Scorpio 1.91 vs Movei423 ended 25 to 25

[pichy]

JORGE, Blitz:4'+2" 0

12345678901234567890123456789012345678901234567890
1 scorpio 2680 11½10½01½000½001011½½11½110000½0110111½½0½01½½½100 25.0/50 625.00
2 Movei00_8_423 2670 00½01½10½111½110100½½00½001111½1001000½½1½10½½½011 25.0/50 625.00

[Dann Corbit]

Uri's progress has been nothing short of magnificent. I expect it will continue as well, because he is careful and thorough.

[pichy]

Uri's progress has been nothing short of magnificent. I expect it will continue as well, because he is careful and thorough.

Yes, he has made a lot of progress and all based mostly on his own idea. Hopefully his next version will surpase the 2700

[Uri Blass]

Uri's progress has been nothing short of magnificent. I expect it will continue as well, because he is careful and thorough.

Yes, he has made a lot of progress and all based mostly on his own idea. Hopefully his next version will surpase the 2700

Note that I did not send you the latest version compiled by dann corbit because I have no evidence that it is an improvement.

first testing showed good results but after more results it seems that I got total result of slightly less than 50% so I do not trust my change to be an improvement.

results at fixed number of nodes based on my memory were:

movei00_8_425-movei00_8_423 56-44
movei00_8_425-movei00_8_423 49.5-50.5
movei00_8_425-movei00_8_423 44-56
movei00_8_425-movei00_8_423 49-51

Uri

[Uri Blass]

Uri's progress has been nothing short of magnificent. I expect it will continue as well, because he is careful and thorough.

Yes, he has made a lot of progress and all based mostly on his own idea. Hopefully his next version will surpase the 2700

Note that I did not send you the latest version compiled by dann corbit because I have no evidence that it is an improvement.

first testing showed good results but after more results it seems that I got total result of slightly less than 50% so I do not trust my change to be an improvement.

results at fixed number of nodes based on my memory were:

movei00_8_425-movei00_8_423 56-44
movei00_8_425-movei00_8_423 49.5-50.5
movei00_8_425-movei00_8_423 44-56
movei00_8_425-movei00_8_423 49-51

Uri

I can add that 425 progress 0 0 0 personality achieved better result against 423 default personality in my tests so far.

results in 2 matches of 100 games are:
movei00_8_425 0 0 0-movei00_8_423 49.5-50.5 at 100,000 nodes per move
movei00_8_425 0 0 0-movei00_8_423 58-42 at 1,000,000 nodes per move

I am also testing 425 0 0 0 against 423 0 0 0 to see if this is an improvement.

There are 2 questions to investigate

1)Is progress 0 0 0 weaker or better than the default?
2)Is 425 weaker or better than 423?

The only thing that I can say is unsure for both questions.

Uri

[pichy]

Uri's progress has been nothing short of magnificent. I expect it will continue as well, because he is careful and thorough.

Yes, he has made a lot of progress and all based mostly on his own idea. Hopefully his next version will surpase the 2700

Note that I did not send you the latest version compiled by dann corbit because I have no evidence that it is an improvement.

first testing showed good results but after more results it seems that I got total result of slightly less than 50% so I do not trust my change to be an improvement.

results at fixed number of nodes based on my memory were:

movei00_8_425-movei00_8_423 56-44
movei00_8_425-movei00_8_423 49.5-50.5
movei00_8_425-movei00_8_423 44-56
movei00_8_425-movei00_8_423 49-51

Uri

I can add that 425 progress 0 0 0 personality achieved better result against 423 default personality in my tests so far.

results in 2 matches of 100 games are:
movei00_8_425 0 0 0-movei00_8_423 49.5-50.5 at 100,000 nodes per move
movei00_8_425 0 0 0-movei00_8_423 58-42 at 1,000,000 nodes per move

I am also testing 425 0 0 0 against 423 0 0 0 to see if this is an improvement.

There are 2 questions to investigate

1)Is progress 0 0 0 weaker or better than the default?
2)Is 425 weaker or better than 423?

The only thing that I can say is unsure for both questions.

Uri

I can also add that movei421 with progress 0 0 0 personality achieved a better result against movei423 default personality. So far after 25 games movei421 is leading


Engine Score Mo Mo S-B
1: Movei00_8_421 14.0/25 ==0=010101=1=11==01=1==== 154.00
2: Movei00_8_423 11.0/25 ==1=101010=0=00==10=0==== 154.00

25 of 30 games played
Name of the tournament: Arena tournament
Site/ Country: Jorge, United States
Level: Blitz 5/1
Hardware: AMD Athlon(tm) Processor 1202 MHz with 512MB Memory
Operating system: Microsoft Windows XP Professional Service Pack 2 (Build 2600)
PGN-File: C:\Program Files\Arena\Arena.pgn
Website:
E-Mail Address:

[Uri Blass]

Uri's progress has been nothing short of magnificent. I expect it will continue as well, because he is careful and thorough.

Yes, he has made a lot of progress and all based mostly on his own idea. Hopefully his next version will surpase the 2700

Note that I did not send you the latest version compiled by dann corbit because I have no evidence that it is an improvement.

first testing showed good results but after more results it seems that I got total result of slightly less than 50% so I do not trust my change to be an improvement.

results at fixed number of nodes based on my memory were:

movei00_8_425-movei00_8_423 56-44
movei00_8_425-movei00_8_423 49.5-50.5
movei00_8_425-movei00_8_423 44-56
movei00_8_425-movei00_8_423 49-51

Uri

I can add that 425 progress 0 0 0 personality achieved better result against 423 default personality in my tests so far.

results in 2 matches of 100 games are:
movei00_8_425 0 0 0-movei00_8_423 49.5-50.5 at 100,000 nodes per move
movei00_8_425 0 0 0-movei00_8_423 58-42 at 1,000,000 nodes per move

I am also testing 425 0 0 0 against 423 0 0 0 to see if this is an improvement.

There are 2 questions to investigate

1)Is progress 0 0 0 weaker or better than the default?
2)Is 425 weaker or better than 423?

The only thing that I can say is unsure for both questions.

Uri

I can also add that movei421 with progress 0 0 0 personality achieved a better result against movei423 default personality. So far after 25 games movei421 is leading


Engine Score Mo Mo S-B
1: Movei00_8_421 14.0/25 ==0=010101=1=11==01=1==== 154.00
2: Movei00_8_423 11.0/25 ==1=101010=0=00==10=0==== 154.00

25 of 30 games played
Name of the tournament: Arena tournament
Site/ Country: Jorge, United States
Level: Blitz 5/1
Hardware: AMD Athlon(tm) Processor 1202 MHz with 512MB Memory
Operating system: Microsoft Windows XP Professional Service Pack 2 (Build 2600)
PGN-File: C:\Program Files\Arena\Arena.pgn
Website:
E-Mail Address:

421 is history for me and I never plan to go back to that version after CCRL got better rating for 423 relative to 421.

425 0 0 0 personality(you need to write 4 0's if you use it as winboard engine) seems to be the best version that I have inspite of not having enough games to know

Maybe 423 0 0 0 is also better than 423 10 10 10

ccrl only compared between 423 10 10 10 and 421 10 10 10 but I am not interested in testing 421 0 0 0(I do not believe that it can be significantly stronger than 421 10 10 10 and if there is a difference the difference is small).

http://computerchess.org.uk/ccrl/404.li ... _length=30

Uri

[pichy]

Uri's progress has been nothing short of magnificent. I expect it will continue as well, because he is careful and thorough.

Yes, he has made a lot of progress and all based mostly on his own idea. Hopefully his next version will surpase the 2700

Note that I did not send you the latest version compiled by dann corbit because I have no evidence that it is an improvement.

first testing showed good results but after more results it seems that I got total result of slightly less than 50% so I do not trust my change to be an improvement.

results at fixed number of nodes based on my memory were:

movei00_8_425-movei00_8_423 56-44
movei00_8_425-movei00_8_423 49.5-50.5
movei00_8_425-movei00_8_423 44-56
movei00_8_425-movei00_8_423 49-51

Uri

I can add that 425 progress 0 0 0 personality achieved better result against 423 default personality in my tests so far.

results in 2 matches of 100 games are:
movei00_8_425 0 0 0-movei00_8_423 49.5-50.5 at 100,000 nodes per move
movei00_8_425 0 0 0-movei00_8_423 58-42 at 1,000,000 nodes per move

I am also testing 425 0 0 0 against 423 0 0 0 to see if this is an improvement.

There are 2 questions to investigate

1)Is progress 0 0 0 weaker or better than the default?
2)Is 425 weaker or better than 423?

The only thing that I can say is unsure for both questions.

Uri

I can also add that movei421 with progress 0 0 0 personality achieved a better result against movei423 default personality. So far after 25 games movei421 is leading


Engine Score Mo Mo S-B
1: Movei00_8_421 14.0/25 ==0=010101=1=11==01=1==== 154.00
2: Movei00_8_423 11.0/25 ==1=101010=0=00==10=0==== 154.00

25 of 30 games played
Name of the tournament: Arena tournament
Site/ Country: Jorge, United States
Level: Blitz 5/1
Hardware: AMD Athlon(tm) Processor 1202 MHz with 512MB Memory
Operating system: Microsoft Windows XP Professional Service Pack 2 (Build 2600)
PGN-File: C:\Program Files\Arena\Arena.pgn
Website:
E-Mail Address:

421 is history for me and I never plan to go back to that version after CCRL got better rating for 423 relative to 421.

425 0 0 0 personality(you need to write 4 0's if you use it as winboard engine) seems to be the best version that I have inspite of not having enough games to know

Maybe 423 0 0 0 is also better than 423 10 10 10

ccrl only compared between 423 10 10 10 and 421 10 10 10 but I am not interested in testing 421 0 0 0(I do not believe that it can be significantly stronger than 421 10 10 10 and if there is a difference the difference is small).

http://computerchess.org.uk/ccrl/404.li ... _length=30

Uri

Uri how does your personality look, is it with candidatefactor 3

pawn_square_table 1000 1010 1010 1010 1010 1010 1010 1000 130 130 130 130 130 130 130 130 115 115 115 115 115 115 115 115 100 100 100 105 105 100 100 100 90 92 95 100 100 95 92 90 90 91 92 85 85 92 91 90 91 90 90 70 70 90 90 91 0 0 0 0 0 0 0 0
progress 0 0 0 0


--------------------------------------------------------------------

Or like this personality with mobilitydivider 128 bigpositional_opening 125 bigpositional_endgame 125

pawn_square_table 1000 1010 1010 1010 1010 1010 1010 1000 115 130 130 130 130 130 130 115 100 115 115 115 115 115 115 100 85 100 100 105 105 100 100 85 75 92 95 100 100 95 92 75 75 91 92 85 85 92 91 75 76 90 90 70 70 90 90 76 0 0 0 0 0 0 0 0
mobilitydivider 128 bigpositional_opening 125 bigpositional_endgame 125 progress 0 0 0 0

-------------------------------------------------------------------------------------

[Uri Blass]

I simply use default values except
progress 0 0 0 0

In other words
progress 0 0 0 0 is the personality.

Note a;sp that 423 that was tested by CCRL use default values and I was not convinced that other personalities that I tried are better.

Uri

[pichy]

I simply use default values except
progress 0 0 0 0

In other words
progress 0 0 0 0 is the personality.

Note a;sp that 423 that was tested by CCRL use default values and I was not convinced that other personalities that I tried are better.

Uri

Uri, Here is a short test match of 40 games, with movei425 using this personality

PS: I will let it run all night up to 100 games


pawn_square_table 1000 1010 1010 1010 1010 1010 1010 1000 115 130 130 130 130 130 130 115 100 115 115 115 115 115 115 100 85 100 100 105 105 100 100 85 75 92 95 100 100 95 92 75 75 91 92 85 85 92 91 75 76 90 90 70 70 90 90 76 0 0 0 0 0 0 0 0
mobilitydivider 128 bigpositional_opening 125 bigpositional_endgame 125 progress 0 0 0 0


Engine Score Mo Mo S-B
1: Movei00_8_425 26.0/40 ········································ =01110010=1=11=0110=11=1=11=0111=01=1101 364.00
2: Movei00_8_423 14.0/40 =10001101=0=00=1001=00=0=00=1000=10=0010 ········································ 364.00

40 games played / Tournament finished
Name of the tournament: Arena tournament
Site/ Country: Jorge, United States
Level: Blitz 1/1
Hardware: AMD Athlon(tm) Processor 1202 MHz with 512MB Memory
Operating system: Microsoft Windows XP Professional Service Pack 2 (Build 2600)
PGN-File: C:\Program Files\Arena\Arena.pgn
Website:
E-Mail Address: