How-To Prepare a scene for LightNet:Cloud

For an onsite scene, the content information is simply a path, you do not need to package and zip the content for onsite rendering. However, this does mean that all onsite PCs must have access to this path. The easiest way to accomplish this is to have a mapped network drive to the content's location. Full network paths with also work. So \\FileServer\CDrive\MyContent will work or Z:\MyContent will work as long as Z:\ is mapped to \\FileServer\CDrive

When a scene is ready for offsite rendering, zip up the content directory's root in an archive.  LightNet will create a new directory on the offsite PCs and unzip the sub directories into it. The easiest way to do this in Windows is to go into the content directory, select your sub-folders, right click and select “Send to Compressed Folder.”

Zipping content directory

Make sure to give your content directory zip file a unique name. If you already have a MyContent.zip that you used for another scene, name the new one MyContent2.zip next MyContent3.zip and so forth. Unless you have multiple scenes that use the exact same content directory.

Lightwave also has a built in content extracting system that exports all of the used objects, image and so forth to a new directory. You can then zip up the contents of that directory. In Lightwave 9.6, the plugin is under Additional -> Content Manager

In Lightwave 10, it is under File -> Package. (Will be adding more information on this soon)

I've been told 9.6's packaging system is unstable and doesn't always work. In any event, as long as all of the files needed by the scene is in your zip file, the scene will render fine. You can easily zip your entire content directory.

In order for the Watcher to pick up the scene properly a content information must be placed in the scene. To do this, I have created a LScript. Add the LScript to Lightwave and add is as a Master Plugin, it will show up as LnC_Settings:

LScript01

Double-click it to bring up it's settings.

LScript02

For an onsite scene, simply put the path in the Content field

LScript03

For an offsite scene, simply put the zip file name in the content field

In the LScript you also have the ability to select the number of segments you want the scene split into. For 1x, it will render normally. For 4x, it will split into 2x2, 16x will be 4x4 and 25x will be 5x5. For segmentation the scene will actually be split into multiple files with a new naming scheme; a letter for the column number and number for row number. For example suffix A1 will be first column, first row, B3 will be second column, third row, D5 will be fourth column, fifth row and so forth. Currently LightNet: Cloud does not reassemble the frames.

Now set your start and end frames in the scene. LightNet: Watcher will use the scene start and end frame settings when it adds the scene to the render queue. If you want to render a single frame, such as if you're rendering a segmented scene where you only need the one frame, set the start and end frame to the same number. So if you only want to render frame 10, set start and end frames both to 10.

Finally you need to enable output on your frames. On your render output select RGB and/or Alpha settings and choose your output formats. The path is irrelevant, LightNet: Cloud will overwrite the output path in order to send the frames back to the sever. But the options MUST be checked or the frames will render with NO output.

Starting with version 1.5, there is also an option to preserve the output prefix, this is filename you enter when you set the output settings. If you choose not to preserve the prefix, LnC will simply use the scene filename as the output.

Starting with version 1.8, you can enter a group or groups of computers to render the scene. Groups are separated by spaces. Each computer, when loading the scene, will check its groups and the scene's groups and if one of the groups match, it will render the scene. If it does not find a match, it will skip the scene and begin checking the next one.

The final frames will still be uploaded to the FTP server, so the path you set for the output files in the Lightwave scene is totally irrelevant, but with this option, the name will be preserved.

Now, simply copy the scene files - and content zip file if it's an offsite scene - into the Watch Folder and the LightNet: Watcher will instamagically begin the render process on it's next check. (IMPORTANT NOTE: In Windows if you drag and drop a file from one folder to another folder on the same drive, by default it will MOVE the file, not copy it. LightNet: Cloud will delete files in the Watch Folder after it has uploaded them, so be sure you COPY the scene files and don't MOVE them)

A note on frame segmentation. If you have a scene where frames have very long render times, do not use frame segmentation in order to make them render faster, they will not. The segment time will be shorter, of course, but the overhead of splitting, downloading and uploading segments and then reassembling them will BY FAR outstrip and kind of perceived faster rendering. Frame segmentation is only useful in splitting very large, individual frames, such as for print graphics.