Do you like your games to rate you?

[MarkShot]

I have Deep Fritz and it only makes difference in the speed of post game analysis. But do you really have the patience to analyze all your games? Most of mine, I don't. When it comes to games, chess is one you can invest a huge amount of time in self improvement. I like learning a little here and there, but I don't want it to become a job.

[Bison]

I thought multi-core support made it think faster.

[MarkShot]

Chess engine computation is measured in KNS.  Kilo Nodes per Second. A node is basically the numerical rating assigned to a board position. KNS increase per cores is approximately linear.

So, it depends how you play your games. If you play friend mode, it doesn't matter. If you play blitz or give Fritz X seconds per move AND IF YOU ARE GOOD ENOUGH, then it matters. Of course, a fast CPU matters.

Where you really see the jump is on post game analysis. Maybe you allow Fritz 15 seconds/move on the analysis. With 1 core, you get 10 ply (move look ahead), but with 4 cores, you get 15 ply.

So, Deep Fritz either gives you the same detail much faster or more detail for the same time. But as I said a lot depends on what you will commit to learn and your own level.

[Bison]

Mark I literally did a dance of joy the first time I beat the computer.  Unfortunately I don't get to do that dance too often.  Best if I gimp the computer as much as possible when it comes to chess!

[MarkShot]

The problem with most game analysis is it gives you blunders and best moves. But it doesn't identify overall strategic problems in your game. For that, you need to study.

[Bison]

If I may represent the response the computer would have once it started to analyze my blunders.

[jomni]

Stats in PBEM in http://theblitz.org/ is useful.  Not only do you know how you are performing, you also know if the scenario is balanced.