In the last tutorial in this part, we’re going to cover the one remaining VRML node that we don’t know about - LOD. OK, I hear what you’re saying - all the other node names made sense! What the hell does LOD mean? Well, LOD stands for Level Of Detail, and it allows you to display different objects depending on how far away the user is.
Why on earth would you want to do that? There are a few reasons - first of all, you could create cool effects with it. That’s not a great reason for an entire node though. After all, you could create your own with a ProximitySensor, Script and Switch. The real reason is for speed. If you want a world that displays quickly, you need to keep the number of faces in the scene down. Now, imagine you’ve got a virtual art gallery, with a very intricate sculpture in one corner. The sculpture has thousands of faces, and is very detailed. Now, from the other side of the room, you can’t see all the detail, but it’s still being drawn, thus slowing down the browser. If you only displayed the fully detailed model when the user was close to it, and used a low-detail version the rest of the time, you wouldn’t waste time displaying detail that you can’t see. Well, this is exactly what LOD does. If displays one of a list of nodes, depending on the distance of the user from a particular point.
So, how do you use the LOD node? Well, it’s really very very simple. It’s very similar to the Switch node, but it selects
the choice automatically. Let’s take a look at the definition of the node:
LOD {
exposedField MFNode level []
field SFVec3f center 0 0 0
field MFFloat range []
}
As you can see, LOD is a pretty simple node. Firstly, the level field is the list of nodes that
are displayed. These can be any valid child node, but bear in mind that the LOD will only display one of
them at a time. If you want to display multiple objects, put them in a Group. center specifies the
point from which the distance to the user is measured. This should normally be the centre of the object. The only other field
is range. This field specifies the distances at which the level changes. The values in this field should be greater
than zero, and specified in increasing order. So, lowest distance first, highest last. There should be 1 less value in this field
than there are nodes in the level field. If the distance from the user to center is lower than the first
entry in range, the first node in level is displayed. Otherwise, if the distance is lower than the second
entry in range, the second node in level is displayed, and so on. If the distance is
not less than the last entry in range, the last node in level is displayed.
You can see the LOD node in action in this example, and see how we’ve coded it here. There are three separate objects for the different levels of detail. ‘lump_near.wrl’ is the highest detail, ‘lump_medium.wrl’ is a medium version, and ‘lump_far.wrl’ is a very low detail model. If you switch between the Viewpoints in your browser, you can see the change in model during the transitions. If you are closer than 15 metres, you see the high-res version; if you are closer than 50 metres away, you see the medium-res version, and otherwise you see the low-res version. Simple.
There’s one more thing you can do with LOD. If you don’t enter anything in the range field, the browser should work out which level to display in order to keep up a good frame rate. If you don’t care when levels change (different browsers and different systems will change level at different times), then you can use it for automatic optimisation. However, I would say it’s safer to specify your own range settings, as this feature is probably not supported in all browsers.
OK well, that about winds it all up for Part 3, VRML Animation. We’ve covered every node in VRML, and now we can do some serious scripting and start creating some really special effects. Hold on!