I think this article is worth a read. I would be curious to know whether it is technically accurate.
Yes but a lot of optimizations seemingly involve turning the 1-in-2-out transaction into a special, optimized version by removing data that isn't needed for that. So these improvements wouldn't apply across the board, and the claimed "free blockspace" increase would also only reach that 1.2MB if every single transaction was a 1-in-2-out.
reply