| Time | Implementation | Compiler | Benchmark date | SUPERCOP version |
| 116241 | neon | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161222 | 20161026 |
| 116252 | neon | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161222 | 20161026 |
| 116312 | neon | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161222 | 20161026 |
| 128285 | neon | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161222 | 20161026 |
| 134470 | neon | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161222 | 20161026 |
| 136985 | neon | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161222 | 20161026 |
| 275621 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161222 | 20161026 |
| 275976 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161222 | 20161026 |
| 486414 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161222 | 20161026 |
| 498539 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161222 | 20161026 |
| 505230 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161222 | 20161026 |
| 506572 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161222 | 20161026 |