Re: "Python is 71x Slower, Uses 75x More Energy, Than C" - YouTube

On 31/07/2024 01:06, Lynn McGuire wrote:

"Python is 71x Slower, Uses 75x More Energy, Than C" - YouTube
    https://www.youtube.com/watch?v=U4c6nFGt1iM

Could that guy's voice be any more annoying?
Anyway, some observations:
* CPython (I assume this is Cpython), is of course just a program written in C. So it's really just a C application ...
* The C results are going to be mostly thanks to the efforts of aggressively optimising compilers, rather than the language itself. I guess they weren't using Tiny C!
* The test programs are all from the computer language benchmarks game, which are mostly useless for comparisons. Because entries can choose their own algorithms, employ multi-threading etc. So it's no surprise there is such a wide range of results.
* When Python is used for a one-off program that runs in a fraction of a second, the fact that is is 70 times slower than the C equivalent is irrelevant. But if you include the build-time for the C, then the Python may well have used fewer machine resources.

 I am not buying that C++ is slower than Rust.
 The referenced paper is:
    https://www.sciencedirect.com/science/article/abs/pii/S0167642321000022
 

It's the computer benchmarks game; I wouldn't pay too much attention.
When I once tested of its programs ('fannkuch') I got these results, although they are several years old (3 out of 24 shown):
  Clang -O3     2.45 secs
  Rustc -O      3.2
  Tiny C       10.1
  Rustc        37.8
Rust in debug mode ran a magnitude slower than optimised releae mode. Maybe it was interpreted? I never found out why.