Uh, plenty of true things, but sometimes just "I have a special case that's a problem": "Certain applications that exist both for Windows and Linux start up fa

Uh, plenty of true things, but sometimes just "I have a special case that's a problem": "Certain applications that exist both for Windows and Linux start up faster in Windows than in Linux, sometimes several times faster. It's worth noting though that SSD disk users are mostly unaffected." Counter anecdata: I worked on the framework to run >5000 tests on Linux and Windows for a "certain program at my old job" in the quasi-CI. For 90% of the tests, a single processes would be done in a few hundred milliseconds, but Windows took much longer than Linux due to program start up time. We initially did a strict splitting of "call $program for step a", then "call $program for step b", but due to the slow program start I did add some logic to merge the two steps where possible (90% of the cases). I can't recall exact numbers, but I think the tests were done in about 70 to 50% of the time. We saved even more time when we moved other steps c & d to Linux, which probably saved a lot of money because we devs got the test results much faster than before. But: I can imagine that different programs yield different results.