| Time | Implementation | Compiler | Benchmark date | SUPERCOP version |
| 90928 | neon | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161211 | 20161026 |
| 90969 | neon | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161211 | 20161026 |
| 91075 | neon | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161211 | 20161026 |
| 111007 | neon | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161211 | 20161026 |
| 112603 | neon | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161211 | 20161026 |
| 114678 | neon | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161211 | 20161026 |
| 236134 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161211 | 20161026 |
| 239749 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161211 | 20161026 |
| 406194 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161211 | 20161026 |
| 411940 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161211 | 20161026 |
| 432662 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161211 | 20161026 |
| 473099 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161211 | 20161026 |