Encoder comparison
Recorded run: enc-2026-10-30-amd (2026-10-30). Raw results · Environment · Run analysis.
All eight datasets contribute equally to each median, including empty and tiny input. For each dataset, speed / C = C mean latency / implementation mean latency; output / C = implementation bytes / C bytes. Medians are taken over those ratios. Qualities are ordered by mbrotli median speed / C, highest first. Burli supports q0–q5 only.
| Quality | mbrotli speed / C ↑ | mbrotli output / C ↓ | Datasets |
|---|---|---|---|
| Quality 9 | 4.323× | 1.000× | 8 |
| Quality 11 | 1.493× | 1.000× | 8 |
| Quality 0 | 1.349× | 1.000× | 8 |
| Quality 10 | 1.243× | 1.000× | 8 |
| Quality 4 | 1.183× | 1.000× | 8 |
| Quality 1 | 1.123× | 1.000× | 8 |
| Quality 3 | 1.089× | 1.000× | 8 |
| Quality 8 | 1.022× | 1.000× | 8 |
| Quality 2 | 1.020× | 1.000× | 8 |
| Quality 5 | 1.017× | 1.000× | 8 |
| Quality 6 | 1.008× | 1.000× | 8 |
| Quality 7 | 1.004× | 1.000× | 8 |