in reply to What's Perl good at or better than Python.
I have heard it said that Python is faster than Perl, but the only actual comparison I have done (parallel implementations of Conway's "Life") says that Perl is slightly faster but Python has a slightly smaller memory footprint. My opinion is that if speed is truly a consideration you will reject both languages in favor of a compiled language. My C implementation of Life had a much larger memory footprint (no readily-available hash implementation) but was orders of magnitude faster than either Perl or Python.
|
|---|
| Replies are listed 'Best First'. | |
|---|---|
|
Re^2: What's Perl good at or better than Python (Game of Life, LLiL, Rosetta and Performance References)
by eyepopslikeamosquito (Archbishop) on Oct 16, 2021 at 22:24 UTC | |
Multiple implementations of Conway's game of life in Perl and C with a focus on performance. Really? Any chance of seeing your source code? A few years back I became a bit obsessed with implementing Game of Life in both Perl and C++, as described in detail at: The benchmark timings above were for running the famous Lidka Methuselah for 30,000 generations. The code that produced the fastest benchmark times can be found here: Summary: The C++ version of the simple GoL algorithm was 450/36 = 12.5 times faster than the Perl version; memory use was 2.8 less, 517,340K v 1,455,004K. For the complex GoL algorithm, C++ was 17/0.08 = 212.5 times faster; memory use was 10.1 times less, 69,632K v 700,000K. The code that produced the fastest early benchmark times for Rosetta Code: Long List is Long can be found here: TODO (maybe): Benchmark Perl vs C++ for Rosetta PGA-TRAM and Rosetta Code: Long List is Long (and write a performance meditation with the results of GoL, PGA-TRAM, Long-List-is-Long). PGA-TRAM Benchmark done: Risque Romantic Rosetta Roman Race Update Instructively, tweaking the Perl code, via a long series of micro-optimizations, reduced the running time from 1635 secs to 450 secs (3.6 times faster), while finding a better algorithm reduced it from 450 secs to 17 secs (26.5 times faster). Similar numbers for the C++ code confirm the old Kernighan and Plauger adage: "Don't diddle code to make it faster -- find a better algorithm". Further Update
> My C implementation of Life had a much larger memory footprint (no readily-available hash implementation) This is very surprising! I suspect your implementation of hashes in C was sub-optimal. :) In my GOL code, the C++ version consistently had a much lower memory footprint than the Perl version. As detailed here the memory footprint of the fastest versions when hit with a nasty three million cell test case: 69,632K for my C++ version, 700,000K for my Perl version, and 18,138,504K (i.e. 18 GB!) for CPAN Game::Life::Infinite::Board. Performance References
Game of Life References
Long List is Long References
Rosetta Code (Many Different Languages)
Extra:
References on Comparing Programming Languages Added Later
See Also
Many references were added long after the original reply was made | [reply] |