| Time | Implementation | Compiler | Benchmark date | SUPERCOP version |
| 78882 | neon | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161222 | 20161026 |
| 78908 | neon | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161222 | 20161026 |
| 78952 | neon | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161222 | 20161026 |
| 87897 | neon | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161222 | 20161026 |
| 91865 | neon | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161222 | 20161026 |
| 93491 | neon | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161222 | 20161026 |
| 191721 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161222 | 20161026 |
| 191903 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161222 | 20161026 |
| 338120 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161222 | 20161026 |
| 350548 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161222 | 20161026 |
| 356191 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161222 | 20161026 |
| 357601 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161222 | 20161026 |