Volker Annuss wrote:
My first reason to for creating a Glaurung clone was a remark by Marco Costalba how hard it is to improve Glaurung or Toga by 20 ELO. In my opinion it is relatively easy for anyone who wrote a chess engine from scratch to improve Glaurung (but not Toga) by 20 ELO. I wrote so in another thread some weeks ago, and I wanted to show it, so I started "cloning".
I wish you good luck with your efforts but it does seem you are out to prove common sense. I decided to take the challenge of improving Toga by 20 elo after reading your remarks. I have so far spent 7 minuets of programming effort. None of the code added to Toga would work as is in Fruit, the changes are really just that simple. I am looking for a trusted person to run some more games and to verify that the added code is infact 7 minuets worth of programming effort. I will send source code only and the person will have whatever rights the GPL gives from there.
Here is a pgn of the 14 games it played while I was at lunch against the version it was derived from. http://www.fruitchess.com/duh.pgn
Volker Annuss wrote:
My first reason to for creating a Glaurung clone was a remark by Marco Costalba how hard it is to improve Glaurung or Toga by 20 ELO. In my opinion it is relatively easy for anyone who wrote a chess engine from scratch to improve Glaurung (but not Toga) by 20 ELO. I wrote so in another thread some weeks ago, and I wanted to show it, so I started "cloning".
I wish you good luck with your efforts but it does seem you are out to prove common sense. I decided to take the challenge of improving Toga by 20 elo after reading your remarks. I have so far spent 7 minuets of programming effort. None of the code added to Toga would work as is in Fruit, the changes are really just that simple. I am looking for a trusted person to run some more games and to verify that the added code is infact 7 minuets worth of programming effort. I will send source code only and the person will have whatever rights the GPL gives from there.
Here is a pgn of the 14 games it played while I was at lunch against the version it was derived from. http://www.fruitchess.com/duh.pgn
I can confirm that it is indeed 7 minutes and 2.4 seconds of programming Ryan.
Thanks very much for this stupendous effort. I would like to name this new program as per your wishes.
I wish to name it TuttiFruity 7.04 (mathematically correct we believe).
Is that OK? I will abide by the GPL and .......interpret that in my own way as is the fashion.
Our team of testers are working really hard right now trying to determine it's strength without running the program. This is very kind on the electricity bill and we believe makes your program the first "green" chess engine.
I'll let you know in due course which functions we think give the most gain in elo and present those results here.
Once again thanks. Global warming just got a little cooler.
Volker Annuss wrote:
My first reason to for creating a Glaurung clone was a remark by Marco Costalba how hard it is to improve Glaurung or Toga by 20 ELO. In my opinion it is relatively easy for anyone who wrote a chess engine from scratch to improve Glaurung (but not Toga) by 20 ELO. I wrote so in another thread some weeks ago, and I wanted to show it, so I started "cloning".
I wish you good luck with your efforts but it does seem you are out to prove common sense. I decided to take the challenge of improving Toga by 20 elo after reading your remarks. I have so far spent 7 minuets of programming effort. None of the code added to Toga would work as is in Fruit, the changes are really just that simple. I am looking for a trusted person to run some more games and to verify that the added code is infact 7 minuets worth of programming effort. I will send source code only and the person will have whatever rights the GPL gives from there.
Here is a pgn of the 14 games it played while I was at lunch against the version it was derived from. http://www.fruitchess.com/duh.pgn
I can confirm that it is indeed 7 minutes and 2.4 seconds of programming Ryan.
Thanks very much for this stupendous effort. I would like to name this new program as per your wishes.
I wish to name it TuttiFruity 7.04 (mathematically correct we believe).
Is that OK? I will abide by the GPL and .......interpret that in my own way as is the fashion.
Our team of testers are working really hard right now trying to determine it's strength without running the program. This is very kind on the electricity bill and we believe makes your program the first "green" chess engine.
I'll let you know in due course which functions we think give the most gain in elo and present those results here.
Once again thanks. Global warming just got a little cooler.
Volker Annuss wrote:
My first reason to for creating a Glaurung clone was a remark by Marco Costalba how hard it is to improve Glaurung or Toga by 20 ELO. In my opinion it is relatively easy for anyone who wrote a chess engine from scratch to improve Glaurung (but not Toga) by 20 ELO. I wrote so in another thread some weeks ago, and I wanted to show it, so I started "cloning".
I wish you good luck with your efforts but it does seem you are out to prove common sense. I decided to take the challenge of improving Toga by 20 elo after reading your remarks. I have so far spent 7 minuets of programming effort. None of the code added to Toga would work as is in Fruit, the changes are really just that simple. I am looking for a trusted person to run some more games and to verify that the added code is infact 7 minuets worth of programming effort. I will send source code only and the person will have whatever rights the GPL gives from there.
Here is a pgn of the 14 games it played while I was at lunch against the version it was derived from. http://www.fruitchess.com/duh.pgn
Congratulations that you achieved where many others failed Ryan!
I pointed Jerry Donald to your post here but I don't think that he has seen it yet. Is there any chance that we could get to see your ideas for the changes in Toga? We would of course be very interested. If it involves ideas you would rather keep for Fruit then that would be totally understandable. But if as you say everything could be programmed on a napkin during a 7 minute lunchbreak leaving sparetime for testing, why not share it with the rest of us? Sending it to Chris will only mean he changes everything back to the original Eden Malum Pinicum 0.1 and he probably swallowed the code after receiving it. Standard procedure for a tester. Nobody will ever see or hear from it again
The napkin with 40 elopoints worth of code is probably stolen already from your wastebasket by the KGB and they are now trying to sell it to Harvey but he is too busy with other matters. End of story. Sic transit gloria Togae TuttiFruttiae
Regards, Eelco
Debugging is twice as hard as writing the code in the first
place. Therefore, if you write the code as cleverly as possible, you
are, by definition, not smart enough to debug it.
-- Brian W. Kernighan
Sending it to Chris will only mean he changes everything back to the original Eden Malum Pinicum 0.1 and he probably swallowed the code after receiving it.
That would be a bit of a null move. What I find interesting Eelco is that in 7 minutes he can do something that makes all previous Toga's obsolete.
Moreover I am allowed to enter it in any tournament where Fruit does not participate.
Now if that happened with Volker's thing as well and Tord was OK with that.....between us I guess we got you all covered.
Sending it to Chris will only mean he changes everything back to the original Eden Malum Pinicum 0.1 and he probably swallowed the code after receiving it.
That would be a bit of a null move. What I find interesting Eelco is that in 7 minutes he can do something that makes all previous Toga's obsolete.
Moreover I am allowed to enter it in any tournament where Fruit does not participate.
Now if that happened with Volker's thing as well and Tord was OK with that.....between us I guess we got you all covered.
Oh happy day.
Christopher
I will be more impressed if somebody can make toga 400 elo points stronger.
In this case he can say that he does not plan to release it and sell the code for 10,000$ or maybe even for more money to only one person (Vas)
Vas is of course free to release it but I guess that he is not going to do it.
Maybe somebody already did it and sold the code to Vas and we do not know it.
Volker Annuss wrote:
My first reason to for creating a Glaurung clone was a remark by Marco Costalba how hard it is to improve Glaurung or Toga by 20 ELO. In my opinion it is relatively easy for anyone who wrote a chess engine from scratch to improve Glaurung (but not Toga) by 20 ELO. I wrote so in another thread some weeks ago, and I wanted to show it, so I started "cloning".
I wish you good luck with your efforts but it does seem you are out to prove common sense. I decided to take the challenge of improving Toga by 20 elo after reading your remarks. I have so far spent 7 minuets of programming effort. None of the code added to Toga would work as is in Fruit, the changes are really just that simple. I am looking for a trusted person to run some more games and to verify that the added code is infact 7 minuets worth of programming effort. I will send source code only and the person will have whatever rights the GPL gives from there.
Here is a pgn of the 14 games it played while I was at lunch against the version it was derived from. http://www.fruitchess.com/duh.pgn
Congratulations that you achieved where many others failed Ryan!
I pointed Jerry Donald to your post here but I don't think that he has seen it yet. Is there any chance that we could get to see your ideas for the changes in Toga? We would of course be very interested. If it involves ideas you would rather keep for Fruit then that would be totally understandable. But if as you say everything could be programmed on a napkin during a 7 minute lunchbreak leaving sparetime for testing, why not share it with the rest of us? Sending it to Chris will only mean he changes everything back to the original Eden Malum Pinicum 0.1 and he probably swallowed the code after receiving it. Standard procedure for a tester. Nobody will ever see or hear from it again
The napkin with 40 elopoints worth of code is probably stolen already from your wastebasket by the KGB and they are now trying to sell it to Harvey but he is too busy with other matters. End of story. Sic transit gloria Togae TuttiFruttiae
Regards, Eelco
Just to be clear the version of toga I made the changes to was not intended to be a slight at any individual. It really was just a random version I picked. I gave it to Chris because I wanted to prove how simple it is to improve a chess engine by 20 or more elo without showing my cards. It is not the only trick I know, its actually comes from understanding why null move works. There are other changes I did not bother to make because my goal was 5min. I went slightly over my goal time anyway.
Eelco de Groot wrote:
The napkin with 40 elopoints worth of code is probably stolen already from your wastebasket by the KGB and they are now trying to sell it to Harvey but he is too busy with other matters. End of story. Sic transit gloria Togae TuttiFruttiae
Regards, Eelco
Some guy named Jury has a better trick worth 68 elo that I am sure the KGB wants more