute. The brilliant part about this way is that you could do lots of things to your second input before it is piped in, as long as the topology remains the same as the first input. = Which is more of a general purpose masked blendshape between the first two inputs of a wrangle, of which you'd supply the original geo in the first input, and your peaked geo in the second.+= which will push the points out more or less based on a multiplier attribute which could be supplied earlier from an Attribute Noise SOP, a Distance to Surface SOP, or even a different wrangle that calculated the attribute in a custom fashion (it's good to not do everything in one wrangle unless it's necessary, so that you can easily swap things out, disable nodes, or generally just not have too many crossed wires).There are two ways to approach doing this in VEX: Peak offsets a point's position along its normal by a distance, but there's no way to give it a map of how much to apply that peak. One example of VEX I like doing for example, is a weighted Peak SOP clone. Sometimes, it's just about doing one specific thing in VEX before getting it back out into the normal node network so that you can play around more and leverage the powerful tools and nodes that you're already used to working with. Sometimes one or a few lines in vex will replace a complicated network of nodes because it can be precise and do exactly what you need, but that doesn't mean it's a great solution for everything. ![]() Knowing how to use VEX is more about knowing what problems it solves well versus when it's better to just use nodes. If you haven't already, one of the links from the sidebar has a great VEX resource, free:
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |