So said Emanuel Lasker. Before a player can begin calculations he has to have something to calculate and where does this "something" come from? Ideas. An idea is suggestion about what to do that you imagine or picture in your mind. If it wasn't for ideas we would have to calculate like an engine and examine and evaluate every move. So, where do ideas come from?
Dutch psychologist de Groot discovered that a key element is the master's ability to recognize patterns and when shown a position, they divide it up into recognizable patterns consisting of 4 of 5 chunks with features that are remembered from other games or positions.
How do you build up your storehouse of patterns? David Bronstein wrote that most tactics are inspired by the recall of previous games that have been played over, so the obvious answer is to play over games...lots of them. Does it really help? Consider the following examples.
Lasker's double Bishop sacrifice against Bauer at Amsterdam 1889 is pretty well known and the Kuzmin's game against Sveshnikov, Moscow 1973 employs a similar idea. In fact, in the tournament at Tashkent 1989 the game Barsegian vs. Garafutdinov was almost identical to Kuzmin's except black varied moves a little bit at the end of the game. Then in St. Petersburg 1914, playing black against Nimzovich, Tarrasch pulled off a double B sacrifice, but the game was denied first brilliancy prize because the judges believed Tarrasch had just reworked the Lasker vs. Bauer game.
In the following to game fragments we see Bogoljubow playing a nice combination that makes you wonder where he got the idea from. It was was based on an idea from a famous Bird vs. Morphy game, London 1858. As Tahl once said, you don't have to reinvent the bicycle. Just be familiar with patterns.