vagrant-cachier
Closed this issue · 4 comments
Did you run the tests with vagrant-cachier installed or no? If you did, it might need to be clarified as well.
Hmm... I'm going to check.
It looks like it wasn't installed on the two 2016 models, but it was installed on my 2015 Pro. That could skew the test a little lot, as the 2015 model wouldn't have had to re-download all the apt packages (and composer too, right?) on each run.
That would also explain the huge delta in the average test speed!
I'll have to re-run the test with cachier uninstalled on my 2015 Pro tonight.
In any case, the 2016 model comparison is still valid, I just need to see if I get largely different results from the 2015. Good catch!
@oxyc - Thanks again for pointing this out! I ran the tests again without vagrant-cachier, and uploaded the results to a separate sheet so you can compare:
https://docs.google.com/spreadsheets/d/16H6TeKCOZRwzsd5bZJM2IHVqN9fU6GZhUrDiu_SK2zU/edit?usp=sharing
It performed one fewer run, and battery runtime was 5.6 hours instead of 4.9. Interesting result, but I don't know if I can keep sacrificing my entire night and part of a morning (draining the battery resets the PMU, which resets the clock, which causes all kinds of crazy with Apps that depend on the correct time) for these tests on one of my main laptops.