We don't have the same cpu though. I have desktop AMD 7950x. You have some recent intel laptop cpu. They have similar geekbench scores, but not exactly the same. Also I disabled boosting beyond 4.5 or 4.7 GHz (don't remember exactly) in the bios, so it's slower than the default desktop configuration.

>orders of magnitude slower

It isn't. It's actually shouldn't be much slower at all. About the same for single threaded tasks. But different architectures are different architectures and could differ significantly from task to task (in tens of percent), enough to render comparative measurement as irrelevantly inaccurate.

>limit for the interpreted code.

class ArrogantSadkov {
public static void main(String[] args) {
int i = 0;

while (i<1000000000) i++;

System.out.println(i);
}
}

time java -Djava.compiler=NONE ArrogantSadkov
1000000000

real 0m3.779s
user 0m3.763s
sys 0m0.010s

$ java --version
openjdk 17.0.11 2024-04-16
OpenJDK Runtime Environment Temurin-17.0.11+9 (build 17.0.11+9)
OpenJDK 64-Bit Server VM Temurin-17.0.11+9 (build 17.0.11+9, mixed mode, sharing)

I mean maybe you meant limit for a dynamically typed language.

It's slower because it's being interpreted, undermining your "it's the limit" argument. See the "-Djava.compiler=NONE" flag. Without the "-Djava.compiler=NONE" flag:

$ time java ArrogantSadkov
1000000000

real 0m0.026s
user 0m0.014s
sys 0m0.013s

And the compiler optimized the loop out in this case.