- Speed x2 in BSGS mode for main version
- fixed some variables names
- fixed bug in addvanity (realloc problem with dirty memory)
- Added option -6 to skip SHA256 checksum when you read the files (Improved startup process)
- Added warning when you Endomorphism and BSGS, THEY DON'T WORK together!
- Legacy version for ARM processor and other systems
- remove pub2rmd
- fixed typos in README
- Speed counter fixed for Compress search without endomorphism check https://github.com/albertobsd/keyhunt/tree/development#Speeds
- Merge of address and rmd160 speeds
- Added option for endomorphism
- Added SAVE bloom filter and table option for adddress, rmd160, minikeys and xpoint
- Improved Makefile options
- Updated random function to use the Linux RNG with the function getrandom
- Minikeys new sequential generator and x2 times more speed
- third bloom filter check for bsgs 20% less memory usage
- Minikeys improvements in speed
- Test to try solve the albertobsd#139 issue
- Solved albertobsd#130
- Minikeys new generator improvements in speed
- Ethereum support
- Double speed for rmd160 mode
- Minikeys mode support
- Stride option
- Solved some bugs: albertobsd#122 albertobsd#111
- Files are going to be updated automatillyca -- from keyhunt_bsgs_3_.blm to keyhunt_bsgs_4.blm -- from keyhunt_bsgs_1_.blm to keyhunt_bsgs_5.blm -- the program will notify you when time to delete the old files
- Fixed the slow bP table generation. -- This fix make obsolete the files keyhunt_bsgs_0_*.blm -- please delete those files, please do:
rm keyhunt_bsgs_0_*.blm
- Added multi vanitysearch for address mode
- BSGS improvements: -- 10x more Speed -- new submodes for BSGS, secuential (default), backward, both, random and dance -- automatic file generation for bloom filter file and bPtable file. -- Good bye to bPfile.
- Memory check periodically for bloom filters and bP Table
- Solved Issues 49, 50 51 See: albertobsd#51 albertobsd#50 albertobsd#49
- Solved Issues 56 albertobsd#56
- Added mutex to the bloom filter for multithread writing
- Full migration from libgmp to secp256k1
- Change the way for keygeneration for modes xpoint, address, and rmd160
- Improve performance for xpoint mode, now is ten times faster
- Change N variable type for modes address,rmd160 and xpoint, from uint32_t to uint64_t
- Added method pub2rmd to search publickeys of the puzzles and other legacy address (Compress publickeys only)
- Small changes to be compiled with mingw on Windows
- Changed sort functions and binary search for modes address/rmd160/xpoint, now those modes can load MAX 2^64 items
- xpoint input file now can contains Comments after the line of data
- from this version all furthers developments will be in the branch
development
- Added a progress counter (this solve bug albertobsd#18 )
- Added multithread for precalculating bP items or reading then from file
- Fixed the code to avoid warnings (this solve the issue albertobsd#19)
- Added xxhash for bloomfilter this hash have better performance than murmurhash2. And it is 64 bits hash :)
- We reduce the number of items of the bPtable in ram using a second bloom filter, thanks @iceland2k14
- The ram saved space is around 80%, so we can use a bigger K value, around 4 or 5 times bigger than previous version
- Solved little error with compress and uncompress new param -l. See albertobsd#17
- function bsgs optimized to use a little less RAM (not related with Pfile)
- Again removed some compile warnings. See albertobsd#16
- Added mode rmd160, this method works two times faster than Address method. This mode can search all the altcoins
- Solved some bug when the publickeys in the input file was invalid but the program keeps running with 0 publickeys
- Now publickeys can be compressed, not only uncompressed
- Added K factor for BSGS
- Added bPfile.c to generate a precalculated file
- Remove unused files about keccak and sha3
- Change Bloom filter limits and % of error from 0.001 to 0.00001 in bloomfilter.
- Added mode BSGS this work with a file with uncompressed keys
- Updated bloom filter to allow More items
- Change Quicksort to Introsort, this solve some edge cases of quicksort.
- Introsort is avaible to keyhunt and hexcharstoraw. worst case. O(N log N).
- Aling of some output text
- Added new tool hexcharstoraw to create a raw binary file for xpoint from a text-hexadecimal file
- Added option -w to work with raw binary file, this file contains xpoint in binary format fixed to 32 bytes
- Fixed some ugly bug in the searchbinary function thanks to Ujang
- Added to stdout the vanitykeys found with -v option
- Fixed search by xpoint.
- Added -e option to skip the sort process whe the file is already sorted.
- Fixed debugcount when upto N is less than debugcount.
- Changed "-R upto" to "-R" and added "-n upto" option.
- Minor bugs fixed.
- First Release
- Thanks to all CryptoHunters to make this code possible