| Time | Implementation | Compiler | Benchmark date | SUPERCOP version |
| 526386 | neon | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161219 | 20161026 |
| 532394 | neon | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161219 | 20161026 |
| 540053 | neon | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161219 | 20161026 |
| 580831 | neon | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161219 | 20161026 |
| 583686 | neon | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161219 | 20161026 |
| 591346 | neon | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161219 | 20161026 |
| 1249805 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161219 | 20161026 |
| 1263386 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161219 | 20161026 |
| 2145313 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161219 | 20161026 |
| 2741581 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161219 | 20161026 |
| 2896971 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161219 | 20161026 |
| 3088497 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161219 | 20161026 |