3 ms·
He does say a little further on that he saw one or two errors when pieces of the opposing colour were on those squares. That's not so hard to correct if you've
by archangel_one 15y ago
He does say a little further on that he saw one or two errors when pieces of the opposing colour were on those squares. That's not so hard to correct if you've tried to detect every square but would be catastrophic if you only looked at one.
- meatsock 15y agotrue. so, test the color of one or two squares detected as empty at random and extrapolate the rest. my point was simply that the pattern rarely changes, and a corner will be white-black or black-white and from that you can determine which way the board is facing, and what every other square's color should be.