libacovea

And with the author's hot off the presses nim flags I get only 1464 fps. So, 1920/1464 = 1.31 for my nim compile flags vs. his new ones, only a little less than the 2521/1626 that was interesting people. For something super jumpy like a simulator, I would find it unsurprising for PGO to make up (or surpass) the difference to Zig in both Nim and C++. 20 years ago there was this ACOVEA [1] project to try to discover great sets of gcc flags that could often find 2X improvements in object code speed for me. The range from build flags/procedures is often much greater than the supposedly interesting cross-language variation. These things often more measure developer experience/persistence than something intrinsic (and build flags/procedures are only part of that experience/persistence). [1] https://github.com/Acovea/libacovea

A long time ago a lone ricer experimented with this in Gentoo to find the best compiler flags: [1] https://web.archive.org/web/20110312082043/http://www.coyote... [2] https://github.com/Acovea/libacovea [3] http://swmath.org/software/28819 [4] https://ieeexplore.ieee.org/document/5380987 I think he was way ahead of the times then. Nowadays? Not so much, integrate it with all sorts of fuzzers, ccache, distcc, whatever and let the "compile cloud" churn.

"Das Balkenspiel ist schlechter Stil." "The bargraph game is lame." edit: To expand on this, there is/was one person who tried to tackle this in a meaningful way. [1] https://web.archive.org/web/20070607182125/http://www.coyote... The bitrotten remains of this can be found here: [2] https://github.com/Acovea/libacovea I'm aware of academic research regarding HPC and superoptimizing, but no one has applied that to generalized distro testing, so far. (At least not in public.) I think this should be a hook in some reproducible build system. But everyboody seems to have caved in against the complexity explosion this would involve. And build times. OpenOffice, Firefox, Chrome, KDE, Gnome....the HORROR!