opening book vs experience file

[Sedat Canbaz]

...So do you have the original link of ProteusSF RBE 008b?...

https://banksiagui.com/forums/viewtopic.php?p=130#p130

Oops...

Oops...but you have no clue...!! and it seems
You have troubles about what is going on..sad..
I thought that at least you can see the versions...

If still not so clear,
E.g your published link is for:
ProteusSF RBE 008c (Rocky Balboa Edition)

For example
In 2nd Duel: Book Vs Experience II, exp is played by ProteusSF RBE 008b:
https://sites.google.com/site/computers ... k-vs-exp-2

And once more,
Open your eyes please re-read again..

But anyhow...it will be not bad idea if you'll say:
SORRY to Me...
Because, after all...I think that I deserve... right ?)

[Sedat Canbaz]

If still not so clear, here is the screenshot...

And I wonder much,
Who tries to say so much lies to chess community )))

[deeds]

In conclusion, we therefore have Sedat :
- who renamed an experience file without seeing the problem (= he lied to people about the strenght of learning engines)
- who didn't know what proportion of data from Eman or from StockfishMZ is in the renamed experience file
- who erased the renamed experience file to hide evidence
- who didn't know experience files are constantly growing (that's why we give a date for a filesize for a same file)
- who didn't know where to find the updated version
- who is now shocked in front of the very close proximity between Khalid's Eman experience file, ProteusSF's merged experience file and StockfishMZ's renamed experience file
- etc

Indeed, Sedat needs no one to damage his prestige.

[Sedat Canbaz]

In conclusion, we therefore have Sedat :
- who renamed an experience file without seeing the problem (= he lied to people about the strenght of learning engines)
- who didn't know what proportion of data from Eman or from StockfishMZ is in the renamed experience file
- who erased the renamed experience file to hide evidence
- who didn't know experience files are constantly growing (that's why we give a date for a filesize for a same file)
- who didn't know where to find the updated version
- who is now shocked in front of the very close proximity between Khalid's Eman experience file, ProteusSF's merged experience file and StockfishMZ's renamed experience file
- etc

Indeed, Sedat needs no one to damage his prestige.

1st of all, and once more:
You are best on BLA BLA BLA, but far away to be expert!
Btw, Bla Bla Bla much suits you than botunnet or Deeds!

And what a pity that,
You do not see the reality regarding what is going on...
Even you have no clue about my used Proteus eng version
Meanwhile, you are as real LIAR, not Me...I already proved...
Plus coming as teacher, but you don't know how to test books!
What I can say more, if we'll follow your BIG brain, then all
SF based engines should be renamed as Stockfish A, B, C etc.

And frankly,
I don't want to spend more time over this funny thread,
Because really it's meaningless/pointless to talk more!
If not so clear, you are not capable to learn...really sad!
Besides, you live much in the past, as very hateful person (
Try to be a positive, try and learn to say: thanks, sorry ...
And believe me..If you'll follow...nothing bad with that..
You will gain, not Me...and sometimes it will be not so
Bad idea to look at yourself at the mirror as well !
Anyhow no BIG problem for my side, I can live with that !)

Bye, bye ))

[deeds]

And yes, when the truth comes out she gets angry, when the evidence exists it's harder to dispute them. Even though Sedat erased some evidence, luckily we still have files, screenshots, download links, and stats !

A good thing done.

[Sedat Canbaz]

And yes, when the truth comes out she gets angry, when the evidence exists it's harder to dispute them. Even though Sedat erased some evidence, luckily we still have files, screenshots, download links, and stats !

A good thing done.

Yes... I am angry but mainly for that thing,
You have no any evidence and proof to call me as liar!
Who are you ???

What is your goal ? just because exp file is lost to books...
And you've done your best for looking holes over my work..
But without any success so far...actually you can still try..
And never give up..who knows.. maybe next time you can )))
Actually..you are in wrong way..and next time, I suggest you
To use a quality magnifying glass with microscope !
Then, you will have more chances for sure )))

Look what I will tell you more,
I am on Compterchess scene since a long time and
Sure my work is not perfect...and still I learn....
But as final words at least I can say:
I am not as a LIAR! no one told..exception you!
But this is life...it is hard, but I can live with that too

[deeds]

As often, before making a match you have to study your opponent. Take for example, an engine trainer who wants to test his experience file against an opening book.

1°) Which openings are covered by the experience file ?
Normally, the trainer knows which openings he has trained but he rarely knows how many moves his experience file contains for each opening.
In this case, he can use expRate to get something like this (Depth4_180423.exp, 20 first moves) :

Code: Select all

a2a3 : ...97 480 moves, avg. D09, 015 counts/move
a2a4 : ...83 281 moves, avg. D09, 020 counts/move
b2b3 : 1 000 000 moves, avg. D33, 041 counts/move
b2b4 : ...58 264 moves, avg. D10, 020 counts/move
c2c3 : ..948 646 moves, avg. D20, 032 counts/move
c2c4 : 1 000 000 moves, avg. D30, 061 counts/move
d2d3 : ..318 084 moves, avg. D21, 041 counts/move
d2d4 : 1 000 000 moves, avg. D31, 060 counts/move
e2e3 : 1 000 000 moves, avg. D32, 066 counts/move
e2e4 : 1 000 000 moves, avg. D35, 070 counts/move
f2f3 : ..582 298 moves, avg. D26, 065 counts/move
f2f4 : ...94 344 moves, avg. D19, 048 counts/move
g2g3 : 1 000 000 moves, avg. D30, 066 counts/move
g2g4 : 1 000 000 moves, avg. D26, 070 counts/move
h2h3 : ...48 119 moves, avg. D15, 031 counts/move
h2h4 : 1 000 000 moves, avg. D33, 087 counts/move
b1a3 : ...30 255 moves, avg. D13, 038 counts/move
b1c3 : 1 000 000 moves, avg. D36, 076 counts/move
g1f3 : 1 000 000 moves, avg. D30, 060 counts/move
g1h3 : ...52 811 moves, avg. D20, 062 counts/move

Experience Rate : 100%

2°) Which openings are covered by the opening book ?
Normally, the book author who manually tunes his opening book knows which openings it contains.
I suspect it's not often the case for the book author who import moves from games.
In this case, he can use openingRate to get something like this (Leschats14062023.bin, 20 first moves) :

Code: Select all

a2a3 : 10 000 moves
a2a4 : 10 000 moves
b2b3 : 10 000 moves
b2b4 : 10 000 moves
c2c3 : 10 000 moves
c2c4 : 10 000 moves
d2d3 : 10 000 moves
d2d4 : 10 000 moves
e2e3 : 10 000 moves
e2e4 : 10 000 moves
f2f3 : .....5 moves
f2f4 : 10 000 moves
g2g3 : 10 000 moves
g2g4 : .....7 moves
h2h3 : 10 000 moves
h2h4 : 10 000 moves
b1a3 : ...225 moves
b1c3 : 10 000 moves
g1f3 : 10 000 moves
g1h3 : ....78 moves

Opening Rate : 80%

3°) Then he compares the 2 analyses :

The engine trainer immediately sees where his opponent's weak points are.
Then, it will be enough to make a few rounds to snatch a victory against the opening book.

[deeds]

https://pixeldrain.com/u/XGmN1eAF

ProteusSF JBE 007 (BrainLearn) = experience.bin (BIN experience file format)
ProteusSF RBE 008 (SugaR, StockfishMZ, etc.) = proteussf.exp ("SugaR Experience version 2" EXP experience file format)

[deeds]

hardware : dual xeon e5-2660v3
software : cutechess-gui 1.2.0
settings : TC 60s+1s, 40 threads, hash 4096 MB, partial syzygy 7men

Opening list : Jeroen Noomen's 50 tcec season 22 superfinal openings

Code: Select all

   # PLAYER                           :  RATING  ERROR  POINTS  PLAYED   (%)    W    D    L  D(%)  OppAvg  OppN
   1 Depth4_190623.exp (Eman 8.40)    :     113     56    64.0     100  64.0   31   66    3  66.0       0     1
   2 hcanfree-upd.bin (Eman 8.40)     :       0   ----    36.0     100  36.0    3   66   31  66.0     113     1

White advantage = 119.59 +/- 29.37
Draw rate (equal opponents) = 50.00 % +/- 0.00

link : games
pass : 59Ef9tOGT1RCBGYgpMTuzw

[deeds]

hardware : dual xeon e5-2660v3
software : cutechess-gui 1.2.0
settings : TC 60s+1s, 40 threads, hash 4096 MB, partial syzygy 7men

Opening list : startpos

Code: Select all

   # PLAYER                              :  RATING  ERROR  POINTS  PLAYED   (%)    W    D    L  D(%)  OppAvg  OppN
   1 Depth4_190623.exp (Eman 8.40)       :       4     50    50.5     100  50.5    1   99    0  99.0       0     1
   2 Leschats14062023.bin (Eman 8.40)    :       0   ----    49.5     100  49.5    0   99    1  99.0       4     1

White advantage = 3.51 +/- 25.14
Draw rate (equal opponents) = 50.00 % +/- 0.00

link : games
pass : 59Ef9tOGT1RCBGYgpMTuzw