| Time | Implementation | Compiler | Benchmark date | SUPERCOP version |
| 2989285 | table | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161210 | 20161026 |
| 3364412 | table | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161210 | 20161026 |
| 3473069 | table | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161210 | 20161026 |
| 3572140 | table | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161210 | 20161026 |
| 4238246 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161210 | 20161026 |
| 4408809 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O3 | 20161210 | 20161026 |
| 5216631 | table | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161210 | 20161026 |
| 5441470 | table | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161210 | 20161026 |
| 8113039 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161210 | 20161026 |
| 10618829 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -O2 | 20161210 | 20161026 |
| 15944525 | ref | gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os | 20161210 | 20161026 |
| 16733533 | ref | gcc -mcpu=native -mfpu=neon-vfpv4 -Os | 20161210 | 20161026 |