On 14. Jan 2008, at 20:04, Gordon Erlebacher wrote: > >>> We tried the lucy.ply file. It has 14,000,000 vertices, takes 1.3 >>> Gbytes >>> of memory. Could not get eqPly.64 to work. >> >> Does it crash, or simply not respond? > > It does not crash. It does not respond. Therefore, it might still > be in the process of creating the k-d tree (on the server node > only?) But what is strange, is that I set the range from 0-0.01, > 0.01 - 0.03, 0.03 - 0.06 (to decrease rendering time). But it still > computes the k-d tree of the entire dataset. An alternative appraoch > would be to be less efficient in terms of load balancing, or adjust > laod balancing "during" the interactive session (somehow).
That would be hard (impossible?) since you need to know the data structure beforehand. The whole point of the kd-tree generation is to organize the unordered triangle soup from the ply file into a spatial data structure. Before that, I see no way of knowing which triangles belong to a given range. Another option is to separate the conversion from eqPly. This makes it easier to optimize it and to add better progress reports. Nobody looked at the conversion so far - I am pretty sure it has a good potential of optimization. Cheers, Stefan. _______________________________________________ eq-dev mailing list [email protected] https://in-zueri.ch/cgi-bin/mailman/listinfo/eq-dev http://www.equalizergraphics.com
