Results seem to vary between runs

[idlewan]

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.

[logicchains]

If I do that, though, performance degrades each run as the cpu heats up.

On 21 December 2014 at 01:26, Erwan Ameil notifications@github.com wrote:

I'm looking at D's results on x86-64 and how they varied today
between the runs (1894
433890d
1799
2cee2d5
2214
1672f9d
1959
83a8397)
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
#8.

[idlewan]

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?)

[hokkos]

Maybe put a sleep between runs on ARM ? Because performances will degrade for languages tested at the end ?