The first set of hints are specific to qemLOSS3 and some of its new capabilities. These are some tips you should be aware of and will provide some reinforcement to things that were hinted at in the tutorials, and maybe even mentioned explicitly, but need to be reinforced one more time. The second section of hints and tips will include items applicable to all versions of qemLOSS.

• If your model has Endomorphs, you want to make sure the base position is being displayed before running qemLOSS3 by choosing "(base)" from the M pull-down menu at the lower right corner of Modeler. After running qemLOSS3, the reduction process usually finishes with one of the Endomorph positions being displayed in Modeler's viewports, rather than the base Endomorph position. I tried to force the qemLOSS3 plugin to select the base position before and/or after the reduction process, but there seems to be an oversight in the Lightwave SDK that doesn't allow me to select the base position (the plugin can select any of the morphed positions, but I could not figure out the proper method to select the base position).  Especially if you have already run qemLOSS3 on your object, and you want to try another reduction, please verify that you have the base Endomorph displayed in the viewports. This is very important, otherwise qemLOSS3 will assume the position being displayed is the base object, and it will perform the reduction as if that is the base object and all relative morphs will be added to that position. (With the proper spin on things, maybe this could be touted as a feature of the plugin :-).

  
   

• You can use the Weight Map Reduction Constraints to reduce just a selected portion of your object. Select all vertices of your object and create a new Weight map (why not call it qemloss) with high initial weight values, 100% or more. Now select that new Weight map and then select all the vertices in the area or areas you want to reduce. Assign a low weight value (possibly even 0 or a negative value) to those vertices. You will also want to make a note of the number of polygons found in the areas you want to reduce and how many polygons should exist in those areas when the reduction is finished, because the difference in those numbers will be needed to calculate the final Reduction Goal . Run qemLOSS3, change the Reduction Goal (it's probably easiest to use the Polygons requester), and enable the "Use Weights as Reduction Constraint" parameter under the WGHT tab. You can use the "Weight Maps" pull-down and "Multiplier" parameter to provide a specific multiplier to that one Weight map if more precise control is needed.

  
   

• Selection Sets (PICK maps) are not discussed at all in the tutorials, but they are just a special case of vertex maps. Any Selection Sets assigned to your object will be preserved during the reduction, but the only controls available for the reduction process are the Discontinuous Vertex parameters. There are no numeric values assigned to a Selection Set that can be used for the geometric error metrics. So unlike the discontinuous TXUV, WGHT, RGB, and RGBA maps, discontinuous Selection Sets will not automatically be handled by the reduction process. However, you can add geometric error weight to the discontinuous vertices found in your Selection Sets in the same way that geometric error weight was added to the discontinuous vertices in a UV map in the second tutorial.

These are some hints and tips from users of qemLOSS3 and previous versions of qemLOSS. If you have had some success using the qemLOSS polygon reduction plugins in a project, and would like to share your own ideas, please send me the tip or hint (marvinl AT email.arizona.edu), and I will be glad to add it to this list. I enjoy receiving feedback from users, so any thoughts you might have are definitely welcome!

• One of the very first users of qemLOSS3, Alexandre Labedade (Alexx) has written a very useful script called Weighter that allows the use of a simple weighted shape to apply weight on a more complex object. Currently the script can be used to simplify the process of assigning weight values to the complex object's Subpatch Weight Map. The object's Subpatch Weight Map can then be used in qemLOSS3 to constrain the reduction of various parts of the object, similar to the standard Weight Map techniques described in the third tutorial.

  
   

• qemLOSS is a good way to produce triangle-only models for applications that require them. The first major step qemLOSS takes is to triple your polygons, which will increase the polygon count if your model contains non-triangular polygons. By providing a Reduction Goal of 100%, the model will be reduced back to the original number of polygons, but now they will all be triangles. If that reduction doesn't preserve your model's shape well enough, don't hesitate to try percentages over 100%, those models will still be smaller than the tripled original. Thanks to Ernie Wright for this tip.

  
   

• This plugin is an excellent way to create low-res stand-in objects for scene layout, or for multi-resolution models and the LOD plugin. Several resolutions of an object can be created quickly and easily using qemLOSS, and then LOD can use the object replacement capability in Layout to substitute a low resolution object in place of an object with a high polygon count when it is far away from the camera.

  
   

• Jon "shine" Jones has created a web page titled Reducing Polygons in Lightwave. He discusses various techniques for discovering polygons that can be eliminated in many Lightwave models. His final step uses qemLOSS to reduce the model to its final polygon count. And even though I like people to read and understand this qemLOSS document, it's still nice to hear his comment about qemLOSS: "...and it'll reduce the polys and put it in a new layer. Check the poly count. 400? But the model looks the same! How can this be? I don't care, it works, that's all that I care about. ;)".

  
   

• Larry Shultz specializes in 3D character modeling and animation, and creates very nice low polygon count characters using spline cages. His online Spline Patching Tutorial describes his method for creating these 3D characters. Larry also posted information to the Lightwave Mailing List about how he uses qemLOSS during the process of creating lower level of detail objects for his animations.