Rendering with Houdini requires a full or an "Engine" license per client.
If you use Mantra, then it require a Mantra render license.
Please enter the number of your Houdini licenses in rrConfig to verify that RR does not start more instances that you have licenses.
You can use takes in Houdini.
Do not forget to use $ACTIVETAKE in your file names if you want to use one output node for multiple takes.
Note: If you set your mantra node to a take, then the rrSubmitter will load this mantra node with this take.
If you have not set a specific take (set to "current"), the rrSubmitter creates a job for each take.
Example scene:
www.RoyalRender.de/download/web/houdini_take.zip
The render config file RR\render_apps\_config\3D06__Houdini_Python.cfg executes a python script with Houdini Hython.
It uses the custom job variable "CustomScript" for the name of the script file to execute.
Test run:
For a test you can try this:
•Place any ROP in the scene.
•Use the default "Submit RRender" menu item to submit jobs in Houdini.
•Check only one ROP as job if there are multiple ROPs.
•Change the renderer to "Python" in the lower left rrSubmitter.
•At the top, expand the Scene Settings.
There is a button to edit the Custom Variables.
Add a variable with name CustomScript and the python script as value.
•You can add 3 more variables named
CustomFlagA
CustomFlagB
CustomFlagC
•Use any kind of commandline flag/arg parser in your python script to read the commandline flags
-scene "<Scene>"
-FlagA "<CustomFlagA>"
-FlagB "<CustomFlagB>"
-FlagC "<CustomFlagC>"
Production:
We recommend to create a submission .xml file.
Or you can pass all required arguments as commandline flags for the rrSubmitter/rrSubmitterconsole.
In this case you have to start the rrSubmitter or rrSubmitterconsole with the commandline flag -NoAutoSceneRead.
Please see the rrHelp section rrSubmitterconsole for instructions for both submission types.
You have 3 options to get arguments into the script:
1.Some scripts support 3 custom arguments <CustomA>,<CustomB>,<CustomC>.
See Custom Job Variables
2.Set environment variables and read them in your script.
Please see rrHelp section Customization/Render Environment
3.Use the rrJob setting AdditionalCommandlineParam.
In this case use the Submitter option "AdditionalCommandlineParam=0~0~Myflag1 Moreflags EvenMoreFlags".
or set it via the submitter or rrControl UI (you might need to press the Show All button).
Note: This option is limited to 250 characters, NON-unicode.
If you use some scripts very often, we recommend to override the default icon as well.
The icon has to be a png, 14 pixel high and 14-28 pixel wide, located in RR\render_apps\_config\
XML submission: <OverrideIcon> NoPathFileName.png </OverrideIcon>
Commandline: -OverrideIcon NoPathFileName.png
•Submit a render.
•Check the render log in rrControl.
•Search for the line "* HtoA source path is set to".
•Create that folder
•Open the folder with your HtoA installation on your local machine.
For example the folder c:\solidangle\htoa\ htoa-3.2.0_r9e1313b_houdini-17.0.352\ htoa-3.2.0_r9e1313b_houdini-17.0.352\
Copy the contents of that folder (Do not copy the folder htoa-3.2.0_r9e1313b_houdini-17.0.352, copy the contents of the folder. (Subfolder "arnold", "configs", ...))
Paste them into the folder that RR has shown in the render log file.
•Submit a render
•Check the render log in rrControl.
•Search for the line "* RedShift source path is set to ...".
•Create this folder
•Copy the contents of your local Redshift folder into that folder. (On Windows, Redshift is installed in c:\ProgramData\Redshift\, on linux /usr/redshift/ )
•Note: Redshift does not supply a new plugin for each Houdini version. But RR expects to have a plugin matching your Houdini version.
Search for (a different line) "Redshift Houdini plugin folder is set to" in the render log.
If the folder does not exist, then please duplicate the next lower version and name it according to the version that you use.
It is possible to export .rs files on a machine that does not have any GPU installed.
Redshift does not need a GPU for writing .rs files, but the Redshift plugin requires the NVidia libraries for its startup.
You have to copy the NVidia libs from a machine with a GPU into the Redshift/bin folder.
On Windows, the libs are located here: C:\Windows\System32\nv*.dll
Note: If you use the same Redshift folder for rendering, then you should ensure that all GPU machines use the same NVidia driver version as the one you copied to /bin.
•Submit a render.
•Check the render log in rrControl.
•Search for these two lines
* Renderman ProServer source path is set to ....\render_apps\renderer_exe\houdini\win_x64\RendermanProServer-.....
|
•Create these folders
•Search for the folder with your Renderman installation on your machine.
For example the folders
c:\Program Files\Pixar\RenderManForHoudini-23.4
c:\Program Files\Pixar\RenderManProServer-23.4
Copy these folders into the Royal Render folder. The folder names should match the stated folder in the render log
•Pixars Renderman license
Please copy your pixar.license file (that should contains your license server name and some other information) into these two new folders.
Or you can define the path to the license file in RR\render_apps\_setenv\all\houdini__inhouse.rrEnv, see line PIXAR_LICENSE=
No output via RR?
We have seen that Renderman 23.4 does not print any error message in batch render. Example: "Camera does not exist".
But you can change your ROP to "Render current frame" and try render inside the Houdini UI.
In this case the node displays a small error icon. Just hover over the ROP with your mouse and click on the error icon that appears on the left side of the node to see the error message
Renderman not loaded?
There have been cases in which Pixars Renderman was installed on the artists workstation, but the artist was using the Renderman ROP version that was shipped with Houdini.
This is an example with Houdini 18 and Renderman v23. As long as Houdini does not update the included Renderman version, you can see the difference in the logo of the ROP: