Wow, that seems counter-intuitive. I suppose it must be due to the computation of the heuristic, combined with how few relaxations are needed in CH.
GraphHopper is really nice work by the way. I believe you are still working on implementing stall-on-demand? Will be very interesting to see how much difference that makes on long-distance queries.
I guess it is the overhead of the heuristic that is too much compared with the saved visited nodes, but this is counter-intuitive for me as well and if I have more time will investigate this again.
GraphHopper is really nice work by the way.
Thanks!
I believe you are still working on implementing stall-on-demand?
Comments
Wow, that seems counter-intuitive. I suppose it must be due to the computation of the heuristic, combined with how few relaxations are needed in CH.
GraphHopper is really nice work by the way. I believe you are still working on implementing stall-on-demand? Will be very interesting to see how much difference that makes on long-distance queries.
I guess it is the overhead of the heuristic that is too much compared with the saved visited nodes, but this is counter-intuitive for me as well and if I have more time will investigate this again.
Thanks!
I've not yet digged into it but others have with less success but would be very interesting: https://github.com/graphhopper/graphhopper/issues/240#issuec...