TurboBench - Compression Benchmark #300

New Issue

claunia · 2026-01-29T20:41:38Z

claunia commented

2026-01-29 20:41:38 +00:00

Originally created by @powturbo on GitHub (Jan 14, 2020).

In recent benchmarks comparing brotli with other compressors, we see that brotli is often tested with a small lz window size. TurboBench-Compression Benchmark is using large window brotli as default in the quality modes 10 and 11. Only the quality mode must be given as parameter to evaluate brotli.

Thanks to the submodule architecture Turbobench is frequently updated as other benchmarks.
I suggest putting a link on the brotli github repository, so to reduce the mistakes done by using sub-optimal options and to reduce the effect of other factors like I/O.
see also the response from Jyrki Alakuijala in this post

Originally created by @powturbo on GitHub (Jan 14, 2020). In recent benchmarks comparing brotli with other compressors, we see that brotli is often tested with a small lz window size. [TurboBench-Compression Benchmark](https://github.com/powturbo/TurboBench) is using large window brotli as default in the quality modes 10 and 11. Only the quality mode must be given as parameter to evaluate brotli. Thanks to the submodule architecture Turbobench is frequently updated as other benchmarks. I suggest putting a link on the brotli github repository, so to reduce the mistakes done by using sub-optimal options and to reduce the effect of other factors like I/O. see also the response from Jyrki Alakuijala in this [post](https://encode.su/threads/2119-Zstandard?p=62927&viewfull=1#post62927)

claunia referenced this issue

2026-01-29 20:51:32 +00:00

[PR #300] [CLOSED] Update decoder #894

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: starred/brotli#300