| Time | Implementation | Compiler | Benchmark date | SUPERCOP version |
| 399475 | arm | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161217 | 20161026 |
| 399914 | arm | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161217 | 20161026 |
| 475151 | arm | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161217 | 20161026 |
| 528847 | arm | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161217 | 20161026 |
| 534790 | arm | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161217 | 20161026 |
| 534795 | arm | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161217 | 20161026 |
| 675091 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161217 | 20161026 |
| 675240 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161217 | 20161026 |
| 728525 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161217 | 20161026 |
| 799238 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161217 | 20161026 |
| 812717 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161217 | 20161026 |
| 827466 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161217 | 20161026 |