Bitcoin password recovery tool by Thegrideon Software

Bitcoin Password by Thegrideon Software is a feature-rich password recovery tool for bitcoin and similar wallets (based on “wallet.dat” database files): Bitcoin, Litecoin, Baha, DASH, … There are hundreds of cryptocurrency clients, forks and versions, thus the best way to check compatibility is to open your wallet (or a test wallet with known password) in trial version. You may also check Password Search FAQ .

Bitcoin and related cryptocurrency Wallets password protection is based on SHA-512 and AES-256 algorithms with thousands of iterations, thus recovery speed is limited and a good CPU / GPU hardware is required to achieve fast recovery speed, but we did our best to enhance this tool performance (recovery speed) with some advanced methods.Bitcoin Password allows you to setup a very precise search range using mixed attacks to exploit known password details (parts and patterns) and to cut the required time of a search comparing to traditional attacks. You can check our password search FAQ for details and examples.

Password Recovery Speed:

Wallets encryption complexity (number of SHA-512 iterations per password) is selected based on computer power available during the encryption process (common range is 50000 – 300000), thus recovery speed is variable from Wallet to Wallet. Spin count 100 000 is used for the tables below.

The final password recovery speed is equal to CPUspeed + GPU(1)speed + … + GPU(N)speed.
For example: Intel i3 + integrated HD 4400 GPU + AMD R9 270 = 140 + 50 + 2 400 = ~ 2 600 p/s.

CPU:
 
Instructions Set:
Speed p/s:
SHA-512 MH/s:

Intel® Core™2 Duo

T7500 @ 2.20GHz

Intel® Core™ i3

2100 @ 3.10GHz

Intel® Core™ i3

4130 @ 3.40GHz

Amazon EC2 “c5.large”

Xeon® Platinum core

SSE2
AVX
AVX2
AVX512

~30
~55
~140
~220

~3
~5.2
~13.9
~21.8

Integrated GPUs
Modern integrated GPUs (like Intel HD and Iris Graphics) can be used for password recovery acceleration. Integrated GPUs are not as powerful as dedicated (discrete) devices and often overlooked but can add up to 30% to CPU performance.

AMD and Nvidia GPUs
Compute units or stream multiprocessors (CU, SM, SMX, SMM) are the main building blocks of GPU calculation power, thus 32xCU GPU card is usually twice as fast as 16xCU GPU card of the same family and architecture running at the same frequency.
You can expect approximately 175-250 p/s performance (spin count 100k) per every Nvidia SM running at 1Ghz and 100-150 p/s per every AMD CU.

Tesla K80 (Kepler 2.0)

2x13xSMX, 820 MHz

GTX 1080 Ti (Pascal)

28xSM, 1580 Mhz

Tesla V100 (Volta)

80xSM, 1380 Mhz

RTX 2080 Ti (Turing)

68xSM, 1540 Mhz

RTX 3090 (Ampere)

82xSM, 1400 Mhz

+ 3 750 p/s
+ 13 000 p/s
+ 20 000 p/s
+ 21 800 p/s
+ 24 100 p/s

3750/26/0.82=175
13000/28/1.58=290
20000/80/1.38=180
21800/68/1.54=200
24100/82/1.4=210

HD 6770 (VLIW5)

10xCU, 850 Mhz

R9 270 (GCN 1.0)

20xCU, 950 MHz

RX 480 (GCN 4.0)

36xCU, 1260 MHz

RX Vega 64 (GCN 5.0)

64xCU, 1300 MHz

RX 6900 XT (RDNA)

80xCU, 1825 MHz

+ 900 p/s
+ 2 200 p/s
+ 5 200 p/s
+ 9 600 p/s
+ 24 500 p/s

900/10/0.85=105
2200/20/0.95=115
5200/36/1.26=115
9600/64/1.30=115
24500/80/1.83=167

It is important to mention that AMD and NVIDIA architectures are quite different and although AMD per CU performance is lower, there are more CUs per card in a similar price range. Moreover GPU performance is somewhat variable with number of factors including GPU/CPU combination and even a driver version. Please contact us if your recovery speed is noticeable different from the data above.