-
Notifications
You must be signed in to change notification settings - Fork 120
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Further places layer plastic surgery #1729
Comments
Probably this line can change: https://github.com/tilezen/vector-datasource/blob/master/vectordatasource/transform.py#L8132
to:
Which would better match how Natural Earth is curated for min_zoom (which will remove the 1.7 features above from the zoom 1 tile, but they'll show up in zoom 2 tile and collide out min_zoom 2.0 features – good). There are some x.5 curated features in Natural Earth. This would also force the Poland example into zoom 3 tile. |
Our 512px tiles are designed to contain the same content as the 4 Also, because clients typically overzoom a zoom I'm happy to make a change as suggested (basically dropping a whole bunch of features needed for overzooming), but we should be aware that it'll be a major change compared with how we've done things in the past. Although it might not matter, if clients aren't evaluating fractional zooms anyway. |
Turns out Tangram doesn't evaluate fractional zooms. I still think it's useful to overstuff the layer a little bit, but for the places layer I think the file size savings are more important / values with x.6 and x.7 etc min_zoom would probably need much more sophisticated runtime label logic than is generally available, too? The reason zoom 0 is still big is |
Verified, closing. |
Originally posted by @nvkelso in https://github.com/tilezen/vector-datasource/issue_comments#issuecomment-447149749
For zoom 0 we're still including too many countries. The same is true for zooms 1 and 2. Addressing this will have significant file size savings.
Seems like we're doing 512 min_zoom math instead of 256 math? Like at zoom 0 we should only see features with min_zoom < 1, but many 1.x graded features are included.
New:
![image](https://user-images.githubusercontent.com/853051/49972295-10a54580-fee6-11e8-98d7-8dc8e84df513.png)
Old:
![image](https://user-images.githubusercontent.com/853051/49972286-07b47400-fee6-11e8-8baa-fa0a3b3b50f6.png)
`
Poland is included at zoom 2, but it's min_zoom is 3.
The text was updated successfully, but these errors were encountered: