| Time | Implementation | Compiler | Benchmark date | SUPERCOP version |
| 870 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161214 | 20161026 |
| 889 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161214 | 20161026 |
| 893 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161214 | 20161026 |
| 894 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161214 | 20161026 |
| 958 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161214 | 20161026 |
| 994 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161214 | 20161026 |
| 1276 | armneon2 | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161214 | 20161026 |
| 1277 | armneon2 | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161214 | 20161026 |
| 1278 | armneon2 | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161214 | 20161026 |
| 1281 | armneon2 | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161214 | 20161026 |
| 1283 | armneon2 | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161214 | 20161026 |
| 1319 | armneon | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161214 | 20161026 |
| 1322 | armneon | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161214 | 20161026 |
| 1334 | armneon | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161214 | 20161026 |
| 1337 | armneon | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161214 | 20161026 |
| 1354 | armneon2 | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161214 | 20161026 |
| 1375 | armneon | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161214 | 20161026 |
| 1375 | armneon | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161214 | 20161026 |