| Time | Implementation | Compiler | Benchmark date | SUPERCOP version |
| 815681 | arm | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161211 | 20161026 |
| 815963 | arm | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161211 | 20161026 |
| 1004271 | arm | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161211 | 20161026 |
| 1030372 | arm | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161211 | 20161026 |
| 1033824 | arm | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161211 | 20161026 |
| 1079334 | arm | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161211 | 20161026 |
| 1626611 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161211 | 20161026 |
| 1668030 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161211 | 20161026 |
| 1812170 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161211 | 20161026 |
| 1843538 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161211 | 20161026 |
| 1871086 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161211 | 20161026 |
| 1877551 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161211 | 20161026 |