Thinking about new tech in term of performance is silly. If you need performance NOW just stick with the known solution because odds are you also need stability and a well tested platform.
For example, if this version already is as quick as comparable software written in C, you know that chances are that it will be even faster in the future and not slower. If it uses an hour to open a port and send a package, that doesn't bode well for future development.
2) "I would like to see any benchmarks comparing this to..." != "I only care about performance and nothing else matters"