| Time | Implementation | Compiler | Benchmark date | SUPERCOP version |
| 142082 | neon | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161211 | 20161026 |
| 143424 | neon | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161211 | 20161026 |
| 154716 | neon | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161211 | 20161026 |
| 156461 | neon | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161211 | 20161026 |
| 163673 | neon | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161211 | 20161026 |
| 181229 | neon | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161211 | 20161026 |
| 497174 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161211 | 20161026 |
| 526585 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161211 | 20161026 |
| 613593 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161211 | 20161026 |
| 622637 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161211 | 20161026 |
| 666138 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161211 | 20161026 |
| 689531 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161211 | 20161026 |