Not 1080p yet, but here is the video of the massive speed win just from copying an inline function into a loop:https://youtu.be/1CVmlnhgT3g?t=3113 …
-
-
Replying to @cmuratori
@cmuratori We did not have time to investigate it fully but it looks like the compiler was able to mod the loop once the inlines were gone.2 replies 0 retweets 2 likes -
Replying to @cmuratori
@cmuratori So, yeah, "zero cost abstraction" isn't even true in the case of inline functions that do nothing but cast or multiply numbers.3 replies 0 retweets 5 likes -
Replying to @cmuratori
@cmuratori You can't really conclude this from one case where the optimizer did two different things for two different pieces of code.1 reply 0 retweets 0 likes -
Replying to @nothings
@cmuratori If the difference always goes in this direction, then yes; but if half the time the inline wins and half the time manual wins, no2 replies 0 retweets 0 likes -
Replying to @cmuratori
@cmuratori VC6 doesn't have inline, only __force_inline2 replies 0 retweets 0 likes
@nothings In the case where I manually copied it in, it did a bunch of additional optimizations regarding the loop.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.