|Page (1) of 5 - 05/15/03||email article||print page|
The LightWave Render FarmRunning Screamernet II on Mac OS X
In this tutorial, we'll be dealing with a simple, hypothetical network configuration, the basics of which are shown in fig 1.1. The term "Host" refers to a computer you designate as your main controller, usually the principle workstation in your setup. "Nodes" are satellite computers, connected via Ethernet to your Host, whose processor power you'll be utilizing for the purposes of network rendering.
Fig. 1.1--This is the network setup used for our tutorial.
Exactly the same principles would apply to a network using more or fewer 'Nodes.'
Our Hypothetical Project
Our hypothetical Content Folder is called [the-worm] and is located on our Host computer. All our Images, Objects and Scene files lie directly within this folder. We strongly recommend you create your own simple project folder, with a very simple scene file correctly pathed for saving within LightWave. It will help you test ScreamerNet (LWSN) a little later.
Fig. 1.2 shows where [the-worm] is located within a disk partition on our Host (called a "Volume" in network lingo). This Volume is a top-level directory which we've given a unique name on our network [WOA_System]. It's a good idea to break down your directory structure logically like this, as it becomes invaluable a little later when we come to re-writing LightWave's CommandLine Files.
Fig. 1.2--This shows the location of a Content Directory [the-worm] used for this tutorial.
Unique naming of your Host computer's partition(s) is also a very good idea.Later we'll see that LWSN (ScreamerNet) cannot directly distinguish betweena local Volume called, say [HardDisk], and a Volume it sees across the network,also called [HardDisk]. This can lead to some confusing behavior!
OK, finally for the introduction, a breakdown of what we'll be tackling during this tutorial and why. It should help clear your head a little before the real mindf**k begins!
Concepts: A very brief description of the workflow involved in our method of network rendering. If it seems confusing at this stage don't worry, we'll gradually build up the amount of detail as the tutorial progresses.
Config Paths: How to successfully move your Config files (Preferences) for LightWave applications. This is a great way to get to grips with concepts important to network rendering and should also improve your general LightWave workflow.
LWSN Host Setup: Setting up your main controller machine (Host) ready for network rendering. We'll be looking at both the system side (file sharing and networking) as well as the application side (creating a re-useable setup for LWSN and giving you a solid understanding of the concepts involved).
LWSN Node Setup: Setting up a single Node machine quickly and efficiently. There's really not much work involved here. We'll show you how to get the networking side of things working correctly and give you a good understanding of what tasks the Node must undertake. Once this is done, you'll simply duplicate the process for all Nodes on your network.
Rendering: We'll be able to start rendering, which means the hard work is done. Most of the slog comes in setting up LWSN for the first time and so once you have scenes rendering successfully, you can kick back and watch the frames arrive!
Troubleshooting: Finally, we've created a section that tackles some of the most common queries and problems confronted when using LWSN. This is done on a Problem--Solution basis.
Let's just briefly outline the concepts behind network rendering in this tutorial.
The program LWSN is the basis for network rendering using LightWave. You cannot simply do it from within LightWave Layout. You'll find LWSN within the [Programs] folder of your LightWave installation.
LWSN needs to know the location of two things in order to work correctly:
- LightWave's Preference file, often referred to as a Config File.
- The Content Directory from which you wish to render. In this tutorial we're working with a Content Directory called [the-worm] inside which all our Images, Objects and Scenes are located.
We tell LWSN where these two items are located by correctly formatting a document called LWSN CmdLine. You'll find an original of this file in the [Programs] folder of your LightWave installation.
Each Node machine will run its own instance of LWSN, with its own slightly different version of the LWSN CmdLine telling it what to do. All these instances will be located on the Host and will be run, across the network, from each Node. This is the principle behind this method of network rendering.
Fig. 2.1--The program LWSN (ScreamerNet) uses its CmdLine file to find out the location of
LightWave's Preferences and your current Content Directory.
Related Keywords:The LightWave Render Farm Running Screamernet II on Mac OS X