Currently measured worst case for some types or code.
I tried it on my primes micro-benchmark (http://hoult.org/primes.txt) and got a 2:1 slowdown on 13th gen i9.
It does a LOT of array access and updating, probably near to worst-case for code that isn't just a loop copying bytes.
The average slowdown is probably more in the same region as using Java or C# or for that matter C++ std::array or std:vector.
alt Hacker News
I tried it on my primes micro-benchmark (http://hoult.org/primes.txt) and got a 2:1 slowdown on 13th gen i9.
It does a LOT of array access and updating, probably near to worst-case for code that isn't just a loop copying bytes.
The average slowdown is probably more in the same region as using Java or C# or for that matter C++ std::array or std:vector.