CGWiki

Appearance

Getting Started ​

So you've decided to take a look at Houdini? Good for you! If you're still not sure and you're a competent Maya user, maybe skim the MayaToHoudini page, it might convince you.

This site assumes you can navigate around Houdini and know the basics. If you don't, don't worry, the core UI is very simple, and shouldn't take more than an hour to learn. If that's you, then you probably want to do these steps first:

  1. Download Houdini Apprentice. It's free, and gives you access to everything you need.
  2. Check out the Sidefx Learning Paths, clear quickstart guides to show you the basics.

If you like video tutorials, but want to play long, maybe watch me blather on in a webinar I recorded for Sidefx. There's a download link in the description to get the same file I'm demonstrating with, gives a bit of an insight to how Houdini is used (skip to about 10 mins in if you don't want to hear my life story):

https://www.youtube.com/watch?v=4d0q-iNXHIo

Once you can move the viewport and get into a SOP network, you can start working through this site, I suggest in this order:

  1. Houdini - lots of quick examples
  2. HoudiniTutorials - conversational, borderline rambling tutorials. Like a text based, hour long vimeo tutorial you'd play at 2x speed
  3. Houdini_Lighting_Shading - how to place lights, assign materials, create an image
  4. HoudiniVolumes
  5. JoyOfVex
  6. HoudiniDops
  7. The rest. 😃

After the first month ​

It comes up frequently on the discord that people get past the basics, then get a bit lost with what to do next. I tend to say the same things over and over, I've got tired of it and decided to summarize here. 😃

This really becomes a metaquestion for learning itself. Why learn Houdini? Why learn anything? If you say 'for a job', well, that can be a dangerous motivator. As Oliver on the discord loves to point out, there's easier jobs that earn more money, why would you choose Houdini? For me the answer is simple, but my answer may not be your answer. Find your answer.

Anyway, my advice on staying motivated comes down to a few broad topics:

  • Be output focused
  • Treat it like learning an instrument
  • Get meaningful feedback
  • Make it stick

Be output focused ​

A co-worker at the university I taught at was fond of quoting this line, now so am I. Every time you watch a tutorial, download a hip, don't just look at it, nod, then close it. Render something, have a screen capture, a flipbook, a rendered still, a full render, of the thing you learned. The quality of the output is up to you, it should be something doable, but strive to have something to show. If you do that for every little Houdini trick you learn, you too can have a wiki full of content that in itself is a great showreel, and a great record of your progress.

Treat it like learning an instrument ​

For some reason people 'get' the concept that if you learn the guitar, there's established things you need to cover first, established waypoints you need to aim at. Broadly it'd be

  • Practice things by rote (scales, chords)
  • Practice improv (solos, making up song structures)
  • Practice playing with others
  • Find a music style you like, focus more attention on that style you want to be

There's loads of theory and opinions on this, but I'd wager most teachers and musicians would agree on those points. I say the same applies to Houdini.

  • Practice things by rote (following along with tutorials, pulling apart hips)
  • Practice improv (can you build a setup from scratch without looking at a tut? Can you modify an example to do something different and cool?)
  • Practice playing with others (eg work with a modeller to ingest their work, do some cool fx stuff, pass your work to a lighter to render it, hang out on a discord to show your work and get critiques)
  • Find a houdini niche you like, focus more attention on that (Charfx? Mograph? Water fx? Lighting? Procedural game tools?)

Meaningful feedback ​

So you have some output, great. What are you going to do with it?

Some people when learning to draw are lucky enough to have a strong and balanced internal critic; they can see whats wrong, correct it, notice weak points in their skills, improve it. The same applies for playing guitar, or going to the gym, or pretty much any skill.

For the rest of us, we often don't have that ability, you don't have perspective on the work you've spent hours doing, and you don't know what you don't know. Hence, asking for feedback. But from where? Family and friends are handy, but if they don't have knowledge of the thing you're learning, their feedback might be empty at best ("I think it looks great son, well done!"), or downright wrong at worst ("I heard everyone is using Unreal these days, you should learn that instead!")

Get feedback from worthy sources, develop your ability to sort good sources from bad sources. Some discords are great for that (I think the cgwiki one is good), maybe getting advice from twitter or youtube comments could be less great. Understand that you need to hear useful critique to get better, you have to seek it out, but also be aware of how meaningful that source of feedback is.

Make it stick ​

I read this and it clarified a bunch of ideas I've had about learning for ages. The book is great:

https://www.amazon.com.au/Make-Stick-Science-Successful-Learning/dp/0674729013

This video summary is also very good:

https://www.youtube.com/watch?v=enI66OSpQTU

Some of the ideas overlap with things I've mentioned above, and my summary here won't do it justice, but broadly its

  • Learn from as many different source types as you can, find methods/teachers/sources that work for you
  • Put it in practice as soon as you can
  • Use spaced self testing to ensure the knowledge is retained
  • Get used to the feeling of 'arrgh, I know this, its on the tip of my tongue...'

Another line I like to quote is 'if it were easy, everyone would do it'. I think Houdini is easier to learn than people claim it is because it is (mostly) very internally consistent, but it does require full brain attention to use it to its full potential. That can be hard to keep in mind when you're staring at an empty network view, and you have no idea what node to put down first. But stick with it, it's worth it.