• High performance precision, zero overhead. One might call it the compression lie detector
• The only benchmark program including LzTurbo
• 100% in-memory benchmark, no I/O overhead
• Include (>70) allmost all popular, latest or fastest compressors in one compiled package
• Benchmarking Entropy Coders, Lz77, Rolz, BWT and Context mixing compressors
• Multiple input files with recursive directories
• Concatenate multiple small files into one multiblock file
• Avoid cache szenario and cpu throttling found in other benchmarks
• Automatic sort, update and merge of result files
• Text, html, csv, markdown output
• ✨ Transfer speed sheet for different connections or devices: GPRS,2G,3G,4G,DSL,Network,HDD,SSD,RAM
• 👍 Html plot: Speedup + Speed/Ratio
• All in one executable, no hassless installing of additional packages, graphic libraries, python,...
• 👍 build in peak memory usage reporting for compression and decompression in html output

• Build or Download turbobench from releases

• List all plugins: "./turbobench -l2"
  
• List all compiled codecs: "./turbobench -l1"
  
• type "./turbobench -h" for help

• see also: Compression Benchmark

CPU: Sandy bridge i7-2600k at 4.4GHz, gcc 6.3, single thread.

• Realistic and practical benchmark with large files

• No PURE cache benchmark

• All compressors with latest versions (2017-05-10)

• File app3.tar binary Portable Apps Suite

(bold = pareto) MB=1.000.000

Hardware: ODROID C2 - ARM 64 bits - 2Ghz CPU, OS: Ubuntu 16.04, gcc 5.3
Compressors with versions 16.08.2016
pd3d.tar - 3D Test Set (RAD Game Tools)

• test all fast compressors in the lz4, lzturbo, zlib class
  (additional groups can be defined in the "turbobench.ini" file)

  ./turbobench -eFAST file
  

• Some codecs can accept additional parameters (see option -l1 or file plugins.cc) (colon ':' as separator)

  ./turbobench -elzma,9mt2:d29:a1:fb273:mf=bt4:mc999:lc8:lp0:pb2 file
  

• individual codec test (output to screen & file.tbb)
  

  ./turbobench -elzturbo,19,29,39/brotli,6/zlib,6 file
  

• retest or test other compressors and merge the results to file.tbb
  

  ./turbobench -eFAST/bzip2 file
  

• Print result file + "transfer+decompression speedup" plot to file.html for browsing

  ./turbobench -p2 -S2 file.tbb

	git clone --depth=1 --recursive https://github.com/powturbo/TurboBench.git
    cd TurboBench
	make

	make NCODEC2=1

    make EC=1

• Windows: MinGW-w64 makefile
• Linux amd/intel: GNU GCC (>=4.6)
• Linux amd/intel: Clang (>=3.2)
• Linux arm: aarch64
• MaxOS: XCode (>=9)
• PowerPC ppc64le
• IBM Z series s390x

• LzTurbo
• Blosc
• BriefLz
• Brotli
• Bzip2
• Chameleon
• CSC
• Density
• Doboz
• FastLz
• Fast-lzma
• glza
• heatshrink
• Intel(R) Intelligent Storage Acceleration Library
• bsc
• Libdeflate
• LibLZF
• LibLz
• LibSLZ
• Lz4
• Lz4ultra
• lzjody 🆕2023.06
• lizard
• Lzfse
• Lzham v1.1
• Lzlib
• Lzmat
• Lzma
• Lzo
• Lzoma
• LZSSE
• LZSA
• Miniz
• ms-compress
• Nakamichi Washigan
• Oodle (windows + linux) 🆕2023.05
• Quantile Compression
• Pithy
• Quicklz
• sap
• shoco
• Shrinker
• Smallz4
• Snappy
• Snappy-c
• Tornado
• Unishox
• wfLZ
• yalz77
• yappy v2011
• zlib
• zlib-ng
• zlib cloudflare
• xpack
• zopfli
• zstd
• zpaq

• TurboRC-Range Coder
• Bitwise RC v2010
• Bitwise vector RC v2012
• AOMedia AV1 entropy coder
• Daala entropy coder
• FastAri v15-10

• TurboAC
• subotin range coder v2000
• Fast AC v2006
• FQZ/PPMD Range Coder v15-03
• PPMD Range Coder v15-03
• sserangecoding 🆕2023.05

• Fpaqc:Asymmetric Binary Coder v07-12

• TurboANX-ANS
• TurborANS
• Finite State Coder v15-05
• Finite State Entropy v16-08

• TurboHF-Huffmann v1.3
• Tornado Huf v0.6a
• zlib Huffmann v1.2.8
• Fast HF v2006
• FPC
• FSE Huff
• Polar Codes v10-07

• inline memcpy
• library memcpy

• TurboRLE Turbo Run Length Encoding
• tcobs short messages compression with COBS framing 🆕2023.06
• TurboBase64 Turbo Base64 Encoding/Decoding
• fastbase64 Base64 Encoding
• base64 Fast Base64 stream encoder/decoder

• bwt:libdivsufsort
• st: bsc schindler transform

Last update: 07 JUL 2023