Yet the term "double spend" is used in so many different ways. Maybe you explain exactly what your plan is.
Didn't he already explain that?
Now before everyone's going to say double-spending is impossible i want to say that i want to make a double-spend attempt means one transaction will confirm and one will drop.
He wants to broadcast 2 different transactions that are both similar size (in bytes), include similar fees, and spend at least 1 identical input. Then he wants just 1 of those transactions to confirm.
From this description:
What I want to make is either:
- Push two transactions at exactly same time to different peers so that will both get relayed and get to different nodes on the network, but in the end only one will confirm
- Or else discover a node which relays double-spend tx's and will allow me to double-spend that transaction at any time.
Edited by me for grammar and clarity.It sounds to me like OP would be satisfied if he could find software that would connect to multiple peers, and allow him to set up 2 different transactions that spend the same output, then broadcast one of those transactions to half of the connected peers while simultaneously broadcasting the competing transaction to the other half of the connected peers. (blockchain.info used to have a web page that would allow this. They seem to have gotten rid of it)
It also sounds to me like he would be satisfied if he could find a mining pool that would accept "most recently received" transaction instead of "first received transaction", and if there were a way to send a competing transaction directly to that pool. (I don't think there are any such pools)
It isn't clear to me if he would be satisfied if he could find a mining pool that would accept "highest fee transaction" instead of "first received transaction". If so, then he could try
BitUndo. (although I don't think BitUndo has very much hashing power, so the likelihood that your replacement transaction would be confirmed is pretty slim).
What amaclin suggest is IMHO not double spend as only one of those two TX will confirm.
Correct, but OP only wants one transaction to confirm. He just seems to want at least a chance of getting a replacement transaction confirmed instead of its competing transaction.
NOTE: Bitcoin is designed to make it VERY difficult to do what the OP is attempting. As a technical exercise, it is possible to write software that will broadcast competing transactions simultaneously to multiple peers, but I'm not aware of any such software that is available and safe to download and run. Aside from simultaneous broadcast, the other things that the OP wants to do would generally require waiting several days for the first transaction to be dropped from most nodes on the network.