tecznotes

Michal Migurski's notebook, listening post, and soapbox. Subscribe to this blog. Check out the rest of my site as well.

Jan 7, 2013 12:31am

loading artifacts, google maps for iOS

The new GMaps for iOS uses the same/similar vector loading process as its longtime Android counterpart. The high-res stuff, including buildings and tiny streets, looks like it comes in right at zoom level 16, which is about where I would’ve put it given the choice.

The transition between zooms is a bit odd. I would not have expected the difference in street widths between the loaded and unloaded sections here. OpenGL does how (as far as I know) have its own concepts like round line caps or joins, so maybe there’s some client-side buffering these that’s expensive to run more than once per geometry load? The “hot dog” street endings show that each tile’s worth of geometry data comes clipped; I haven’t tested to see whether street names straddle z16 tile boundaries or not.

Interestingly, GMaps makes no attempt to model elevated freeways, terrain, or anything else vertical beyond buildings. In use, this feels like a satisfactory compromise. The road names here are in perspective, applied to the ground as textures, while the route numbers are all billboards facing forward.

I’m not sure why the Grand Avenue label below is not tilted back like Elita or Perkins.

The translucent handling of the building is very good. It’s sketchy enough that you can recognize them and navigate without being distracted by their incompleteness.

Comments (1)

  1. If I recall correctly I heard in a google talk about webgl and maps that they were rendering circles to create the roads joins (and endings). I though they were not cutting the geometry but your image points they are.

    Posted by javisantana on Friday, January 11 2013 5:41pm UTC

Sorry, no new comments on old posts.

December 2018
Su M Tu W Th F Sa
      
     

Recent Entries

  1. How It’s Made: A PlanScore Predictive Model for Partisan Elections
  2. Micromobility Data Policies: A Survey of City Needs
  3. Open Precinct Data
  4. Scoring Pennsylvania
  5. Coming To A Street Near You: Help Remix Create a New Tool for Street Designers
  6. planscore: a project to score gerrymandered district plans
  7. blog all dog-eared pages: human transit
  8. the levity of serverlessness
  9. three open data projects: openstreetmap, openaddresses, and who’s on first
  10. building up redistricting data for North Carolina
  11. district plans by the hundredweight
  12. baby steps towards measuring the efficiency gap
  13. things I’ve recently learned about legislative redistricting
  14. oh no
  15. landsat satellite imagery is easy to use
  16. openstreetmap: robots, crisis, and craft mappers
  17. quoted in the news
  18. dockering address data
  19. blog all dog-eared pages: the best and the brightest
  20. five-minute geocoder for openaddresses

Archives