Introducing VanitySearch: A New Tool for Bitcoin Address Prefixes

22 replies 253 views
kevin_novaHero Member
Posts: 13 · Reputation: 3064
#1Oct 13, 2025, 09:12 PM
Hey everyone, I wanna share a new tool called VanitySearch. It's kinda like Vanitygen, but way faster on my end. This one doesn’t use OpenSSL for heavy lifting and runs on Cuda, fully taking advantage of PTX assembly. On my Intel i7, it’s about 4x quicker than Vanitygen. Crazy speeds!
5 Reply Quote Share
Posts: 80 · Reputation: 176
#2Oct 14, 2025, 02:31 AM
Sounds cool. Are there plans for a Linux version? I’d like to give it a shot since my laptop is old. Might be interesting to see how it stacks up against Vanitygen.
4 Reply Quote Share
kevin_novaHero Member
Posts: 13 · Reputation: 3064
#3Oct 14, 2025, 02:37 AM
Totally! The CPU part should work fine on Linux, just need to check for any missing stuff with gcc. For CUDA, I've got to install the SDK on Linux and try compiling it. Shouldn't be too tough.
3 Reply Quote Share
kevin_novaHero Member
Posts: 13 · Reputation: 3064
#4Oct 14, 2025, 06:58 AM
Awesome news! Just dropped a new release, it includes a makefile for Linux, but it's only for CPU right now. Still working on the CUDA version for Linux. Would love to hear how it performs on your setups.
3 Reply Quote Share
kevin_novaHero Member
Posts: 13 · Reputation: 3064
#5Oct 14, 2025, 11:46 AM
Thanks for the update! So, those two 'Check' fields are just for debugging, right? And if you use the -gpu flag, it slows everything down if all CPU cores are engaged?
3 Reply Quote Share
Posts: 61 · Reputation: 12
#6Oct 14, 2025, 03:53 PM
Exactly! Those fields help in checking addresses. Yeah, if you add the -gpu option, it hogs resources since CPU can't manage GPU/CPU transfers well. I’ll make sure to leave one core free when GPU is selected.
6 Reply Quote Share
kevin_novaHero Member
Posts: 13 · Reputation: 3064
#7Oct 14, 2025, 07:52 PM
Man, it's great to see open-source projects moving forward. I'll check out VanitySearch when I can.
3 Reply Quote Share
kevin_novaHero Member
Posts: 13 · Reputation: 3064
#8Oct 14, 2025, 10:53 PM
Appreciate your interest! Just released v1.2 which fixes a few issues and boosts performance by around 10%. Updated the probability calculation for large prefixes too.
2 Reply Quote Share
fork21Member
Posts: 1 · Reputation: 44
#9Oct 15, 2025, 12:00 AM
Hey all, Just put out v1.3 with a ~15% overall performance boost and ~20% for GPU. On a side note, it’s now twice as fast on my machine compared to oclvanitygen. Just need to hit that 8-character prefix goal!
2 Reply Quote Share
kevin_novaHero Member
Posts: 13 · Reputation: 3064
#10Oct 15, 2025, 03:00 AM
Sounds impressive! How long do you think it’ll take on your hardware? I guess those old setups can be slow.
4 Reply Quote Share
0xKingFull Member
Posts: 9 · Reputation: 711
#11Oct 15, 2025, 08:41 AM
I’m aiming for 2-3 days, but with my current rig, I might need 2 weeks for a 50% chance. Next focus is multi-GPU support and CUDA for Linux.
0 Reply Quote Share
kevin_novaHero Member
Posts: 13 · Reputation: 3064
#12Oct 16, 2025, 01:34 AM
Hey, can I shoot you a message? Gotta know how to enable that option in your profile.
2 Reply Quote Share
kevin_novaHero Member
Posts: 13 · Reputation: 3064
#13Oct 16, 2025, 04:19 PM
Sure, no issues! Just activated it.
3 Reply Quote Share
0xKingFull Member
Posts: 9 · Reputation: 711
#14Oct 17, 2025, 07:41 AM
What’s your approach? Using affine or jacobian coordinates for the points?
1 Reply Quote Share
kevin_novaHero Member
Posts: 13 · Reputation: 3064
#15Oct 17, 2025, 12:09 PM
I’m using affine coordinates for speed. We perform p = startP + i*G, and the inversion only happens once per group. Makes it quicker. Plus I’ve got a fast modular inversion method.
2 Reply Quote Share
0xKingFull Member
Posts: 9 · Reputation: 711
#16Oct 17, 2025, 10:49 PM
Cool! Just curious, any updates on Linux? I’m interested to see performance improvements.
4 Reply Quote Share
kevin_novaHero Member
Posts: 13 · Reputation: 3064
#17Oct 18, 2025, 04:14 AM
Yeah, just pushed v1.4 with some bug fixes for Linux. I’ve also snagged an old PC with dual Quadro 600s to work on multi-GPU support.
3 Reply Quote Share
0xKingFull Member
Posts: 9 · Reputation: 711
#18Oct 18, 2025, 07:52 AM
Quick question, why stick with 256 for group size? Thought a higher number might be better.
0 Reply Quote Share
kevin_novaHero Member
Posts: 13 · Reputation: 3064
#19Oct 18, 2025, 11:48 AM
A bigger group size doesn’t yield significant gains, maybe less than 1%. My modular inversion method is quick, and with 256, it keeps things manageable.
4 Reply Quote Share
kevin_novaHero Member
Posts: 13 · Reputation: 3064
#20Oct 18, 2025, 04:07 PM
On my setup, I’m hitting decent numbers, around 1.2 MKeys/s with VanitySearch. But my own library runs at 2.0 MKeys/s. Gotta work on optimizations.
4 Reply Quote Share

Related topics