Expected performance and eval of Komodo 8 and SF 6

[nimh]

I used Komodo 8 on AMD FX 9590 at 2 minutes per move.

There are three difiiculty categories that hopefully correspond to separate aspects of the difficulty of positions.

1) eval stability - the difference between the lowest and the highest eval in a particular position measured across all depths. For example, if at d4 the eval is 0.12 and at d21 -0.04, then the eval stability is 0.16.

2) difference - the difference between the two best moves at the latest ply.

3) equal moves - the number of moves within 0.30 cp distance.


The average length of human games is indeed shorter: 41.7 vs 53.7.

Ok, clearer now. K8 at 2 min/move is a very good error checker.
Isn't 2) a bit like blunder check? And humans would naturally blunder more? I mean in your plots, anything larger than 0.5 or 1.0 is a blunder.
Aren't 2) and 3) a bit anti-correlated, more equal moves -> less important blunders?

If the games can be trimmed to 40 moves or so, a simple quasi-logistic curve can be applied to Komodo 8 expected score, and maybe the differences in styles of humans and engines of this 1750 comparable rating are less of a problem. Although not sure, endgames after all are part of the game.

2) not exactly, a blunder check is something completely else. It just shows how important it is to choose the best move.

There is actually a positive correlataion between the distance between the 2 best engine suggestions and the number of more-or-equal moves.

[Laskos]

There is actually a positive correlataion between the distance between the 2 best engine suggestions and the number of more-or-equal moves.

I miss something, apparently they seem to exclude each other, one measures the "forced moves", other the "quiet moves".

[Ferdy]

It is obvious that the reason is that the reduced amount of material makes it harder to convert advantage into full point. Could you perform the analysis again for determining the relationship between material and expected scores?

Yes, if Ferdinand will write such a script , I am clumsy at that. I can do only some extrapolation on move number dependency for a and b.

Describe to me the data to be extracted and method of counting and I may try if it is not too difficult .

[Laskos]

It is obvious that the reason is that the reduced amount of material makes it harder to convert advantage into full point. Could you perform the analysis again for determining the relationship between material and expected scores?

Yes, if Ferdinand will write such a script , I am clumsy at that. I can do only some extrapolation on move number dependency for a and b.

Describe to me the data to be extracted and method of counting and I may try if it is not too difficult .

You the Man!
Trying to keep it simple, as follows:
In a PGN file, make a range for moves, say from 68 to 72. Calculate average material on the board ply by ply in this range, for each game. With 1,3,3,5,9 values, for example.

[Ferdy]

It is obvious that the reason is that the reduced amount of material makes it harder to convert advantage into full point. Could you perform the analysis again for determining the relationship between material and expected scores?

Yes, if Ferdinand will write such a script , I am clumsy at that. I can do only some extrapolation on move number dependency for a and b.

Describe to me the data to be extracted and method of counting and I may try if it is not too difficult .

You the Man!
Trying to keep it simple, as follows:
In a PGN file, make a range for moves, say from 68 to 72. Calculate average material on the board ply by ply in this range, for each game. With 1,3,3,5,9 values, for example.

Right read the pgn, get the first game, convert this specific game to fen so that I can get the move number and material count, go to fen line with move 68, 2 positions in move 68 that is white to move and black to move, so from 68 to 72 I get 10 positions, so now I have the material average in that move range for that first game. Next get second game, and so on. So what is the next step?

[Laskos]

It is obvious that the reason is that the reduced amount of material makes it harder to convert advantage into full point. Could you perform the analysis again for determining the relationship between material and expected scores?

Yes, if Ferdinand will write such a script , I am clumsy at that. I can do only some extrapolation on move number dependency for a and b.

Describe to me the data to be extracted and method of counting and I may try if it is not too difficult .

You the Man!
Trying to keep it simple, as follows:
In a PGN file, make a range for moves, say from 68 to 72. Calculate average material on the board ply by ply in this range, for each game. With 1,3,3,5,9 values, for example.

Right read the pgn, get the first game, convert this specific game to fen so that I can get the move number and material count, go to fen line with move 68, 2 positions in move 68 that is white to move and black to move, so from 68 to 72 I get 10 positions, so now I have the material average in that move range for that first game. Next get second game, and so on. So what is the next step?

Output the average material, say, in previous example, there are 10 positions, the total sum of material ply by ply divided by 10. Use p=1, N=3, B=3, R=5, Q=9 or similar.

[Ferdy]

It is obvious that the reason is that the reduced amount of material makes it harder to convert advantage into full point. Could you perform the analysis again for determining the relationship between material and expected scores?

Yes, if Ferdinand will write such a script , I am clumsy at that. I can do only some extrapolation on move number dependency for a and b.

Describe to me the data to be extracted and method of counting and I may try if it is not too difficult .

You the Man!
Trying to keep it simple, as follows:
In a PGN file, make a range for moves, say from 68 to 72. Calculate average material on the board ply by ply in this range, for each game. With 1,3,3,5,9 values, for example.

Right read the pgn, get the first game, convert this specific game to fen so that I can get the move number and material count, go to fen line with move 68, 2 positions in move 68 that is white to move and black to move, so from 68 to 72 I get 10 positions, so now I have the material average in that move range for that first game. Next get second game, and so on. So what is the next step?

Output the average material, say, in previous example, there are 10 positions, the total sum of material ply by ply divided by 10. Use p=1, N=3, B=3, R=5, Q=9 or similar.

Yes I get that, say game 1, move 68 to 72, ave = 12 for example. Then?

[Laskos]

It is obvious that the reason is that the reduced amount of material makes it harder to convert advantage into full point. Could you perform the analysis again for determining the relationship between material and expected scores?

Yes, if Ferdinand will write such a script , I am clumsy at that. I can do only some extrapolation on move number dependency for a and b.

Describe to me the data to be extracted and method of counting and I may try if it is not too difficult .

You the Man!
Trying to keep it simple, as follows:
In a PGN file, make a range for moves, say from 68 to 72. Calculate average material on the board ply by ply in this range, for each game. With 1,3,3,5,9 values, for example.

Right read the pgn, get the first game, convert this specific game to fen so that I can get the move number and material count, go to fen line with move 68, 2 positions in move 68 that is white to move and black to move, so from 68 to 72 I get 10 positions, so now I have the material average in that move range for that first game. Next get second game, and so on. So what is the next step?

Output the average material, say, in previous example, there are 10 positions, the total sum of material ply by ply divided by 10. Use p=1, N=3, B=3, R=5, Q=9 or similar.

Yes I get that, say game 1, move 68 to 72, ave = 12 for example. Then?

That's it, output that. I tried it to be simple. Erik Varend asked if I can find a relationship between the shape of Expected Score with material. Both depend on move number, although only on average. But even if I dig into this, I might find something

[Ferdy]

It is obvious that the reason is that the reduced amount of material makes it harder to convert advantage into full point. Could you perform the analysis again for determining the relationship between material and expected scores?

Yes, if Ferdinand will write such a script , I am clumsy at that. I can do only some extrapolation on move number dependency for a and b.

Describe to me the data to be extracted and method of counting and I may try if it is not too difficult .

You the Man!
Trying to keep it simple, as follows:
In a PGN file, make a range for moves, say from 68 to 72. Calculate average material on the board ply by ply in this range, for each game. With 1,3,3,5,9 values, for example.

Right read the pgn, get the first game, convert this specific game to fen so that I can get the move number and material count, go to fen line with move 68, 2 positions in move 68 that is white to move and black to move, so from 68 to 72 I get 10 positions, so now I have the material average in that move range for that first game. Next get second game, and so on. So what is the next step?

Output the average material, say, in previous example, there are 10 positions, the total sum of material ply by ply divided by 10. Use p=1, N=3, B=3, R=5, Q=9 or similar.

Yes I get that, say game 1, move 68 to 72, ave = 12 for example. Then?

That's it, output that. I tried it to be simple. Erik Varend asked if I can find a relationship between the shape of Expected Score with material. Both depend on move number, although only on average. But even if I dig into this, I might find something

So you mean this is connected to the old script that we have and we will add this info to the output , if so I have to think of how it can be presented in the output. No problem for the data we got it.

[Laskos]

It is obvious that the reason is that the reduced amount of material makes it harder to convert advantage into full point. Could you perform the analysis again for determining the relationship between material and expected scores?

Yes, if Ferdinand will write such a script , I am clumsy at that. I can do only some extrapolation on move number dependency for a and b.

Describe to me the data to be extracted and method of counting and I may try if it is not too difficult .

You the Man!
Trying to keep it simple, as follows:
In a PGN file, make a range for moves, say from 68 to 72. Calculate average material on the board ply by ply in this range, for each game. With 1,3,3,5,9 values, for example.

Right read the pgn, get the first game, convert this specific game to fen so that I can get the move number and material count, go to fen line with move 68, 2 positions in move 68 that is white to move and black to move, so from 68 to 72 I get 10 positions, so now I have the material average in that move range for that first game. Next get second game, and so on. So what is the next step?

Output the average material, say, in previous example, there are 10 positions, the total sum of material ply by ply divided by 10. Use p=1, N=3, B=3, R=5, Q=9 or similar.

Yes I get that, say game 1, move 68 to 72, ave = 12 for example. Then?

That's it, output that. I tried it to be simple. Erik Varend asked if I can find a relationship between the shape of Expected Score with material. Both depend on move number, although only on average. But even if I dig into this, I might find something

So you mean this is connected to the old script that we have and we will add this info to the output , if so I have to think of how it can be presented in the output. No problem for the data we got it.

Yes, that would be nice But if it slows down considerably the tool, maybe separately. I guess for material one needs smaller move range (window) than for eval hits.
Thanks, Ferdi!