BitCrack
BitCrack copied to clipboard
3070 , 3080 , 3090 Speed.
Has anyone been able to test the speed on those cards ? From previous issue someone posted 1b/s for 3080 on the cl version (yes i know that cuda version doesn't work on 3080). More interested if anyone has tested for 3070 and 3090 (or if anyone has those and can run basic test for speed with screenshots ofcourse :) ).
[2020-12-16.14:51:35] [Info] Compression: compressed [2020-12-16.14:51:35] [Info] Starting at: 00000000000000000000000000000000000000000000000000000000057EE7CA [2020-12-16.14:51:35] [Info] Ending at: 000000000000000000000000000000000000000000000000FFFFFFFFFFFFFFFF [2020-12-16.14:51:35] [Info] Counting by: 0000000000000000000000000000000000000000000000000000000010000000 [2020-12-16.14:51:35] [Info] Compiling OpenCL kernels... [2020-12-16.14:51:35] [Info] Initializing GeForce RTX 3060 Ti [2020-12-16.14:51:37] [Info] Generating 33,554,432 starting points (1280.0MB) [2020-12-16.14:51:41] [Info] 10.0% [2020-12-16.14:51:41] [Info] 20.0% [2020-12-16.14:51:42] [Info] 30.0% [2020-12-16.14:51:42] [Info] 40.0% [2020-12-16.14:51:42] [Info] 50.0% [2020-12-16.14:51:42] [Info] 60.0% [2020-12-16.14:51:42] [Info] 70.0% [2020-12-16.14:51:42] [Info] 80.0% [2020-12-16.14:51:43] [Info] 90.0% [2020-12-16.14:51:43] [Info] 100.0% [2020-12-16.14:51:43] [Info] Done [2020-12-16.14:51:43] [Info] Loading addresses from 'C:/Users/Desktop/exe.txt' [2020-12-16.14:51:43] [Info] 159 addresses loaded (0.0MB) [00:00:14] 2048/8192MB | 159 targets 613.69 MKey/s (18,446,744,063,416,801,333 remaining) [ETA 9.53 centuries]
-b 128 -t 256 -p 1024
Nvidia RTX 3090 Bitcrack CL version 1779 Mkey/Sec
Nvidia RTX 3090 Vanitysearch cuda version speed 4025 Mkey/sec
I modified the code to compile with Nvidia cuda 11.1 and able to load the Bitcrack Cuda version but its giving 2 types of error-
1.if its single address input the errors is "misaligned address" 2.if its input file the error after allocating bloom filter is "invalid value"
try bitcrack -b 82 -t 256 -p 2096 -b 82 -t 256 -p 3096 -b 82 -t 256 -p 4096 -b 82 -t 512 -p 2096 -b 82 -t 512 -p 3096 -b 82 -t 512 -p 4096
also try with cuBitCrack, compile cuda ver and try to use that
@hamnaz may I know where is updated compiled version of cubitcrack ??
try bitcrack -b 82 -t 256 -p 2096 -b 82 -t 256 -p 3096 -b 82 -t 256 -p 4096 -b 82 -t 512 -p 2096 -b 82 -t 512 -p 3096 -b 82 -t 512 -p 4096
And -t 0 is must
-t 0 is for kangaroo not for bitcrack
-t 0 is for kangaroo not for bitcrack
I am talking about his second screenshot where he use -t 7
Building in Linux Using make:
Build CUDA:
make BUILD_CUDA=1 Build OpenCL:
make BUILD_OPENCL=1 Or build both:
make BUILD_CUDA=1 BUILD_OPENCL=1
Nice stats on all cards compared to the 2080 ones . Too bad no one showed 3070 , but i`m guessing will be the same as 3060 TI .
OpenCL (!) version
-b 128 -t 256 -p 1024
RTX 3080 Asus TUF, 427k targets
P.s.: for 1 target - 1462 Mkey/s.
root@hash:/home/hash/BitCrack/bin# ./clBitCrack -b 82 -t 256 -p 4096 1FshYsUh3mqgsG29XpZ23eLjWV8Ur3VwH [2021-01-13.18:14:41] [Info] Compression: compressed [2021-01-13.18:14:41] [Info] Starting at: 0000000000000000000000000000000000000000000000000000000000000001 [2021-01-13.18:14:41] [Info] Ending at: FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFEBAAEDCE6AF48A03BBFD25E8CD0364140 [2021-01-13.18:14:41] [Info] Counting by: 0000000000000000000000000000000000000000000000000000000000000001 [2021-01-13.18:14:41] [Info] Compiling OpenCL kernels... [2021-01-13.18:14:41] [Info] Initializing GeForce RTX 3090 [2021-01-13.18:16:10] [Info] Generating 85,983,232 starting points (3280.0MB) [2021-01-13.18:16:29] [Info] 10.0% [2021-01-13.18:16:29] [Info] 20.0% [2021-01-13.18:16:30] [Info] 30.0% [2021-01-13.18:16:30] [Info] 40.0% [2021-01-13.18:16:30] [Info] 50.0% [2021-01-13.18:16:30] [Info] 60.0% [2021-01-13.18:16:30] [Info] 70.0% [2021-01-13.18:16:30] [Info] 80.0% [2021-01-13.18:16:31] [Info] 90.0% [2021-01-13.18:16:31] [Info] 100.0% [2021-01-13.18:16:31] [Info] Done GeForce RTX 3090 7872 / 24268MB | 1 target 1667.26 MKey/s (533,010,055,168 total) [00:05:16]
A 3080 RTX Prism does 1600m/k/s on openCL easy, so I would suggest you to tweak the -b value a bit (e.g. 128), as the 3090 should be doing way better.
Fyi. The CL version is broken (https://github.com/brichard19/BitCrack/issues/263). Your power will be used for void.
Goodluck hunting 👍
GeForce RTX 3060 Laptop GPU
cuBitCrack.exe
-b 64 -t 512 -p 1024 564.09 MKey/s [00:00:32]
-b 96 -t 128 -p 1024 605.41 MKey/s [00:00:30]
-b 128 -t 256 -p 1024 666.64 MKey/s [00:00:27]
-b 82 -t 256 -p 2096 704.18 MKey/s [00:00:25]
clBitCrack.exe
-b 128 -t 256 -p 1024 613.69 MKey/s [00:00:29]
-b 128 -t 256 -p 756 609.83 MKey/s [00:00:29]
-b 128 -t 256 -p 1024 618.68 MKey/s [00:00:29]
-b 82 -t 256 -p 2096 620.38 MKey/s [00:00:29]
Anyone want to help me out here? I have an RTX 3090 and with settings you guys are running Im getting less than 200mkey/s
3070
For -b 82 -t 256 -p 1024 3070 Ti NVIDIA GeForce R 3493 / 8191MB | 16294 targets 1095.95 MKey/s (4,351,568,379,904 total) [01:07:27]
3080 -b 128 -t 256 -p 1024