Specifying a unified simplification for a given zoom level for all tiles
Nobody has claimed this yet.
- Dominant language
- C++
- Stars
- 3.1k
- Forks
- 430
- PR merge metrics
- No merged PRs in 30d
Description
I'm not sure this is an issue or something I don't know how to do. I've searched other issues which resemble my issue but can't seem to find the right option.
I've used the mapbox studio tileset service and when I upload a simplified geojson, I get tilesets ranging from z4 to z9 which look perfect. By using tippecanoe, I wanted to add a feature id to my geojson by using `--use-attribute-for-id=insee_code`. Unfortunately, during generation of the tilesets, I get tiles which are too detailed and thus too big.
```For layer 0, using name "2020communes"
35018 features, 10775303 bytes of geometry, 275463 bytes of separate metadata, 711485 bytes of string pool
tile 2/2/1 size is 545482 with detail 12, >500000
tile 3/4/2 size is 791842 with detail 12, >500000
tile 3/4/2 size is 517850 with detail 11, >500000
tile 4/8/5 size is 971075 with detail 12, >500000
tile 4/8/5 size is 760785 with detail 11, >500000
tile 5/16/11 size is 1013722 with detail 12, >500000
tile 5/16/11 size is 756407 with detail 11, >500000
tile 5/16/11 size is 592690 with detail 10, >500000
tile 6/32/22 size is 729739 with detail 12, >500000
tile 6/32/22 size is 514876 with detail 11, >500000
```
If I upload this to mapbox studio, I get non-uniform outlines for a given zoom level:

I've tried many many options by forcing detail level to 9, forcing minimum zoom to 4, using `-pS`, following the solution from this similar issue https://github.com/mapbox/tippecanoe/issues/452, but I can't get to have the same tileset result from a geojson uploaded to mapbox studio and a simple addition of ID to features in tippecanoe. My main issue is that by using tippecanoe the result doesn't look uniform over all the map. Is there a way to specify to tippecanoe how to have the same simplification logic for all tiles for a zoom level (so that there is no strange tile-specific "glitches") ?
The dataset itself is the outlines of all administrative areas in france with 2 properties per feature (which are small but would be nice to keep).
Thanks for the help,
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by reproducing the reported tippecanoe output with the France administrative-area dataset and the --use-attribute-for-id=insee_code option, then compare its tile simplification with the Mapbox Studio result. Done means establishing whether a unified simplification rule per zoom level is supported or defining the required behavior and documenting the relevant option or limitation.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp
- Domain
- data, tooling
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100