3 ms·
Compiling a single C file program, linking glib, libarchive and some other stuff (repeated runs, to allow for caching inputs): Orange Pi 3: real 0m0.916s
by megous 4y ago
Compiling a single C file program, linking glib, libarchive and some other stuff (repeated runs, to allow for caching inputs):
Orange Pi 3:
real 0m0.916s
user 0m0.839s
sys 0m0.074s
5950X
real 0m0.080s
user 0m0.064s
sys 0m0.016s
Slower, but not nearly 1000x in practice. Cortex-A8 is a bit slower, still, but not by much.
- cozzyd 4y agowell, one project I work on compiles ~100 times faster on my desktop (Ryzen 5 3600) than the BBB if building from scratch, and that's not accounting for multiple cores, so 1000x might be reasonable for a project with sufficient compile-time parallelization. But it's still fast enough (< 10 seconds ) on the BBB that setting up a cross-compilation environment isn't worth it (and obviously incremental compiles are faster).