I'm watching this talk on performance measurements youtu.be/r-TLSBdHe1A
Ten minutes in I think this is secretly a lecture on statistics

I'm like a child watching Dora the Explorer when she asks what to do next. "Do a student's t-test!"

I'm gonna predict what's coming up next. I bet next he's gonna say something about cached data

Never mind this is about programming after all

Seeing the phrase "link order" on the screen and immediately getting up from my desk to go do smth else

Okay it's a new day and I am ready! to! learn!

You know I think I'm safe from all these what-ifs with my minimum 15x slowdown

I should run the entire file someday
But not on my computer... it will die...

I've never heard of analysis of variance
Hmm... it would be useful if I had measured more than one file

Here is the point where the talk is going to call me out for profiling incorrectly

Testing the hypothetical effect of speeding a component up by slowing everything else down instead... Very clever

Latency... throughput... I remember these terms from a class

Follow

I think it would be interesting to try out coz but it looks like the progress points are for putting in C code

Sign in to participate in the conversation
types.pl

A Mastodon instance for programming language theorists and mathematicians. Or just anyone who wants to hang out.