Yet it's also about solving a problem, that goes out of memory. If there is an equal algorithm that doesn't go out of memory isn't this a valid criticism? The

Yet it's also about solving a problem, that goes out of memory. If there is an equal algorithm that doesn't go out of memory isn't this a valid criticism? The author probably spent some time writing the code, modifying and extending the algorithm. He wrote deduplication code, memory mapping etc. Wouldn't it be interesting to know how iterative deepening performs?