First of all - thank you sharing those videos for those of us who can't participate to the webinaar

My question is: how does the cooking work in Houdini?

In one point, Scatter is used to distribute points to the grid, and then later, stamping is used in Copy-sop to add rotation information to those points: when its time to “evaluate” - don't know the correct word for this in Houdini lingo - the Copy-sop, is the Copy evaluated as many times as there are points? Where does Houdini know how many times the $PT returns a valid value?

The forrest exsample was nice - it really shows the power of Houdini!

It's the Copy SOP itself that determines how things evaluate within the Houdini geometry engine which sorts out all the dependencies for you.

There are two types of SOPs: Generators (create geometry) and Modifiers (modify input geometry). All modifier SOPs (such as the Copy SOP) have a dependency to the geometry input in to it. If the geometry changes up stream, then this SOP is marked “dirty” and has to “cook” (evaluate). Simple.

SOPs can also have dependencies via expressions such as the stamp() function when it points at a Copy SOP, channel references and data expressions such as point(). Not so simple but there's over 25 years of R&D to make this all happen for you so again, no worries. If you wish to get a sence of just what is evaluating, you can open up the Performance Monitor and press the record button then change a parameter to see what other operators evaluated.

You can MMB on any SOP icon and choose the dependencies menu option to see what dependencies and references exist for this node. You can also open up the network display preferences (eye icon or press “d” in the network view) and go to the Dependencies tab and turn on the various dependency links and options to see additiona visual cues for external dependencies beyond the input geometry dependency.

Each SOP is it's own procedure and each SOP has it's own methods. Once you know how to wire two SOPs together and can crawl about the Houdini interface, you focus on learning each SOP as it's own little programme procedural. Many SOPs have help cards that show you standard set-ups, what SOPs normally go in conjunction with the current SOP in question.

As for the Copy SOP, it is right on the top of the list as a key tool in Houdini as it allows you to create plain duplicates, visualize complex instance set ups and do stamping on upstream nodes (or any nodes actually) to do powerful data construction and a whole bunch more.

The Copy SOP does plain old duplciates if you don't use the second right template input. As soon as you wire geometry in to the second template input, you can use a few local variables in the Copy SOP (see the Copy SOP help card on Local Varaibles) such as $PT and $ID which allow you to access the point number and the particle id attribute number in any parameter to the Copy SOP.

Again study the Copy SOP's help card and spend some time going through the help cards, reading all the node comments (MMB on the SOP icon) and the notes.

Thank you!