| Time | Implementation | Compiler | Benchmark date | SUPERCOP version |
| 47511 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161216 | 20161026 |
| 48150 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161216 | 20161026 |
| 57815 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161216 | 20161026 |
| 57839 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161216 | 20161026 |
| 58644 | ref64 | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161216 | 20161026 |
| 58662 | ref64 | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161216 | 20161026 |
| 58706 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161216 | 20161026 |
| 58978 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161216 | 20161026 |
| 70913 | ref64 | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161216 | 20161026 |
| 71063 | ref64 | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161216 | 20161026 |
| 71758 | ref64 | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161216 | 20161026 |
| 73490 | ref64 | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161216 | 20161026 |