| Time | Implementation | Compiler | Benchmark date | SUPERCOP version |
| 638312 | table | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161218 | 20161026 |
| 644306 | table | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161218 | 20161026 |
| 644347 | table | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161218 | 20161026 |
| 659097 | table | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161218 | 20161026 |
| 706097 | table | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161218 | 20161026 |
| 756391 | table | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161218 | 20161026 |
| 2991122 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161218 | 20161026 |
| 3068360 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161218 | 20161026 |
| 5007645 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161218 | 20161026 |
| 7188607 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161218 | 20161026 |
| 8816406 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161218 | 20161026 |
| 9149426 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161218 | 20161026 |