|On this page|
The Wedge ROP will render the specified ROP multiple times. Each time it will vary some parameters in the .hip file. This is useful for exploring how changing parameters will change the look for expensive functions.
The WEDGE variable will be set to a string that describes the current
set of parameters. Using this in filenames output can ensure each
wedge doesn’t conflict with the other wedges. Alternatively, if one
is wedging to MPlay, the individual wedges will render together. The
output file could be set to
filename_$WEDGE_$F.pic to ensure each
wedge gets its own image sequence.
If wedging images, you can also use the hwatermark program to stamp the images with the wedge name. Setting the Post Frame Script to
unix hwatermark -x 4 10 -m \"Wedge: $WEDGENUM\" `chs("vm_picture")` `chs("vm_picture")` $HFS/houdini/fonts/Helvetica-Bold.pfa 24
The WEDGENUM variable will be set to the currently processed wedge number. This ranges from 0 to one less than the number of wedges. If you render a specific wedge with the Wedge Range, WEDGENUM will be set to that wedge number.
Runs the specified render operation once for each wedge.
This prefix is added at the front of the WEDGE variable. This can be used to separate different sets of wedges.
All Wedges will cause every computed wedge to be rendered at once. Alternatively, one specific wedge can be rendered. The latter is useful if manually splitting up the wedges across machines, which may be necessary for simulations.
Which ROP will be run for each wedge. The ROP should have a parameter called "take" for take based wedging to work. If it outputs to a file, the filename should have $WEDGE in it to avoid writing the same file twice.
Wait for Render to Complete
Sets the Block Until Render Completes temporarily on the Output Driver. This avoids all the wedges being started simultaneously, which can inconvenience a single-user machine. However, by not blocking, you can return control to Houdini quickly while waiting for the renders to run in background.
Controls how the scene is varied.
Each of the Wedge Parameters is varied according to its number of steps and range and applied to its specified channel. All combinations of the parameters are then wedged. The total number of wedges is the product of all of the step values.
One wedge is performed for each immediate child of the given parent take. The output driver’s take parameter is set to each child take’s value.
In By Take wedging, this is the name of the parent take. Only immediate children of this take are used for wedges.
Instead of uniformly sampling each range, a random value will be picked inside the range for each sample.
Number of Samples
Determines how many wedges will be run when in random sampling mode.
The seed used for generating the random samples. Changing this will create a different set of random samples.
The number of parameters to use in By Channel wedging. Each parameter is a single float value which is divided uniformly across a given interval.
The english name of the channel. This is added to the WEDGE variable. If blank, the channel name is used.
The channel path to vary. This is a full channel path, a path
to a node followed by the name of a channel. Note that it is
not a parameter path:
/obj/geo1/tx is legal but
The minimum and maximum values that the channel will vary over.
Number of Steps
The number of steps that the range will be broken into. Ignored if random sampling is performed.
This example shows how to setup the Wedge ROP to automatically create a bunch of variants of a network.
The following examples include this node.