You could be looking at something in excess of 200 valid seeds assuming a relatively similar "hit" rate of 9-10 words per position... So how would you know that you'd generated the correct one?
I can think of a solution for this: simply extract a list of all Bitcoin addresses that ever had a balance, and search those 200 addresses in the list.
Quandl.com gives the number of unique addresses used per day, but I'm not sure how accurate this is. Let's assume on average 200k addresses for 8 years, say 600 million in total, 20 GB data. My simple Linux-laptop could search that 200 times in a few minutes. It could be much faster if I split the list into parts that fit my RAM. Setting up a database would be more work, but could speed up the search in case many more addresses need to be tested.
I'm now intriqued to be able to do this. I need:
-a command-line method to extract addresses from a word seed
-a method to extract unique addresses from my Bitcoin Core database
Once I have both, I can easily script them together to run a search.
this would take several weeks to brute-force, but if the incentive is big enough, it might still be worth it.
My estimate is a lot less, assuming I can figure out the two required tools I mentioned above. It would be a nice think to master. On the forum here someone once shared a program that quickly searches up to 5 missing characters from a private key. I saved it, it might come in handy some day.
I was hoping maybe if I had most words,
How do you know for sure you have all but one word? Is this based on the total number of words you have?
