TIL: MacBooks are slower when not plugged in!
# general
h
TIL: MacBooks are slower when not plugged in!
2
😮 1
😂 1
Am benchmarking something either storage- or CPU-bound (still not sure), and seeing the benchmarks take about 50% longer when I unplug
You can check this for yourself with some simple CPU-bound thing, like
Copy code
$ time python -c 'x = 0
for i in range(0, 100000000): x += 1'
b
They must use some tricks to get a 18 hours long battery life 😅
g
This one is also important to keep in mind when measuring storage on Macbooks: https://twitter.com/marcan42/status/1494213862970707969 ... without flushing to stable data loss is possible.
🤯 1
h
Nothing has meaning any more
Weird that POSIX is not more emphatic about this
reminds me of mongodb achieving amazing transaction throughput by buffering writes and not fsyncing them
🤣 1
and hiding all this behind the “write concern” concept