LPATHBench
LPATHBench copied to clipboard
logicchains
Results seem to vary between runs
I'm looking at D's results on x86-64 and how they varied today between the runs (1894 1799 2214 1959) without (as far as I could tell) D's code or command line changing between these specific commits.
It might be beneficial to take either the best out of 10 runs, or the average of 10 runs to smooth out the differences and get more consistent results.
If I do that, though, performance degrades each run as the cpu heats up.
On 21 December 2014 at 01:26, Erwan Ameil [email protected] wrote:
I'm looking at D's results on x86-64 and how they varied today between the runs (1894 https://github.com/logicchains/LPATHBench/commit/433890d6bea50d3da92c0dfbbb00b45cd87496a4 1799 https://github.com/logicchains/LPATHBench/commit/2cee2d5a4e27984b9ab07fe2399e6f0c9941bee5 2214 https://github.com/logicchains/LPATHBench/commit/1672f9d39bb70add8b12eb171df77160a104a87a 1959 https://github.com/logicchains/LPATHBench/commit/83a839739d0c5b117293c4cbfd603902ae1583d5) without (as far as I could tell) D's code or command line changing between these specific commits.
It might be beneficial to take either the best out of 10 runs, or the average of 10 runs to smooth out the differences and get more consistent results.
— Reply to this email directly or view it on GitHub https://github.com/logicchains/LPATHBench/issues/8.
Oh, I didn't know about that aspect. The ARM will limit itself and downclock when it heats up too much? How long does it take for it to do that? (I was pointing at x86-64 results. Is it also affected?)
Maybe put a sleep between runs on ARM ? Because performances will degrade for languages tested at the end ?
