| Time | Implementation | Compiler | Benchmark date | SUPERCOP version |
| 1743401 | arm | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161211 | 20161026 |
| 1746586 | arm | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161211 | 20161026 |
| 1936041 | arm | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161211 | 20161026 |
| 1957620 | arm | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161211 | 20161026 |
| 1965733 | arm | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161211 | 20161026 |
| 2019666 | arm | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161211 | 20161026 |
| 2562443 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161211 | 20161026 |
| 2705888 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161211 | 20161026 |
| 2717149 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161211 | 20161026 |
| 2718144 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161211 | 20161026 |
| 2774033 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161211 | 20161026 |
| 2836556 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161211 | 20161026 |