Patrick Kilian wrote: >> Looking at the code for _unstable (which is identical to stable at the >> moment, IIRC,) there's a function to recalibrate maxTilesetComplexity >> based on processing time. IMO, it's fixing one problem (people not >> having a good idea of where there complexity should be,) but causing >> others (I see a couple things I consider logic errors, and it ignores >> the fact that, for some of us, the bigger problem with complexity is >> available RAM and not available CPU cycles.) > I'm the one who is to blame for that code. Input is welcome.
For one thing, it seems to ignore the "0" setting. That should not be done. I had it at 0 and it started declining tiles with a complexity of just 100000. On the other hand: why do we need such a setting? People will see soon enough if their setting is too low (or at least: they should notice). Maarten _______________________________________________ Tilesathome mailing list [email protected] http://lists.openstreetmap.org/listinfo/tilesathome
