Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Those benchmarks are pretty heavily gamed. Think about things like hardcoding "Content-Length: 123" instead of computing the length at runtime.


I wish benchmark suites would reject those kinds of shenanigans, it's completely ridiculous and makes the results practically worthless.


Do you raise those issues as specific bug reports against specific programs?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: