# bzip3

12 Hacker News commenters recommended bzip3 and 12 criticized it, November 2025 to November 2026. It was recommended in 2 of 13 months.

**Verdict:** Recommended for strong compression on some data, such as Wikipedia XML dumps, and speed in some tests. Critics say decompression is slow and memory-heavy and it often trails zstd.

## Why people recommend it

- Strong compression on some data (4 comments): "I got some really good results with bzip3 compression Wikipedia XML dumps" ([source](https://news.ycombinator.com/item?id=49600649)) "Impressive compression benchmark. Four times smaller than z standard." ([source](https://news.ycombinator.com/item?id=49598648)) "The latter compressed better and was about 8x faster." ([source](https://news.ycombinator.com/item?id=47384371))
- Faster than zstd in some tests (3 comments): "Tested it myself and found that it often also does slightly worse than zstd -19 in compression ratio but faster" ([source](https://news.ycombinator.com/item?id=49598853)) "The latter compressed better and was about 8x faster." ([source](https://news.ycombinator.com/item?id=47384371)) "bzip3 was 7x faster and had ~20% better compression" ([source](https://news.ycombinator.com/item?id=47384298))

## What people warn about

- Decompression is slow and memory-heavy (5 comments): "But the decompression (memory and speed) is a different story." ([source](https://news.ycombinator.com/item?id=49602743)) "4-core zstd decompression was 2.05 seconds vs a whopping 297 seconds for bzip3 -dj4" ([source](https://news.ycombinator.com/item?id=49599953)) "512GB is a huge block size for bzip3, as it needs to basically all be in memory" ([source](https://news.ycombinator.com/item?id=49599148))
- Often worse than zstd on other data (3 comments): "the initial extremely favorable conclusion was short-lived, because trying later bzip3 on other data files gave worse results than zstd" ([source](https://news.ycombinator.com/item?id=49598743)) "Tested it myself and found that it often also does slightly worse than zstd -19 in compression ratio but faster" ([source](https://news.ycombinator.com/item?id=49598853)) "later I have also found other cases where the bzip3 performance was not so good" ([source](https://news.ycombinator.com/item?id=47386524))
- Name taken from bzip project (3 comments): "How is this not just taking their established name for your own project with the version number +1?" ([source](https://news.ycombinator.com/item?id=47393302)) "it's a bit weird to release, say, Windows 12 without permission from the authors of Windows 11" ([source](https://news.ycombinator.com/item?id=47384346)) "It's distasteful to use such a name when it's not created by the bzip authors." ([source](https://news.ycombinator.com/item?id=49600467))
- Benchmarks look cherry-picked (2 comments): "The benchmarks are disingenuous, to the point of looking cherry-picked." ([source](https://news.ycombinator.com/item?id=49598996)) "how is this an honest benchmark" ([source](https://news.ycombinator.com/item?id=49598717))

## Alternatives hackers recommend

- [Zstandard (zstd)](https://hackerapproved.com/p/zstd.md): 111 recommend
- [Brotli](https://hackerapproved.com/p/brotli.md): 24 recommend
- [gzip](https://hackerapproved.com/p/gzip.md): 23 recommend
- [tar](https://hackerapproved.com/p/tar.md): 19 recommend
- [NNCP](https://hackerapproved.com/p/nncp.md): 15 recommend

## Cite as

Hacker Approved, "bzip3", https://hackerapproved.com/p/bzip3/ (counts from 1,795,640 Hacker News comments, November 2025 to November 2026).