You find it unlikely that a compiled language might be 15x faster than a interpreted language doing a compute-bound workload, when said compiled language is often 30x faster in posted synthetic benchmarks... I think you are extremely skeptical...

I have ported node.js applications to Go and seen similar substantial speedups. In my cases though I was running in a multi-threading enviroment. not GAE and the performance difference (improvement) was much greater under concurrent load versus single request latency. But that seems reasonable since node.js is closer to Go in performance than Python is.

I guess I just believe it because I have seen it. :)

>>posted synthetic benchmarks<<

The benchmarks game shows toy programs not synthetic benchmarks ;-)

You find it unlikely a compiled language is faster than an interpreted one...?

Sometimes, with smart data structures for specialized cases (persistency, better asymptotic behaviour etc.), it can happen that memory accesses dominate the actual execution. If that happens, and you can't saturate CPU's execution units, the difference between an efficient interpreter and a compiler gets smeared very quickly.

Python is compiled. That is the c in .pyc

Sorta, not really the point though. http://stackoverflow.com/questions/2998215/if-python-is-inte...