| Time | Implementation | Compiler | Benchmark date | SUPERCOP version |
| 893750 | arm | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161217 | 20161026 |
| 922293 | arm | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161217 | 20161026 |
| 922904 | arm | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161217 | 20161026 |
| 925992 | arm | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161217 | 20161026 |
| 933650 | arm | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161217 | 20161026 |
| 941074 | arm | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161217 | 20161026 |
| 1266405 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161217 | 20161026 |
| 1308632 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161217 | 20161026 |
| 1324977 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161217 | 20161026 |
| 1330783 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161217 | 20161026 |
| 1348993 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161217 | 20161026 |
| 1375520 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161217 | 20161026 |