| Time | Implementation | Compiler | Benchmark date | SUPERCOP version |
| 90858 | e/hongjun/v0 | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161216 | 20161026 |
| 91341 | e/hongjun/v0 | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161216 | 20161026 |
| 92497 | e/hongjun/v1 | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161216 | 20161026 |
| 92985 | e/hongjun/v1 | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161216 | 20161026 |
| 94083 | e/hongjun/v1 | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161216 | 20161026 |
| 94647 | e/hongjun/v1 | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161216 | 20161026 |
| 96541 | e/hongjun/v0 | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161216 | 20161026 |
| 96747 | e/hongjun/v1 | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161216 | 20161026 |
| 98612 | e/hongjun/v0 | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161216 | 20161026 |
| 98676 | e/hongjun/v0 | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161216 | 20161026 |
| 111741 | e/hongjun/v1 | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161216 | 20161026 |
| 116213 | e/hongjun/v0 | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161216 | 20161026 |