The question with this is whether there is enough time when less than the ideal number of sub-processes are active to make the added complexity and memory usage worthwhile. I've been experimenting on my system with a change where the largest zoom levels are assigned to the first (n-1) children and all remaining zoom levels to the last child. On my two-core machine, rendering tiles for z17 takes about as long (actually slightly longer) as rendering z12-z16 combined, so there is much less time with a core sitting idle than with the stable client's method.
On Wed, Dec 17, 2008 at 13:32, Matthias Julius <[email protected]> wrote: > Matti Viljanen <[email protected]> writes: > >> Hi >> >> Just crossed my mind... >> >> I have a quad-core processor, so I no doubt use Fork=4. Using Fork >> sure cuts rendering time, but I think it still could do better. I'm >> referring to this kind of output: >> >> [...] > > The development version of the client has a threading branch > (tilesAtHome/branches/threading) where René Wunderlich is trying to > improve the performance of the client. > > The current forking scheme is clearly sub-optimal. > > Matthias > > _______________________________________________ > Tilesathome mailing list > [email protected] > http://lists.openstreetmap.org/listinfo/tilesathome > -- David J. Lynch [email protected] _______________________________________________ Tilesathome mailing list [email protected] http://lists.openstreetmap.org/listinfo/tilesathome
