| Time | Implementation | Compiler | Benchmark date | SUPERCOP version |
| 627696 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161217 | 20161026 |
| 630960 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161217 | 20161026 |
| 652608 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161217 | 20161026 |
| 655539 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161217 | 20161026 |
| 724475 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161217 | 20161026 |
| 742806 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161217 | 20161026 |
| 7932434 | 8bit | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161217 | 20161026 |
| 8132587 | 8bit | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161217 | 20161026 |
| 8182526 | 8bit | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161217 | 20161026 |
| 9492915 | 8bit | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161217 | 20161026 |
| 10546547 | 8bit | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161217 | 20161026 |
| 10690945 | 8bit | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161217 | 20161026 |