Measured - byte-exact - 2026

Pedulli vs pbzip2

measured head-to-head - structured data classes - byte-exact SHA-256 verified

TL;DR - honest
pbzip2 is multi-core bzip2 (roughly linear scaling with cores): same ratio as bzip2 -9, multi-threaded. Pedulli is a best-of-N racer: it races xz, zstd, brotli and your data's SRD math and keeps the smallest verified output, always byte-exact (SHA-256 roundtrip) with a +1 byte never-worse floor on its own output. On ratio Pedulli wins outright on structured classes and ties the best codec on already-optimal data, so it is never larger than the best standard codec it races. pbzip2's edge is a different axis entirely: multi-core wall-clock throughput on long natural-language text.

The measured table

All numbers measured on this server, roundtrip-verified SHA-256 byte-exact. Proofs available on request.

Inputpbzip2 -9Pedulli (best-of-N)Δ
1 MiB of zeros47 B13 Bsmaller (redundant input)
JSON 31 KB2,937 B1,265 B-56.9%
HTML 161 KB37,210 B37,210 Btie (best-of-N selects the equal best)
MP4 master 10 MB9,789,123 B9,430,108 B-359 KB
Random bytes 1 MB1,048,620 B (+44 B)1,048,577 B (+1 B)43 B less overhead

What pbzip2 does better (honest)

What Pedulli does that pbzip2 does not

When to switch

Use pbzip2 when you have many cores, want maximum multi-core wall-clock throughput, and bit-compatible bzip2 output is mandatory. Use Pedulli whenever ratio matters: it races xz, zstd, brotli and your data's SRD math and keeps the smallest verified output, so it is never larger than the best standard codec it races, with a per-file +1 byte never-worse floor on its own output.

More comparisons: zstd - xz - 7-Zip - brotli - gzip