camosoul
Well-known member
I reached out to a few of the devs and people on the periphery of DASH coding, but it's a weekend, so...
I don't post here often, and that's intentional. I'm on the edge of losing my mind and that's pretty obvious. I know what I'm like and I try not to spread it around too much... Misery loves company, but it can't be pleasant for the company... So I try to keep it to myself. But, sometimes it gets the best of me... What little you see of me is when I can't contain it anymore. I really do try... Those of you who've dealt with me behind the scenes know I'm really not so terrible... I may be cranky and abrasive, but no one can call me a liar or a thief. ...I'm not doing so hot, and it shows...
On with it:
I stored a large amount of DASH in my Trezor T using DMT to put it in random locations. In case I got $5 wrenched, it'd still be hard to find the addresses. The DMT config has been lost due to the destruction of my laptop and both of the backup drives. So, the addresses/BIP32 Paths are no longer known even though the Trezor works fine. I need to search the BIP44/BIP32 paths of an enormous span, and compare them all to a known segment of the blockchain with TXes of a certain size. This will definitely fit in RAM. I need to do it really, really fast... There needs to be close to Terrachecks per second... If a GTX 1080 Ti can approach 0.4GHash/s... And that in any way correlates, this might just be doable. Since it's merely a comparison of a memory-resident value and a BIP32 calculation, which is less work than hashing... I'm guessing it can be done fairly fast. Parallellism will be key. I hope RAM bandwidth will be more of a concern than the actual compute power.
The Trezor is intact.
If I can get the account and address data from an xpub using trezorctl or some such, I should be able to run through the entire mess looking for any addresses with a balance. Add that address and BIP32 path to a list. Then, I can put them back into DMT and use DMT to send the funds to a new address.
Example:
m/44'/5'/123456789'/0/123456789
I need to search everything
from: m/44'/5'/0'/0/0
to: m/44'/5'/999999999'/0/999999999
I need to pull the xpub data in a manner that does not require me to manually operate the Trezor every time. Once extracted in whole, it can be worked on freely.
I can then scan all addresses from 0 to 999999999 in all accounts from 0 to 999999999 for a balance, and save to a list of all the addresses and the associated bip32 paths found which have funds on them.
I can then put this path/address into DMT to send those funds out to a new address.
Perhaps multiple machines working in parallel on different sections of the address space? That's the idea. Put the crypto community's old video card mining rigs to use.
I can verify that I am searching the correct Master/root because I have some very old backups which contain addresses that previously had a balance. I can look them up on the blockchain and if they match up, I know I've got the correct xpub Master/root.
999,999,999 x 999,999,999 =
999,999,998,000,000,000 addresses to check.
Every millisecond will matter... It has to be very efficient. All it really has to do is generate the address from the BIP32 path and then see if it exists in the selected list of possible addresses in the blockchain. If it finds one, it then checks to see if it has a balance greater than X DASH.
Put the address, the BIP32 path, any TXID and the current value in a list/file. It will have to move very fast to find anything in my lifetime... I'll need video cards, and lots of them... More than I can fit in my house. Aren't there a lot of video card mining rigs out there?
We can also narrow it down to a fairly small section of the blockchain. I know a date range, and I know an amount below which there's no reason to be looking, and that rules out almost everything... The list of possible addresses will be short and will definitely fit in RAM. Break it up into work sections with a forked distributed.net client?
I'm not sure that everything here is even possible... I certainly can't do it alone. If thousands of machines are on the job, it'll be done in less than a month... Weeks maybe. Those of you who have helped me out in the past know I don't screw around when it comes to compensating people who help me. I'm still working out the math... I'll come back and edit this with a number when I figure it out. It will be paid in DASH, that much I can tell you... I want to keep it more like a Lottery Ticket. I don't want to distribute rewards like mining does. This way there's a chance that a guy alone with his one video card might hit a decent jackpot. Those with big farms might miss, but they won't have lost much if that's the case because this'll go fast. There's 10 of them in this address space. So, that means there's over 10,000 DASH to be found in total.
EDIT: I've decided the pot needs to be sweeter. There are 10 addresses to be found. Each one has at least 1000 DASH on it because it was a masternode. Which is still running... For each one of them found, the bounty is 30% of the total found.
So, a minimum of 300 DASH for each one found, probably a little more. The coder(s) who make this happen get 20% of that, and they get it for all 10. The finders get 80% for each one they find.
Payout occurs immediately when the funds move off of the address.
That's a total bounty of roughly $200,000 at the current price, which is terrible and likely to go up quite a bit by Feburary... I'm not asking for charity. I'm not screwing around. This is real money. To add to it, since they'll likely happen one at a time, you'll see that I'm not lying when I pay out the first one, and the other 9 still remain. this is not some bullshit scam, I'm not asking you to send me anything, not asking you to invest anything, I don't want your private keys, no moonshot nonsense, none of the crap... It's a salvage mission. The coins are in there. You just have to find them. and that's going to take math. Lots of it.
A little backstory... They're all masternodes I was running/had set up for a boomer. I already paid him for my fuckup out of pocket. I'm trying to mitigate that loss and get some back.
The objective is not to check the entire keyspace for a single objective, like distributed.net doing RC5-72...
The goal is to find exactly 10 Easter Eggs scattered about that space. Even if I only find some of them, it's worth it. But if video card mining rigs get thrown at it, it'll be far more profitable than mining because it won't take long.
A mechanism much like distributed.net would make sense. Except it wouldn't really take up any space or bandwidth, because it would just be coordinating a really big counter. "Hey, you check this group of BIP32s, see if you find a match. Let me know if its empty and we should look elsewhere, or if you found an Easter Egg in that thicket you get some DASH!"
Maybe some of the video card mining code out there can be bent to the task? Distributed.net's client already supports a ton of video cards for their purposes. It might be a copy pasta fork...
I guess my misfortune can benefit some people. We're all hurting due to the politically weaponized flu bs. Who doesn't need some money?
Hopefully there's no showstopper up near the top of this hair-brained plan... Asking for input from a few people smarter than I... Not that it's hard to be smarter than me...
@UdjinM6 @moocowmoo @antouhou
I fucked up real bad... As in, life ending fuckup. I still can't believe it happened. I did everything right. I'm almost hoping the stress will finally be the end of me... I don't see a way to live past this shit, or any reason to bother trying. I'm not going to demand a chain rollback or free stuff, no gofundme, no donations... I own this even if it ends me. You help me, I help you. I can't code for shit... I can't do this. I wouldn't subject you to looking at me if there wasn't a good reason...
AND DAMMIT! AS I'M EDITING THIS A TREE FELL ON MY HOUSE!
I don't post here often, and that's intentional. I'm on the edge of losing my mind and that's pretty obvious. I know what I'm like and I try not to spread it around too much... Misery loves company, but it can't be pleasant for the company... So I try to keep it to myself. But, sometimes it gets the best of me... What little you see of me is when I can't contain it anymore. I really do try... Those of you who've dealt with me behind the scenes know I'm really not so terrible... I may be cranky and abrasive, but no one can call me a liar or a thief. ...I'm not doing so hot, and it shows...
On with it:
I stored a large amount of DASH in my Trezor T using DMT to put it in random locations. In case I got $5 wrenched, it'd still be hard to find the addresses. The DMT config has been lost due to the destruction of my laptop and both of the backup drives. So, the addresses/BIP32 Paths are no longer known even though the Trezor works fine. I need to search the BIP44/BIP32 paths of an enormous span, and compare them all to a known segment of the blockchain with TXes of a certain size. This will definitely fit in RAM. I need to do it really, really fast... There needs to be close to Terrachecks per second... If a GTX 1080 Ti can approach 0.4GHash/s... And that in any way correlates, this might just be doable. Since it's merely a comparison of a memory-resident value and a BIP32 calculation, which is less work than hashing... I'm guessing it can be done fairly fast. Parallellism will be key. I hope RAM bandwidth will be more of a concern than the actual compute power.
The Trezor is intact.
If I can get the account and address data from an xpub using trezorctl or some such, I should be able to run through the entire mess looking for any addresses with a balance. Add that address and BIP32 path to a list. Then, I can put them back into DMT and use DMT to send the funds to a new address.
Example:
m/44'/5'/123456789'/0/123456789
I need to search everything
from: m/44'/5'/0'/0/0
to: m/44'/5'/999999999'/0/999999999
I need to pull the xpub data in a manner that does not require me to manually operate the Trezor every time. Once extracted in whole, it can be worked on freely.
I can then scan all addresses from 0 to 999999999 in all accounts from 0 to 999999999 for a balance, and save to a list of all the addresses and the associated bip32 paths found which have funds on them.
I can then put this path/address into DMT to send those funds out to a new address.
Perhaps multiple machines working in parallel on different sections of the address space? That's the idea. Put the crypto community's old video card mining rigs to use.
I can verify that I am searching the correct Master/root because I have some very old backups which contain addresses that previously had a balance. I can look them up on the blockchain and if they match up, I know I've got the correct xpub Master/root.
999,999,999 x 999,999,999 =
999,999,998,000,000,000 addresses to check.
Every millisecond will matter... It has to be very efficient. All it really has to do is generate the address from the BIP32 path and then see if it exists in the selected list of possible addresses in the blockchain. If it finds one, it then checks to see if it has a balance greater than X DASH.
Put the address, the BIP32 path, any TXID and the current value in a list/file. It will have to move very fast to find anything in my lifetime... I'll need video cards, and lots of them... More than I can fit in my house. Aren't there a lot of video card mining rigs out there?
We can also narrow it down to a fairly small section of the blockchain. I know a date range, and I know an amount below which there's no reason to be looking, and that rules out almost everything... The list of possible addresses will be short and will definitely fit in RAM. Break it up into work sections with a forked distributed.net client?
I'm not sure that everything here is even possible... I certainly can't do it alone. If thousands of machines are on the job, it'll be done in less than a month... Weeks maybe. Those of you who have helped me out in the past know I don't screw around when it comes to compensating people who help me. I'm still working out the math... I'll come back and edit this with a number when I figure it out. It will be paid in DASH, that much I can tell you... I want to keep it more like a Lottery Ticket. I don't want to distribute rewards like mining does. This way there's a chance that a guy alone with his one video card might hit a decent jackpot. Those with big farms might miss, but they won't have lost much if that's the case because this'll go fast. There's 10 of them in this address space. So, that means there's over 10,000 DASH to be found in total.
EDIT: I've decided the pot needs to be sweeter. There are 10 addresses to be found. Each one has at least 1000 DASH on it because it was a masternode. Which is still running... For each one of them found, the bounty is 30% of the total found.
So, a minimum of 300 DASH for each one found, probably a little more. The coder(s) who make this happen get 20% of that, and they get it for all 10. The finders get 80% for each one they find.
Payout occurs immediately when the funds move off of the address.
That's a total bounty of roughly $200,000 at the current price, which is terrible and likely to go up quite a bit by Feburary... I'm not asking for charity. I'm not screwing around. This is real money. To add to it, since they'll likely happen one at a time, you'll see that I'm not lying when I pay out the first one, and the other 9 still remain. this is not some bullshit scam, I'm not asking you to send me anything, not asking you to invest anything, I don't want your private keys, no moonshot nonsense, none of the crap... It's a salvage mission. The coins are in there. You just have to find them. and that's going to take math. Lots of it.
A little backstory... They're all masternodes I was running/had set up for a boomer. I already paid him for my fuckup out of pocket. I'm trying to mitigate that loss and get some back.
The objective is not to check the entire keyspace for a single objective, like distributed.net doing RC5-72...
The goal is to find exactly 10 Easter Eggs scattered about that space. Even if I only find some of them, it's worth it. But if video card mining rigs get thrown at it, it'll be far more profitable than mining because it won't take long.
A mechanism much like distributed.net would make sense. Except it wouldn't really take up any space or bandwidth, because it would just be coordinating a really big counter. "Hey, you check this group of BIP32s, see if you find a match. Let me know if its empty and we should look elsewhere, or if you found an Easter Egg in that thicket you get some DASH!"
Maybe some of the video card mining code out there can be bent to the task? Distributed.net's client already supports a ton of video cards for their purposes. It might be a copy pasta fork...
I guess my misfortune can benefit some people. We're all hurting due to the politically weaponized flu bs. Who doesn't need some money?
Hopefully there's no showstopper up near the top of this hair-brained plan... Asking for input from a few people smarter than I... Not that it's hard to be smarter than me...
@UdjinM6 @moocowmoo @antouhou
I fucked up real bad... As in, life ending fuckup. I still can't believe it happened. I did everything right. I'm almost hoping the stress will finally be the end of me... I don't see a way to live past this shit, or any reason to bother trying. I'm not going to demand a chain rollback or free stuff, no gofundme, no donations... I own this even if it ends me. You help me, I help you. I can't code for shit... I can't do this. I wouldn't subject you to looking at me if there wasn't a good reason...
AND DAMMIT! AS I'M EDITING THIS A TREE FELL ON MY HOUSE!
Last edited:
