Has anyone seen the initially estimated performanc...
# k2-adopters
e
Has anyone seen the initially estimated performance improvements that K2 was supposed to bring? From my personal projects, speaking to colleagues on different projects, and public posts, the consensus seems to be that there were minimal, if any, improvements.
c
In my experience, K2 is indeed much faster than K1. However, builds are usually dominated by Gradle configuration and execution, so the time taken by the Kotlin compiler is just a small percent of the overall build time—even if the Kotlin compiler gets faster, it doesn't change much in the experience.
u
Yea I'm the same. I reframed it to them needing to refactor as to provide new language features. Gogo unions and value semantics!
r
Same, initial benchmark showed near to 25% of build improvement. Gonna run, more benchmark on different tasks, but that should be similar to initial result.