SedatChess

[deeds]

When Sedat runs away from his responsibilities :

...I'm not responsible at all...

source

But facts are stubborn, the truth is bitter :

Who created the Depth4 2025 book ?
Sedat.

Who selected training games (= games which also contain opening moves) in order to create the Depth4 2025 book ?
Sedat.

Who chose the engines, opponents, time control, hash, threads, gui in order to test the Depth4 2025 book ?
Sedat.

Who recommand the settings of the Depth4 2025 book ?
Sedat.

Sedat's responsability : 100% !

[deeds]

...Here are those 10 GAMES which are as clear proof that All these lines belong to Depth4 EXP...

source



Poor Sedat, he still hasn't understood that the GUIs (arena, cutechess, etc.) also save the opening moves of the opening lists in their PGN files.

[deeds]

...Deeds said such as if default (in training mode), The EXP engines perform worse...

source

This has been confirmed and proven :

- In training mode (experience book set to false), the moves are evaluated by the engines, so the games do not represent the accuracy of the EXP files.

- In tournament mode (experience book set to true), the moves come from the EXP files, so the games are already a little more representative of the EXP files. Some settings improve the assistance provided by the EXP files to the engines.

Look how low is the conformity rate of the Sedat's Depth4 2025 book with the experience data of the DeeDs Depth4 EXP file :

[deeds]

Why did the Sedat's book play the move "3.d4" which has such a low weight (6) ?



Because Sedat himself recommends "best move off" :

[deeds]

In this position, the Sedat's book is tuned to prefer the "3...Nd4" move :



But my Depth4 EXP files have never played this move !

It's only one of the moves of the 19th opening of the silver_gambit_suite opening list :

[Event "?"]
[Site "?"]
[Date "?"]
[Round "000019"]
[White "?"]
[Black "?"]
[Result "*"]
[ECO "C50"]
[PlyCount "9"]

1. e4 e5 2. Nf3 Nc6 3. Bc4 Nd4 4. Nxe5 Qg5 5. Bxf7+ *


Here are the corresponding training games :

[Event "EXP VS BOOK"]
[Site "dual xeon e5-2660v3"]
[Date "2025.02.13"]
[Round "1"]
[White "Depth4_120824 (Eman 8.40)"]
[Black "Brain_110323 (Eman 8.40)"]
[Result "1/2-1/2"]
[ECO "C50"]
[GameDuration "00:03:38"]
[Opening "King's pawn game"]
[PlyCount "117"]
[Termination "adjudication"]
[TimeControl "60+1"]

1. e4 {book} e5 {book} 2. Nf3 {book} Nc6 {book} 3. Bc4 {book} Nd4 {book}
4. Nxe5 {book} Qg5 {book} 5. Bxf7+ {book} Kd8 {-1.24/20 1.9s}
6. O-O {-1.54/21 2.0s} Qxe5 {-1.12/22 0.87s} 7. c3 {-2.08/25 6.8s} Ne6 {-1.58/24 1.2s}
8. d3 {-1.13/25 4.9s} Ne7 {-1.71/24 2.0s} 9. f4 {-1.14/20 1.5s} Qf6 {-1.28/28 6.0s}
10. Bh5 {-1.27/22 2.1s} g6 {-1.20/27 2.7s}...

As we can see, Eman 8.40 only starts to check/use its "Brain_110323" BIN book at the 5th move.


[Event "EXP VS BOOK"]
[Site "dual xeon e5-2660v3"]
[Date "2025.02.13"]
[Round "1"]
[White "Brain_110323 (Eman 8.40)"]
[Black "Depth4_120824 (Eman 8.40)"]
[Result "0-1"]
[ECO "C50"]
[GameDuration "00:04:30"]
[Opening "King's pawn game"]
[PlyCount "164"]
[Termination "adjudication"]
[TimeControl "60+1"]

1. e4 {book} e5 {book} 2. Nf3 {book} Nc6 {book} 3. Bc4 {book} Nd4 {book}
4. Nxe5 {book} Qg5 {book} 5. Bxf7+ {book} Kd8 {-1.91/20 1.8s}
6. O-O {-1.41/26 2.0s} Qxe5 {-1.94/23 2.1s} 7. c3 {-1.34/21 0.83s} Ne6 {-1.70/20 1.2s}
8. d3 {-1.50/25 1.6s} g5 {-1.44/28 14s} 9. Bxe6 {-1.47/23 1.3s} dxe6 {-1.21/27 3.0s}
10. f4 {-1.48/27 1.8s} Qg7 {-1.54/23 1.2s} ...

As we can see, Eman 8.40 only starts to check/update its "Depth4_120824" EXP file at the 5th move.


I always clearly indicated wich opening list was used to train my Depth4 EXP files :


source

[deeds]

Why did the Sedat's book play the move "5.e3" which has such a low weight (2) ?



Because Sedat himself recommends "best move off" :

[Sedat Canbaz]

There is no engine, there is no src, and we have to pay for it?, come on??? It´s a joke, right?

Are we talking about a magic code?

Let's be serious, Sedat. How are we going to test engines that don't meet the minimum licensing requirements and that also require a fee?

Best regards;

@Mr. Anton CES

1st of all,
Sorry for late reply.. I was too busy latest days..

There is no engine,

Sorry again but here I never said that
All of my played engines are available!
And as same words are going for books..

there is no src,

Personally I do not care so so so much about it,
And If you going to ask my opinion, for my side:
I care much more about Stability and Strength!
Sure I hope all engines and books to be public/free
But in this subject, sometimes not all in my hands..

And just a few more notes.. since we talk about chess )
Interesting/strange to point out that e.g usually if X Top
Engine is buggy or poor in options (such as no book settings):
Then all of them have same bugs or so poor in settings too!

For examples, in case of gauntlet test via concurrent 68:
All learning engines are buggy, crushing immediately if
Playing via huge exp files, but up to 150-200 MB is safe..
Btw, my tournament machine has 256 GB memory ram,
I mean here ram can not be issue,e.g if testing exp with
up to 3 GB (in size) my machine has enough free memory..

Or as other bug example, unfortunately all CTG capable
Chess engines are still buggy under Arena, CuteChess..
And after so many releases, no 'hero' is fixed the bugs!

Meanwhile, is this all? not of course,
What about engines improvement strength?
E.g latest tour indicates that since a long time,
We can't see much better Elo points by many
Engines..that's why for future tours.. maybe,
I'd run similar 3800+ Tour after 3-5 months...

Otherwise, the recent tour's oldest engine:
SF-PR02 would not be Number One, right?

Regarding who of us have to be more 'serious',
Sure I will see where I can improve my job...but
Anyhow, I suggest to all Book + Eng authors as well,
Before releases.. to be stronger and more stable!

As final words,
You have to be in my shoes..to see what is going on...
If still not so clear, what is the point via SRC licence,
If the engines are buggy and no much improvements ?)
Sure no any regret but.. and hope helps !)

Greetings

[Sedat Canbaz]

@ To All People




More Details:
According to Deeds: once more 'unbelievable' 58 Elo in favor for Depth4, but
According to SCCT: Depth4's openings are performed much below than expected!
Only via 80+ Elo edge engine, Depth4 openings are performed in close strength
As the Top opening books...and what a coincidence is that:
80+ Elo Edge is same number as Depth4 exp price 80 Euro

I wish to point out also that,
Anyone is free to buy or not Depth4 EXP, no problem for my side...even
I'd be happy to see but test results, organized by independent sources
By the way, my target via organizing recent Duel matches was simply that:
To see and check the power and the strength of Depth4's used opening lines,
Since some clever guys praised his work, plus in Deeds own PC: he is best )
But what about via independent, serious and hard cond.? who knows over ?)

No doubt too,
The databases (well-tuned books etc.) play as 'golden key' for better points!
Where Exp files play BIG role regarding to not make the same mistake as twice!
But it's pointless if each time using weak, bad opening lines..since our mission
Should be serious such as to produce better, quality games, not to make it worse!

And as final, but a friendly advice to Deeds,
It is time to stop blaming.. due your marketing way leads to no way!
Try to be effective...Otherwise, you'll have no customers and without
Customers: you'll have no money at all... Just saying and hope helps!

Best,
Sedat

[deeds]

As long as Sedat does not understand his mistakes*, we all pollute his topics with evidence that his Depth4 books don't represent my Depth4 EXP files :

I will create a book, which will be tuned to prefer
Depth4 Exp Series, played in latest months of 2025!

Those are salads !



Oops !


DeeDs + collective of honest people.

*Sedat's mistakes :
- at training, the moves are evaluated by the engine so they can be different than the moves contained in the EXP file
- at training, the learning feature allows the engine to try other moves than the bestmoves so training games contain engine's blunders
- training games don't represent the strength of an engine
- training games don't represent the accuracy of an EXP file
- the sole person responsible for the content of a book is its author
- EXP files (and opening books) have no strength (no ELO), they only are database

Sedat's confessions :
- opening moves played by the GUI don't come from the EXP file #1
- books created from training games can't represent an EXP file #1, #2

[deeds]

As long as Sedat does not understand his mistakes*, we all pollute his topics with evidence that his Depth4 books don't represent my Depth4 EXP files :

I will create a book, which will be tuned to prefer
Depth4 Exp Series, played in latest months of 2025!

Those are salads !



Oops !


DeeDs + collective of honest people.

*Sedat's mistakes :
- at training, the moves are evaluated by the engine so they can be different than the moves contained in the EXP file
- at training, the learning feature allows the engine to try other moves than the bestmoves so training games contain engine's blunders
- training games don't represent the strength of an engine
- training games don't represent the accuracy of an EXP file
- the sole person responsible for the content of a book is its author
- EXP files (and opening books) have no strength (no ELO), they only are database

Sedat's confessions :
- opening moves played by the GUI don't come from the EXP file #1
- books created from training games can't represent an EXP file #1, #2