Re: [OSM-dev] GraphHopper Maps 0.1
I took the liberty of comparing the result on that page to the direct call. How realistic are the results? If I run a route e.g. from Heroldsberg to Biograd, I get greatly different times: http://apmon.dev.openstreetmap.org/routing: 9,5s directly on http://graphhopper.com/maps/: 0.017s So it seems that the result of the compare page is off by a factor of more than 500 times. Why? bye, Nop The actual requests get sent from the server backend on the osm dev server (rails app) and are then transcoded from json to kml before getting passed on to the user. Unfortunately, the dev servers rails system seems to be incredibly slow! Even just simply loading the pages takes a long time. Would be nice if you could somehow include the actual response size and time in the output to have a feeling for that too. Would you mind to print an error message if some vehicles or routing types are not supported instead of printing results for the default? Eg. graphhopper does not support shortest in the web API and OSRM does not support foot etc PS: there seems to be a small bug when I enter a new location so I have to click twice on 'find route'. ___ dev mailing list dev@openstreetmap.org http://lists.openstreetmap.org/listinfo/dev
Re: [OSM-dev] GraphHopper Maps 0.1
Hi Am 24.07.2013 06:52, schrieb Kai Krueger: I was curious to see how it compares in speed and quality of calculated routes to the other engines, so I took the liberty to add it to http://apmon.dev.openstreetmap.org/routing It seems the labeling is wrong. When using the Grasshopper Router it always prints Route was calculated by Cloudmade. Peter ___ dev mailing list dev@openstreetmap.org http://lists.openstreetmap.org/listinfo/dev
Re: [OSM-dev] GraphHopper Maps 0.1
Peter Körner wrote Hi Am 24.07.2013 06:52, schrieb Kai Krueger: I was curious to see how it compares in speed and quality of calculated routes to the other engines, so I took the liberty to add it to http://apmon.dev.openstreetmap.org/routing It seems the labeling is wrong. When using the Grasshopper Router it always prints Route was calculated by Cloudmade. Should be fixed now. This is only a demonstration page though and not as powerful as the web interface that comes with OSRM or Graphhopper. Although I do very much hope that something like this will eventually make its way onto osm.org. Kai -- View this message in context: http://gis.19327.n5.nabble.com/GraphHopper-Maps-0-1-tp5771017p5771183.html Sent from the Developer Discussion mailing list archive at Nabble.com. ___ dev mailing list dev@openstreetmap.org http://lists.openstreetmap.org/listinfo/dev
Re: [OSM-dev] GraphHopper Maps 0.1
Hi Peter, Nice to see new stuff. Works nice in long distances. I noticed that it does not like roads with tram rails. (highway=*, railway=tram) See http://graphhopper.com/maps/?point=56.949977%2C24.120226point=56.952645%2C24.122061 Viesturs On Tue, Jul 23, 2013 at 10:22 AM, Peter K peat...@yahoo.de wrote: Hi there, yesterday we released the first public version of our fast and Open Source routing engine called GraphHopper. This could be especially interesting for Java developers. You can also try our web applicationhttp://graphhopper.com/maps/with world wide coverage. See the full anouncement here.https://karussell.wordpress.com/2013/07/22/graphhopper-maps-high-performance-and-customizable-routing-in-java/ Let me know if you encounter problems or if you have questions! Regards, Peter. ___ dev mailing list dev@openstreetmap.org http://lists.openstreetmap.org/listinfo/dev ___ dev mailing list dev@openstreetmap.org http://lists.openstreetmap.org/listinfo/dev
Re: [OSM-dev] GraphHopper Maps 0.1
Thanks! I've created an issue for it! Peter. Hi Peter, Nice to see new stuff. Works nice in long distances. I noticed that it does not like roads with tram rails. (highway=*, railway=tram) See http://graphhopper.com/maps/?point=56.949977%2C24.120226point=56.952645%2C24.122061 Viesturs On Tue, Jul 23, 2013 at 10:22 AM, Peter K peat...@yahoo.de mailto:peat...@yahoo.de wrote: Hi there, yesterday we released the first public version of our fast and Open Source routing engine called GraphHopper. This could be especially interesting for Java developers. You can also try our web application http://graphhopper.com/maps/ with world wide coverage. See the full anouncement here. https://karussell.wordpress.com/2013/07/22/graphhopper-maps-high-performance-and-customizable-routing-in-java/ Let me know if you encounter problems or if you have questions! Regards, Peter. ___ dev mailing list dev@openstreetmap.org mailto:dev@openstreetmap.org http://lists.openstreetmap.org/listinfo/dev ___ dev mailing list dev@openstreetmap.org http://lists.openstreetmap.org/listinfo/dev
Re: [OSM-dev] GraphHopper Maps 0.1
Hi, Here's some feedback: 1. It seems turn restrictions are not taken into account in the routing? 2. It seems that the internal routing graph has too few nodes from where you can start or end. I think this might be a result of the optimizations you have performed to make the router fast. So if you wanted to route to a POI along a one-way street, you might get routed to the point at the end of the one-way street and the route does not pass through that one-way street. Here's an example to make the second issue clearer: http://graphhopper.com/maps/?point=14.557141%2C121.021051point=14.559306%2C121.020455 I would expect the route to pass along the one-way street. Eugene On Tue, Jul 23, 2013 at 3:22 PM, Peter K peat...@yahoo.de wrote: Hi there, yesterday we released the first public version of our fast and Open Source routing engine called GraphHopper. This could be especially interesting for Java developers. You can also try our web applicationhttp://graphhopper.com/maps/with world wide coverage. See the full anouncement here.https://karussell.wordpress.com/2013/07/22/graphhopper-maps-high-performance-and-customizable-routing-in-java/ Let me know if you encounter problems or if you have questions! Regards, Peter. ___ dev mailing list dev@openstreetmap.org http://lists.openstreetmap.org/listinfo/dev
Re: [OSM-dev] GraphHopper Maps 0.1
Hi Eugene, thanks for your feedback! 1. It seems turn restrictions are not taken into account in the routing? yes 2. It seems that the internal routing graph has too few nodes from where you can start or end. I think this might be a result of the optimizations you have performed to make the router fast. So if you wanted to route to a POI along a one-way street, you might get routed to the point at the end of the one-way street and the route does not pass through that one-way street. Here's an example to make the second issue clearer: http://graphhopper.com/maps/?point=14.557141%2C121.021051point=14.559306%2C121.020455 Indeed. This will be fixed with the next release as it is not only important for car but also for hiking etc: https://github.com/graphhopper/graphhopper/issues/27 Regards, Peter. Hi, Here's some feedback: 1. It seems turn restrictions are not taken into account in the routing? 2. It seems that the internal routing graph has too few nodes from where you can start or end. I think this might be a result of the optimizations you have performed to make the router fast. So if you wanted to route to a POI along a one-way street, you might get routed to the point at the end of the one-way street and the route does not pass through that one-way street. Here's an example to make the second issue clearer: http://graphhopper.com/maps/?point=14.557141%2C121.021051point=14.559306%2C121.020455 I would expect the route to pass along the one-way street. Eugene On Tue, Jul 23, 2013 at 3:22 PM, Peter K peat...@yahoo.de mailto:peat...@yahoo.de wrote: Hi there, yesterday we released the first public version of our fast and Open Source routing engine called GraphHopper. This could be especially interesting for Java developers. You can also try our web application http://graphhopper.com/maps/ with world wide coverage. See the full anouncement here. https://karussell.wordpress.com/2013/07/22/graphhopper-maps-high-performance-and-customizable-routing-in-java/ Let me know if you encounter problems or if you have questions! Regards, Peter. ___ dev mailing list dev@openstreetmap.org http://lists.openstreetmap.org/listinfo/dev
Re: [OSM-dev] GraphHopper Maps 0.1
On 23.07.2013 09:22, Peter K wrote: yesterday we released the first public version of our fast and Open Source routing engine called GraphHopper. I'm curious which of the more tricky features of the OSM data model are already supported by GraphHopper. You commented on turn restrictions, but what about: * pedestrian routing across areas? * barrier nodes? * conditional restrictions? * destination and other access values besides yes? ___ dev mailing list dev@openstreetmap.org http://lists.openstreetmap.org/listinfo/dev
Re: [OSM-dev] GraphHopper Maps 0.1
Hi, On 23.07.2013 09:22, Peter K wrote: yesterday we released the first public version of our fast and Open Source routing engine called GraphHopper. I'd like to understand better where this fits in between the purely A* gosmore and the CH-based osrm. With osrm, it is difficult to have multiple routing profiles on one machine because you have do compile a routing graph for each profile - a world-wide setup for car, bike, foot would take something like 150 GB of memory on the server. With gosmore, this is easy since you only need one routing graph to support a multitude of profiles - not only car, bike, foot, but also motorcycle, HGV, and others. This flexibility comes at a noticeable speed penalty. I read that I can run GraphHopper with or without CH. Does that mean that when I run it without, I get the gosmore-like flexibility to evaluate edges at runtime, or is the speed profile baked into the routing graph (and therefore your demo server needs something like 64 GB of RAM because it has three routing graphs)? Bye Frederik -- Frederik Ramm ## eMail frede...@remote.org ## N49°00'09 E008°23'33 ___ dev mailing list dev@openstreetmap.org http://lists.openstreetmap.org/listinfo/dev
Re: [OSM-dev] GraphHopper Maps 0.1
Hi Frederik, I read that I can run GraphHopper with or without CH. this is true. Currently you have the choice between A*/Dijkstra (simple, bidirectional, bidirectional CH) or is the speed profile baked into the routing graph The online demo is CH based and requires 3*16GB. If you'd use a none-CH graph you would have all the flexibility BUT as you already said you have to take care that the shortest path tree won't eat all the available RAM (e.g. use a memory bound dijkstra). Or you'll have to make compromises in the quality/'exactness' of the routes. With graphhopper you have the choice. If you only need a rather smallish area like bavaria (or even germany can work with tuning) you don't need CH. All that was said under the fact that you are using the in-memory data access method of graphhopper. But you can try graphhopper via a memory mapped data access which is the default for Android and you won't need lots of RAM which could be want you want for the CH-based stuff and multiple profiles. But of course this will slow things down too, but I don't know how much (will still depend on how many RAM is available). Regards, Peter. Hi, On 23.07.2013 09:22, Peter K wrote: yesterday we released the first public version of our fast and Open Source routing engine called GraphHopper. I'd like to understand better where this fits in between the purely A* gosmore and the CH-based osrm. With osrm, it is difficult to have multiple routing profiles on one machine because you have do compile a routing graph for each profile - a world-wide setup for car, bike, foot would take something like 150 GB of memory on the server. With gosmore, this is easy since you only need one routing graph to support a multitude of profiles - not only car, bike, foot, but also motorcycle, HGV, and others. This flexibility comes at a noticeable speed penalty. I read that I can run GraphHopper with or without CH. Does that mean that when I run it without, I get the gosmore-like flexibility to evaluate edges at runtime, or is the speed profile baked into the routing graph (and therefore your demo server needs something like 64 GB of RAM because it has three routing graphs)? Bye Frederik ___ dev mailing list dev@openstreetmap.org http://lists.openstreetmap.org/listinfo/dev
Re: [OSM-dev] GraphHopper Maps 0.1
The online demo is CH based and requires 3*16GB Forgot to mention that if one would reuse the base graph of ~9GB one could reduce this to something like 25GB (would take a bit development effort) Peter. ___ dev mailing list dev@openstreetmap.org http://lists.openstreetmap.org/listinfo/dev
Re: [OSM-dev] GraphHopper Maps 0.1
Hi, this looks like an interesting addition to the set of routing engines already available for OSM data. I was curious to see how it compares in speed and quality of calculated routes to the other engines, so I took the liberty to add it to http://apmon.dev.openstreetmap.org/routing That page allows you to use all of the main routing engines (OSRM, YOURS, Mapquest Open, Cloudmade and now Graphhopper) through a single web interface and therefore allows to compare the calculated routes quite easily and how they behave with the various OSM constructs and tagging schema. I hope adding it was OK with you. If not, then I will of cause remove it again immediately. But given that that demo page produces pretty much no traffic and graphhopper seems pretty fast, I thought it wouldn't cause any issues. Kai On 07/23/2013 01:22 AM, Peter K wrote: Hi there, yesterday we released the first public version of our fast and Open Source routing engine called GraphHopper. This could be especially interesting for Java developers. You can also try our web application http://graphhopper.com/maps/ with world wide coverage. See the full anouncement here. https://karussell.wordpress.com/2013/07/22/graphhopper-maps-high-performance-and-customizable-routing-in-java/ Let me know if you encounter problems or if you have questions! Regards, Peter. ___ dev mailing list dev@openstreetmap.org http://lists.openstreetmap.org/listinfo/dev
Re: [OSM-dev] GraphHopper Maps 0.1
Kai Krueger wrote I was curious to see how it compares in speed and quality of calculated routes to the other engines, so I took the liberty to add it to http://apmon.dev.openstreetmap.org/routing I took the liberty of comparing the result on that page to the direct call. How realistic are the results? If I run a route e.g. from Heroldsberg to Biograd, I get greatly different times: http://apmon.dev.openstreetmap.org/routing: 9,5s directly on http://graphhopper.com/maps/: 0.017s So it seems that the result of the compare page is off by a factor of more than 500 times. Why? bye, Nop -- View this message in context: http://gis.19327.n5.nabble.com/GraphHopper-Maps-0-1-tp5771017p5771164.html Sent from the Developer Discussion mailing list archive at Nabble.com. ___ dev mailing list dev@openstreetmap.org http://lists.openstreetmap.org/listinfo/dev
Re: [OSM-dev] GraphHopper Maps 0.1
NopMap wrote Kai Krueger wrote I was curious to see how it compares in speed and quality of calculated routes to the other engines, so I took the liberty to add it to http://apmon.dev.openstreetmap.org/routing I took the liberty of comparing the result on that page to the direct call. How realistic are the results? If I run a route e.g. from Heroldsberg to Biograd, I get greatly different times: http://apmon.dev.openstreetmap.org/routing: 9,5s directly on http://graphhopper.com/maps/: 0.017s So it seems that the result of the compare page is off by a factor of more than 500 times. Why? bye, Nop The actual requests get sent from the server backend on the osm dev server (rails app) and are then transcoded from json to kml before getting passed on to the user. Unfortunately, the dev servers rails system seems to be incredibly slow! Even just simply loading the pages takes a long time. Same thing with OSRM and the other routing backends, which also feel very slow on that page. Sorry, I should have said, that speed comparisons aren't realistic. But never-the-less, comparing the output between the routers should give a valid picture of how they behave. Kai -- View this message in context: http://gis.19327.n5.nabble.com/GraphHopper-Maps-0-1-tp5771017p5771166.html Sent from the Developer Discussion mailing list archive at Nabble.com. ___ dev mailing list dev@openstreetmap.org http://lists.openstreetmap.org/listinfo/dev
Re: [OSM-dev] GraphHopper Maps 0.1
Thanks! Hi, this looks like an interesting addition to the set of routing engines already available for OSM data. I was curious to see how it compares in speed and quality of calculated routes to the other engines, so I took the liberty to add it to http://apmon.dev.openstreetmap.org/routing That page allows you to use all of the main routing engines (OSRM, YOURS, Mapquest Open, Cloudmade and now Graphhopper) through a single web interface and therefore allows to compare the calculated routes quite easily and how they behave with the various OSM constructs and tagging schema. I hope adding it was OK with you. If not, then I will of cause remove it again immediately. But given that that demo page produces pretty much no traffic and graphhopper seems pretty fast, I thought it wouldn't cause any issues. Kai On 07/23/2013 01:22 AM, Peter K wrote: Hi there, yesterday we released the first public version of our fast and Open Source routing engine called GraphHopper. This could be especially interesting for Java developers. You can also try our web application http://graphhopper.com/maps/ with world wide coverage. See the full anouncement here. https://karussell.wordpress.com/2013/07/22/graphhopper-maps-high-performance-and-customizable-routing-in-java/ Let me know if you encounter problems or if you have questions! Regards, Peter. ___ dev mailing list dev@openstreetmap.org http://lists.openstreetmap.org/listinfo/dev
