high call rate `style` function on `RLayerVectorTile`, hits performance badly about rlayers HOT 4 CLOSED

If your function returns a different style for every feature used, you have no way to improve the performance - whether you use rlayers or you use directly OpenLayers. In this case you have 2 choices: either reduce the number of features or use clustering.
If some features share the same styles - you can use the cache parameters - in this case rlayers wil call only your hashing function to check if a new style has to be computed and constructed.

from rlayers.

You got it - the function returns 3 different styles depending on the feature type. One of those types makes for 99% of the features rendered... what cache parameters are you making mention of, Do you mean I should build a cache layer depending on the 'feature' and 'resolution' parameters sent to the function, or is there a mechanism offered by rlayers/openlayers?

Unfortunately clustering does not seem to be an option for our use case (but I'll look into it more).

Thanks for such a fast response btw 🚀

edit: I just noticed cacheSize and cacheId on the clustering example... I'll see what I can do with it.

from rlayers.

@vgrafe if one feature type represents 99% of the features, your best option, performance-wise, would be to move these features to a separate vector layer with a single fixed style

from rlayers.

Makes sense. Thank you for the help!

from rlayers.