Topic: VanitySearch (Yet another address prefix finder) - page 7. (Read 31225 times)

That means the CUDA code for generating address checksums is wrong, which is not surprising to me since VanityGen was retrofitted using Kangaroo source code (which does not have this subsystem), so probably wasn't tested thoroughly enough.

Just tried this. No luck! I also tried using all the available codes for my GPU which I got with
VanitySearch -check always returning errors.

Which Nvidia drivers and CUDA version do you use, can you post some more details ?[/code]

Hmm, that's weird since I'm not experiencing that using Vanitysearch v1.19 (release).

Here's the result of ./vanitysearch -check command:

A quick test using your command: ./vanitySearch -stop 1cit:
WIF private key and address are valid and correct.

Have you compiled directly from the master branch?
If so, try to build from the 'releases tags' instead, specifically v1.19 (link).
The issue may have been introduced by the last few commits after Aug 10, 2020.

==> Only Priv (HEX) is correct, the rest is wrong.

==> Only Priv (HEX) is correct, the rest is wrong.

According to the output on the second search while omitting the GPU flag it shows only "Number of CPU thread used" = 1 and no hint about GPU. But the error is still there, the calculation is wrong.

Just like the result in your vanitySearch -check command, the checksum here is also invalid.

The first four Bytes of SHA256[SHA256(80F30DBBB1FFF76EBF9741202CCE4676B6258ACCAF648C040844D9796F239AD99801)] should be: 0xA3B0A194
while the invalid WIF private key has: 0x6B4E251C, so it didn't pass the checksum.

Same with the address, 1citboYAWtgbJCySZe8C7U7y36cxGfZUZ has an invalid checksum of 0xB071873E
The first four Bytes of SHA256[SHA256(0006C186C1421FF64024B3FC356AF312061DE52688)] should be: 0x054A4884

---

Someone has reported a similar issue (also using a GPU): https://github.com/JeanLucPons/VanitySearch/issues/124
BTW, I'm not experiencing this when only using CPU.

Hey there. First of all BIG THANKS to Jean_Luc for that tool which I downloaded and compiled on Linux. I am running Ubuntu 22.04 (Jammy Jellyfish) and have the GPU NVIDIA GeForce RTX 3070 in use with all drivers up-to-date. I can access it without issues with various applications. However I cannot make use of it with the VanitySearch tool.


I tried to build with
but I have no folder /usr/local/cuda* on my disk so the compilation fails. I could only compile it without GPU support. Do I need to install additional CUDA stuff on my machine, do I need CUDA SDK ? I've done some research and the answer obviously is YES, I do.

I installed Nvidia Toolkit as shown HERE and it went well. Now there are folders /usr/local/cuda* on my disk. I retried running the build command of VanitySearch with GPU support, but it fails:

I do not have a /usr/local/cuda-8.0 folder on my system, but there is a /usr/local/cuda-11.7 folder which contains the nvcc file. So i had to modify the line 34 inside Makefile of VanitySearch and put the corresponding version which is installed on my system:

then I retried the build process which again failed:

Seems it doesn't understand the required compute capability 2.0. Uhmm?? Let's dig further ...


I understand that this has something to do with compute capability but honestly said I have no clue what exactly these numbers represent. (EDIT: SOLVED! see bottom of this post!) I guess and hope the higher the value the better So I tried editing VanitySearch' Makefile once more and manually assign the value 87 to the particular variable (note: this is wrong, I had to use 86, check the solution on the bottom of my post)
[/quote]
It failed again with error stating not founding g++-4.8 on my system. Indeed, I don't have this executable but instead I have /usr/bin/g++-11 and g++, latter is just a symlink to it. Now modifying again the Makefile ...

and the build finished without errors. BTW: At this step I realized that I can ommit the CCAP=2.0 in the command line *lol*   because I already hard-fixed it in Makefile. So far SO GOOD but ...

I cannot run VanitySearch with GPU support, see here:

when I omit the -gpuId 0,1 parameter as shown in developers example I get this error

Simply said, I dont have two GPUs, only one. Being said I need to either use -gpuId 0 or ommit the switch so default setting (0) take effect.

EDIT: ==> SOLVED !!! YEAH !!! I just had to continue reading on Jean_Luc's GIT repository at the section "Docker". There is a link to Wikipedia explaining CUDA computing capabilities and the various GPUs, alternatively check also this link. In my case (Nvidia RTX 3070Ti)  it is computing capability 8.6 but I used 8.7 in my initial try. After replacing with 8.6 and rebuilding the tool it runs perfectly (... I thought  )


this took <5 seconds   I'm curious what rates other can achieve with their GPU. Anyone out there willing to share his GPU model and the achieved calc rate with VanitySearch, please? I get 1,680 MKey/s with my Nvidia 3070 Ti.

Now the bad news are --> All addresses with prefix 1 or 3 that were found are unknown/invalid on the blockexplorer. Only adresses with prefix bc1q that I found are valid but their private keys are invalid, too. When I run VanitySearch with the -check switch it outputs warnings, see here:

That confirms that only the calculated bc1q addresses are valid. Going one step further I tried to import one of the found bc1q addresses into a fresh new Electrum wallet. But the private key seems to be invalid, too. I cannot import it in Electrum. See following example of a valid bcq1 that was found by VanitySearch:

In Electrum I click [File] --> New/Restore --> "Import Bitcoin addresses or private keys" --> and the description says:
I tried entering the Priv (WIF) that was output by VanitySearch
but I cannot click next. Even when I omit the p2wpkh: string and just entering the key solely. It doesn't work either, Electrum does not accept it. I tried some private keys which I can ensure they are valid and Electrum accepts them so I can be sure Electrum has no bug   As I experienced VanitySearch partially outputs invalid bitcoin addresses, now it seems that this is true also for the calculated private keys. I did further tests today just to realize that all private keys VanitySearch is outputting are invalid. Did anyone else experience this and can confirm? Is this a known bug or did I miss anything? Any help appreciated.

EDIT: Here some examples ...


VanitySearch' resulting "Priv (HEX)" is correct, but the Priv (WIF) is invalid. When I convert on my own from Priv (HEX) --to--> Address I get this result
1TryMeBmkZbYQRh24QZUSCjZfMuzJsK32 # manually calculated
1TryMeBmkZbYQRh24QZUSCjZfMuzgbqd6 # calculated and shown in VanitySearch



One more example with legacy address ...

VanitySearch calculated and shown Priv (WIF):
L5NB8KdCFkL7q8ZTAKAuHHeJhsP41XoJijqtuiZrXJffMEfzwaZy

But the correct corresponding compressed Priv (WIF) for the above Priv (HEX) should be correctly:
L5NB8KdCFkL7q8ZTAKAuHHeJhsP41XoJijqtuiZrXJffMEhSXy1u

That results in VanitySearch showing the wrong address:
1citboYAWtgbJCySZe8C7U7y36cxGfZUZ

Whereas correctly should be:
1citboYAWtgbJCySZe8C7U7y36cstvVjR

What's wrong here? Any clues?

Anyway. I hope this post can help other like-minded Linux users to get VanitySearch run on latest GNU/Linux version with CUDA support. If you have failed like I did, you can try this little tutorial shown here which in my case is related to Ubuntu 22.04 Jammy Jellyfish with Nvidia CUDA running. I wish you success and keep my fingers crossed. Thanks and good luck!!!

citb0in

With VanitySearch  engine
Can anyone help to modify BTCCollider to can custom rank search?

https://github.com/JeanLucPons/BTCCollider

if possible I would like to custom start seed from random 256 bit start to any seed want
I would like to start seed from 63bit and 119 bit for both puzzle64 and puzzle120

No, I've seen some replies who tested this with their Ryzen CPUs like this one (link)
He might have misunderstood the info that it doesn't work with AMD GPUs (Video Cards).

Somebody told me Vanity/BTCcollider could be used on Intel processors only. Is this true?
Can it be AMD as well?

We are currently messing with pricing matrices. some of you keep submitting things that have a heavier byte count therefore introducing a pivot going off length alone isnt ideal. (we've lowered the free difficulty for the time being)
It saves time if your running a single 1080 vs whats under our hood  but if you have a farm with a script you can can easily do this (the average retail investor that would want this doesnt have gpu farms)
but i think your referring to a simple .bat such as
save that as a .bat and right click and edit your prefix before running.
starting off with a lowercase d vs a D is a difference in a starting power of 23 vs 1354
so DKBiT9 can be found almost in about 1 min but dkbit9 would take like 30 mins.
we just want people to quit using junk like this http://vanitygen.net/ we can make it easy ,trusting and a little educational in the process.
even that guy charges for more then 3 digits , and then he sweeps your coin after you load it.
People have been getting scammed by services like these since samr7 released the first vanity search software.
if we can provide a consistent service maybe people wont fall pray to that kind of behavior anymore.

 

Asking owner of forum to ban/kill your account it's very definition of suicide and I think his words posted above will be enough, so please let's not continue going off-road with this.
I only mentioned him here because I created few vanity addresses with his name, that's all.
Now let's get back on topic.

I don't think the account was ever terminated. I think Lauda left voluntarily. All one has to do is change the password to xxxxxxxxxx-some-random-long-password, and no one else will be able to log into it. (obviously, from a secure device.)

I am far from being an expert in generating vanity addresses.
All I did was downloading the Vanitysearch open source software and typing few commands, it was as simple as it gets, and it took me only a few minutes to finish everything.
Someone could easily make one click script that would do this job even quicker, maybe something similar was already created but I am not aware of this.

I appreciate you are trying to make something easier to use, but I am not sure how many kids and grandmas really need vanity addresses.

I will think about testing your service and report later with my feedback, but like I said before in my case it will not save any time and in reality I would probably need more time for creating new email address.
It could be faster if I want to create address with more characters, but then I would have to pay more for this, and I don't want to do that just for fan.
So far I didn't found a way how to generate address using your website for free.

Oh Lauda was the one of the OG's of bitcointalk and he didn't really passed away in real life (at least I don't think he did) but his forum account was terminated by theymos.

I don't think this was true, during my testing.

For what it's worth: I don't think I've ever had any complaints from users who messed up. But it happened once or twice that someone forgot their part private key after they gave me the public key.

Literally one line above your quote stated we aren't targeting people who know how to do it,
At this current point in time there is only 3 legitimate services relevant to vanity splitkey generation
LoyceV's service,
bitcoin-uni.de's service
and our's.
all the rest are scams or not yet scams as they do not provide split key generation options at all.
We still haven't zero 'ed in on pricing but are getting close.
We do not log base keys (since we use rekey options we don't even see them in cli, we have learned switching up base keys generally yields results quicker, some basekeys simply dont yield results even on low difficulty wallets ),
All we have logged is the following.
Pubkey
Wallet Address
Partial Priv
and email
without a base key, the only flaws we can consider but won't are end users security practices.
even if someone's email where hacked or phished there isnt much that can be done with the partial priv and wallet besides speculating. unless they imported the wallet to some online source and it too was comprimised.
If end user truly kept the partial priv offline and lack of basekey brute forcing is to a moot end.
we tried to create a zero liability system by design.
We should be launching a finished product and announcing it on may 5th-8th'ish as our spamscore will be green light.
In concerns to the pricing  our competitors start charging $2 after 3 digits,  We don't start charging $2 until around the 7th digit or a certain difficulty, so it may feel expensive but its not.
bitcoin-uni.de 1ABCDEFGH=$432         -c $130.41,   1ABCDEFGHi -Cannot even submit this order 11111111 with -c enabled $26.91 and an email/refund apologizing that it cant be done.
 1splitkey.com 1ABCDEFGH=$117.50     -c $58.70,    1ABCDEFGHi -  $6818.10 and will deliver around 1636~ hours later. (66 days) 11111111 without -c enabled (since it cant be done that way) = $166799.90 we will get it done, it might take a year or 2 but we have ways to accomplish this weather it be on vast.ai rentals or A100 instances or an agreement with lambda labs

The difference in our services is we've done the research we had a viable product even before bitcoin-uni.de came out but saw many issues and challenges to overcome in our beta product.
We wanted everyone to be able to use our product not just the technically inclined.
When my child whom is under 10 years old was able to navigate and utilize the service we knew we where onto something. However my wife ,not so much. One's in 3rd grade the other is perusing a MBA.
One thing we are trying to factor in and solve is a proper workflow for assumptive behavior online.  

@dkbit98 I would love it if you tried the service in full fashion and then came back here to report the process, test out 1dkbit, use a burner email we dont care maybe next time you need something like 1Lauda you can know and trust our service and save a lot of time. (not that its time consuming or anything if you know how)
The neat thing about our service its all automated.
Ona side note are you referencing user Lauda I was unaware they passed away. Regardless of whom your referring to I send my condolences.

I don't think that sending private keys via email is a good option, especially when we know most people are using one email address and it's most likely gmail.
Just for fun I tested the pricing using various characters and I think price is fair (few bucks only) if I pick up to six characters, but it grows much higher adding just one more character.
I have to be honest and say that I honestly don't know why would I use 1splitkey when I can just generate my own vanity address for free with my computer.

Last year I did something similar, just for Luada's funeral:

reality is that if you're too long on one address it goes to aml blacklist. And then you best option is black exchange rate . So i don't see much sense in all this stuff anymore.

Thank you for vectorizing the logo and churching it up , some of the work you asked for was eye opening as well such as the difficulty differences.

I had never heard of vanity addresses until WhyFhy introduced me to them. The process I went through to create the public/private key was a breeze thanks to his instructions. I have zero experience in coding and/or brute forcing, so this was a neat project to be a part of. At no point was I asked for the private key. In my experience, no compromise in security. Now i've got a Vanity Address to show off and share my experience.

We are building something special, still working on a difficulty calculator like no other so we can offer up to 7 digits free
We could use some help solving this here as we have discovered JL's VS difficulty calculator is inaccurate at some points/pivots/bits.
problem is our difficulty calculator has to be more accurate than even vanitysearch's, so our quotes will be accurate so we dont operate at loss or email you guys saying we cant do it.
We run about 16,000 MK/S with -sp applied (so far)
We run a combination eth mining rigs to achieve this with proprietary software assigning workers, logging speeds and an api callback server relaying said data.
We've essentially created a real vanity searching mining pool(workaround) ,
We plan to use smi or wmic to pull card data and catagorize speeds against collected data such as DaveF's list and auto configure arguments.
and relay to website for time estimates. even a simple 7digit case insensitive can be done in under a few mins, It's like not hitting a mining share for few mins.
Our quotes are still broken, until we solve for a few x's on difficulty calculations via pivots/bits
We have a few other bugs to work out.
emails being marked as spam, or our personal smtp server not sending email in the first place thinking we are sending spam
We do not want to quote on wallet length like all the competition does. Our process if fully automated.
This has been a work in progress for about 16 months. Some of you tried our google form based beta. We took all the feedback with grace and went back to the drawing board.
We are not targeting guys like you that already know how to do it but if you lack power please feel free to use our service when it's officially announced.
We accept BTC and LN ,(implemented and tested working)
Again, the service isn't ready but at this point right now feedback is crucial as we near launch. We strive to be the definitive service in splitkey generation.
I offer you guys a preview at 1splitkey.com