- Notifications
You must be signed in to change notification settings - Fork 3.8k
Open
Labels
Description
I'm usually not interested in route information apart from duration, distance and geometry, so I'm always using OSRM with steps=false. My understanding is that the matching data used to generate the steps array in response (RouteStep.name and RouteStep.ref values among others) is still loaded in RAM. Depending on the extract size, this may amount to a lot of unused data.
First of all, I'd like to figure whether trying not to load this data is worth the trouble. I can see osrm.names is not that big, but osrm.icd is. Is there a way to evaluate the potential gain here?