Chess and automaticity

I have played an awful lot of chess to be as bad at it as I am. I have a chess app that sees two moves ahead, and I beat it about one time in ten. The problem is that I play intuitively, doing things that I believe will be good ideas, usually because I've done them before - they are the kind of thing I do. I long ago took the correct insight that conscious processing of all possibilities will not be powerful enough to always win, and now act as though unconscious/automatic processing will refine itself over time into an unbeatable victory engine. In fact, conscious processing is the way to begin accumulating unconscious patterns to allow deeper insight into the future. Without puzzling it out, I have not improved.

This is a bit of a bad mental habit for me, which may be related to my mother praising 'cleverness' rather than 'effort.' This was heavily reinforced by schooling in a system that required respectful behaviour and a home life that fore-armed me with ideas. I think I actually did about 5% of my assigned homework, and so my facility with everything skilled, from algebra to music to Japanese to essay composition remained almost static.

At the moment I am struggling to become a skilled researcher instead of just a magpie-minded popinjay. But this long habit of believing that effort is a sign you should just give up, because you are innately unqualified and can't become qualified is hard to break. I need to get a virtuous dopamine circle going around reading things and learning, instead of a vicious shame-spiral at not already knowing them.

Published with Blogger-droid v1.6.7

Game Theory

I've been listening to Yale's Ben Polak on Game Theory.  It's very good.

Here are some highlights:

• Never play a strictly dominated strategy (if given a choice).
• Look for a strategy or mix of strategies that will make your opponent's choice of strategy indifferent, and is also a 'best response' for you, and you have a Nash equilibrium.
• Evolutionarily Stable Strategies are Nash Equilibria (an ESS is an evolution game that has reached an end point - male sea elephants have to be huge to hold down a big harem and have a lot of offspring; but the non-huge bulls can sneak around among the females. The ratios of effort spent on these two strategies prove to be a Nash equilibrium).
• Sometimes more choices make your situation weaker - burning your boats or otherwise making a total commitment can increase your chance of a positive outcomes; but it is important that your adversary knows you're committed, and that it is that sort of game.
• A game known as 'battle of the sexes' assumes that two people going on a date have slightly different taste in film but would prefer to meet up, rather than not. But if they both go to their least favourite film in the hopes of meeting the other, they lose out.  What was particularly interesting was that the example 'couple' for the class exercise won out every time because the woman went to her favoured film, and the man went to her favourite film.  The traditional practice that in courtship, the effort should be on the man's side, beat the rational game rules.

BattleTech

I read my first BattleTech novel today.  It was a gift from my brother-in-law, who rolled his eyes as he handed it over.  It was not a very good novel, and if anything a closer examination of the world of BattleTech made it seem less convincing. Powerful, machine-mounted laser weapons missing?  At ranges of a few hundred meters?  Interstellar travel is mastered, but nukes and anti-matter and space bombardment and artillery are not effective?

 But, I'm pretty sick and exhausted, and the characters were better than you might expect, and I was reminded that I had quite enjoyed playing MechCommander.

It's always worth Googling things, too, because I was amazed to find Mechcommander 2 was released as an example with source code by Microsoft. It is minus all the supporting video and animation, and I'm not sure about the in-game art.  Unfortunately it was a Windows XP game, I'm not sure I'll be able to get it running on Windows 7.  Maybe Wine?  Or I do still have the windows Me machine...

Update: Full release (which was made available free by the original publishers at the same time that Microsoft used it as an example) allegedly available here. Smaller download too.

Illustration: Story & Game

Here is about the simplest game you can imagine.  A team is divided into two even halves and they are positioned in squares facing each other, with one empty square between them.  Each member can step ahead into an empty square, or jump a player from the other team into an empty square. They can only move forward.  The game is cooperative - the teams can talk and agree how to handle the situation.

This setup means that at each turn, there are only two possibilities: red or blue can move to occupy the space, creating a space where they were before.  Our group of ten solved this on the third attempt, but some groups never solve it. The 'choice map' is my attempt to show these potential paths - but almost all of them will end up in blockage on the next move or two.

The thing that is like Story (as McKee describes it) is that interest arises from high-stakes choices between apparently equal 'goods'. Dramatica assumes that the point of a 'grand argument' narrative is to teach a lesson, to illuminate a world in a fruitful way.  The best stories furnish us with the moves and types and experiences that make us a little more awake, a little wiser, a little lighter in the step, by illustration.

Our group solved the problem when, on the third iteration, we had learned something about which choices not to make.  No one of us remembered the map of the whole field, but each of us remembered the particular mistakes near us, or involving us, which caused a breakdown in the earlier attempt.  The difficulty for a story-teller is to present regular, realistic challenges that offer this sort of dilemma - things that rather than build character, reveal it.  The story-teller too has a choice, but with this illustration, the better choice may come a little easier.

Robot Defense Free

Today i completed the 'huge score' achievement of 50,000,000 lifetime points in Robot Defense, one of the more popular Android games. Given it takes me about 20 mins even on the highest rated game available to accumulate 200000 points, this is at least 80 hrs investment, and probably much more.

The game is well-balanced, and a very good model of drama. In the course of the game you make many choices in the dilemma of apparently equal goods - missiles or ant-air cannon? Augment the slow tower or convert it to a mine? And the opposition steadily mounts until you fail, a tragedy, for which the seeds almost always appear to have been sewn early (poor layout, not enough anti-vehicle, not enough anti-air, oh my foolish and misplaced enthusiasms!). If you complete 100 levels intact, all is well after all, a comedy, with the affirming moral that human, not robot kind will forever dominate the earth.

Published with Blogger-droid v1.6.7