| Time | Implementation | Compiler | Benchmark date | SUPERCOP version |
| 103256 | faster | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161218 | 20161026 |
| 103442 | faster | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161218 | 20161026 |
| 105497 | faster | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161218 | 20161026 |
| 106065 | faster | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161218 | 20161026 |
| 112878 | faster | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161218 | 20161026 |
| 113907 | faster | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161218 | 20161026 |
| 181450 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161218 | 20161026 |
| 187560 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161218 | 20161026 |
| 193206 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161218 | 20161026 |
| 218511 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161218 | 20161026 |
| 240978 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161218 | 20161026 |
| 243825 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161218 | 20161026 |