为什么System / mscorlib代码更快？特别是对于循环？

This is just a personal project I've been digging into. Basically, I parse a text file (say from 20mb up to about 1gb) using StreamReader. The performance is pretty solid, but still... I've been itching to see what would happen if I parse it in binary. Don't misunderstand, I'm not prematurely optimizing. I am defintely micro-optimizing on purpose just "to see".This is just a personal project I've been diggi