Twisted Logic: Pros and Cons

[swami]

Since Edsel asked me about my overall evaluation/review of Twisted Logic, I decided to study its games. These are noticeable Pros and Cons and the frequency of similar mistakes/similar good ideas exhibited by the engine has caught my attention, so I decided to take notes of it.
This engine has good scope and will probably make its way to the top freewares.


This analysis is based on studies of atleast 70 games, frequency of usual mistakes happening that is commonly exhibited by the engine is outlined below.


Pros:

-Early attacker. Opens up king side pawns earlier in the game sometimes to launch an attack.
- Closed/Open Position: Plays excellent aggressive style, Better prefer open position chess.
- Excellent Piece Activity at any stage of the game.
- Good understanding of Space advantage, and Open files and Diagonals
- Good understanding of some pawn structures: Especially value of connected center pawns, Pawn Chains
- Good at making long range plans in closed position.
- Attack and Defense: Pretty good attacker, good defender as well.
- Outpost and Overprotection: No problem in there.
-Openings preferable: Audy's book is pretty good anyway.

Cons:

- Doesn't fully understand the pawn structures: Underestimates passer pawns, isolated pawns.
- Plays occasional computer like incomprehensible waiting moves, but rarely.
- Little understanding of Weak and Strong squares.
- Careful when opening up the king side pawns. Doesn't care king side pawns sometimes, In normal position, Instead of capturing the white bishop on f6 with Bishop on e7, it captured the piece with g7 pawn instead thus leading to lesser concern for kind side pawns.
- Little understanding of positional chess relative to other stronger engines.
- Bishop pawn endings need improvement.
- Better time management needed especially in blitz, slow at first and faster and bolder in the end.

Test Sets: Positional Chess Test Score: 50 out of 73: 68.4%

[BubbaTough]

Interesting post, looks like a very valuable analysis. A few questions:

1. Is this based on playing it, or looking at games it played against other computer programs, or both?
2. How strong are you . A range is fine such as (2000-2200, 2200-2400, 2400+).

-Sam

[swami]

Interesting post, looks like a very valuable analysis. A few questions:

1. Is this based on playing it, or looking at games it played against other computer programs, or both?
2. How strong are you . A range is fine such as (2000-2200, 2200-2400, 2400+).

-Sam

Thanks Sam.

It is based on analysis of 70+ games TL has played against the other engines in my tournament. I don't play tournament nowadays, last tournament I played seriously was 4 years ago, but my current playing strength at Playchess is 2350 in blitz. Playchess is what I have been using almost every weekend.

[BubbaTough]

Cool...sounds like you are at least as strong as me in chess then, probably stronger (something I have been looking for in a tester ). Any interest in me sending you a version of my program for you to take a look at? I have been working on time management recently, but a) still think it has problems and b) was terrible at it myself when I used to play tournaments and thus am robbed of my normal method of coding inspiration (introspection).

-Sam

[swami]

Cool...sounds like you are at least as strong as me in chess then, probably stronger (something I have been looking for in a tester ). Any interest in me sending you a version of my program for you to take a look at? I have been working on time management recently, but a) still think it has problems and b) was terrible at it myself when I used to play tournaments and thus am robbed of my normal method of coding inspiration (introspection).

-Sam

No problem Sam, I will look into it and send you report in case if you are interested, I could play a games against the engine myself and annotate it.
I've just sent you a pm.

Regards.

[Graham Banks]

Cool...sounds like you are at least as strong as me in chess then, probably stronger (something I have been looking for in a tester ). Any interest in me sending you a version of my program for you to take a look at? I have been working on time management recently, but a) still think it has problems and b) was terrible at it myself when I used to play tournaments and thus am robbed of my normal method of coding inspiration (introspection).

-Sam

Sam - just in case it helps you, LearningLemming 0.24 how has 300+ games in the CCRL 40/40 list.
You're able to download them if you wish.

Cheers, Graham.

[BubbaTough]

Sam - just in case it helps you, LearningLemming 0.24 how has 300+ games in the CCRL 40/40 list.
You're able to download them if you wish.

Cheers, Graham.

Thanks Graham!

I downloaded some of those when I first started trying to get back into programming, very helpful. I appreciate all the work you guys have done.

There have been a number of changes in the new version I am testing right now (I will send you guys a copy when things look "good enough", since CCRL spends so much time testing each copy I don't want to bog you down by spamming you with versions). Some of it is preparation for a multi-threaded version whose only effect may be to introduce bugs, but there is also some eval, search, time management, and a little "don't hork the processor" stuff (horking the processor helps make the engine a little stronger, but as I start preparing for public release I need to make it a little friendlier).

One thing CCRL games can't help with though is time management, because the ponder off and the X moves in Y minutes stuff is quite different from how the engine is normally run (on ICC, in tournaments, or in casual play). In addition to exploiting Swami's chess brain, I look forward to picking his brain about time management. I think this is an under-appreciated area for improving chess strength, and I don't think its well understood how humans should do it, let alone how to teach computers to do it.

Another thing CCRL games don't do is give me an excuse to send Swami a version, which is a sign of trust and thanks for donating his time helping me out .

-Sam

[Edsel Apostol]

Thanks Swami for doing my request. This is a very valuable analysis. I'm going to print this out to study and this will serve as a basis on where I'm going to work to to improve Twisted Logic.

I am actually aware of some factors you've listed here and its nice that it was confirmed. Sometimes I don't know which things to prioritize and feedback like this seems to help me start sooner in the right direction.

Big thanks again!

[swami]

Thanks Swami for doing my request. This is a very valuable analysis. I'm going to print this out to study and this will serve as a basis on where I'm going to work to to improve Twisted Logic.

I am actually aware of some factors you've listed here and its nice that it was confirmed. Sometimes I don't know which things to prioritize and feedback like this seems to help me start sooner in the right direction.

Big thanks again!

Hi Edsel,

Actually the report is not yet fully complete! I was planning to add more weaknesses I had taken note of, soon as I confirm its frequency and number of times the same mistake occured in other games.

I wrote this report yesterday based on 70 games alone.

- Now there is another 30 games (from Grand Blitz Tourney)
- and 60 games from Division 5 tournament,
- and few game I play against it...
- My own selected Tactical Test Suite scores for the engine.

These are all pending.So kindly please wait before I make the complete corrected list in the next few days

Ps: Ulysses might like to try out this engine if he likes Chess Thinker, One noticeable thing about Twisted Logic engine is its pure uniqueness in handling the pieces overall, the piece activity.

Best Regards.

[Ovyron]

When I have the time . I've put Twisted Logic on my top priority list.