Fritz (12) and the limit strength feature

[Alexander Schmidt]

I tested your engines with a human (myself). I wouldn't play it with a long control, every game was played at 5 minutes or something like two seconds per move. Maybe you should try a bit more in shorter control. Nobody would bother to sit and play a computer for a standard control. I also don't find much improvement in your LS settings against the default settings in Stockfish.

Forget about StockFish and Fruit. They where a kind of first try, not well tested. StockFish is no improvement in the LS version because I use the StockFish algorythm based on a selection of MultiPV moves. I just calibrated it to ELO values a little. Fruit is also too good to get a good style at fast timecontrol with low ELO because it is slowed down too much.

SlowChess is the version to go. If this one also is too week at fast timecontrol I can speed it up. Just give me some numbers what ELO you have to set that it plays at your level.

About the timecontrol: You are wrong. I started with this engines to play long timecontrol games. I created a hand made wodden board with a 800MHz thin client. I wanted a adjustable engine for that with good tuned ELO values. And I indeed play vs this machine long timecontrol games. This is the only reason why this mods exist

[kinderchocolate]

Wow, that's amazing. I didn't know that, you should have announced on somewhere with a website or something like that Are you planning to sell it : - )

I'm not sure but if you are interested in offering your knowledge in limit strength for a chess engine to thousands and thousands of chess players around the world, you might want to help me. I'm the developer of SmallFish, a port of the latest Stockfish on the iOS platform urgently needs a new and more reliable way to reduce the strength. The current strength reduction algorithm implemented by the Stockfish team is not very good. It is an open source project, so if you want to benefit the entire world, please contact me. You will be fully credited. SmallFish has a very high download rate, your efforts certainly won't be wasted.

I will try SlowChess, maybe one day I will port it...

[Alexander Schmidt]

I won't sell it, it is an open source project Everything you need to build your own is available here:

http://www.miclangschach.de/forum/

I have no better idea for reducing the strength of StockFish. In my opinion it is better to reduce the nodes instead of messing up the evaluation. This works well for SlowChess, but to get below 2000 ELO with StockFish I still had to use the original weak levels. Here is a nice thread in talkchess about this topic: http://www.talkchess.com/forum/viewtopic.php?t=45795

[kinderchocolate]

Alexander, do you think you can write a patch for the latest Stockfish using whatever reduced method you think the best? I think the UCI Limit in the Stockfish chess engine is broken, it just doesn't work.

[Alexander Schmidt]

I think the UCI Limit in the Stockfish chess engine is broken, it just doesn't work.

You have to start a new game after reducing the strength, then it works.