collectd configuration

Hi team, really liking tdengine so far, and wondering why it doesn't come up more with the crowd on hackernews etc...

About to implement simple server monitoring via collectd and if I understand it correctly, the documentation at https://docs.tdengine.com/third-party-tools/data-collection/collectd/ has the binary vs tsdb protocols backwards?

It says to use write_tsdb for binary and network with the [opentsdb_telnet] tdengine configuration, but as far as I can tell (and the only way to get it to work) is use network over UDP for the [collectd] tdengine configuration, and write_tsdb for TCP to [opentsdb_telnet] configurations. Is this generally correct?

And second when i don't define a schema, it creates a supertable for each Plugin, and tags for value types and type_instance instead of putting all readings into one supertable with multiple values like cpu, disc, memory etc. Is this the best way to handle collectd data or could this be done more efficiently by defining a different schema or changing the ingestion on the collectd side?

Thanks a ton, hope the answers are also useful for others.