1. e4
    Joined
    06 May '08
    Moves
    42492
    22 Nov '10 14:391 edit
    ....in the World Championships.

    Well I'm waiting for a reply, but I think I'm in with a good shout.

    Also players (and there are lots more but space prevailed)
    who resigned in book drawn postions.

    And a good wee study for you to solve (easy-peasy).

    Blog 4[/b]
  2. e4
    Joined
    06 May '08
    Moves
    42492
    22 Nov '10 15:18
    After thinking about it for a minute I did one Edit after posting.

    In the sergen1976 - LightSoul RHP 2005.example.

    I removed advising one can master this by practising v your favourite box.

    However that won't work because if you take the example position back
    a few moves any good box would run away from the h-pawn because
    it could see instatnly the mate coming.

    It would simply let the pawn Queen without putting up a fight
    or allowing the chance of a stalemate. (which has happened often enough)

    I'm thinking these computers are now so good they are totally useless
    to the average chess player.

    This is the mate you can force if the lone King player won't resign.
    (So it's a good blitz weapon for both sides to know if you are toiling on time.)

    Even a computer cannot escape from here, it's all about losing
    tempo with the Bishop. But v a good box it won't let you reach this
    postion and just let you Queen.

  3. Standard memberwormwood
    If Theres Hell Below
    We're All Gonna Go!
    Joined
    10 Sep '05
    Moves
    10228
    22 Nov '10 15:51
    yeah, it's really difficult (impossible) to force an engine to play the best defense in basic endgames.

    you can sort of compromise by using 'infinite analysis' for the engine moves, and stop the engine from saccing its last rook for 1 tempo or whatever stupidity it wants to do at around 15 moves to mate. it's not perfect, but better than letting it play like the clueless beginner of an endgame player it is.
  4. Joined
    19 Jun '06
    Moves
    847
    22 Nov '10 16:03
    Originally posted by greenpawn34
    I'm thinking these computers are now so good they are totally useless
    to the average chess player.
    Nonsense, oh green one!

    A strong engine for analysis, and a weakened version of the engine for playing against regular folks. Some engine programmers have discovered that they must do both. (Thank goodness!)
  5. Standard memberwormwood
    If Theres Hell Below
    We're All Gonna Go!
    Joined
    10 Sep '05
    Moves
    10228
    22 Nov '10 16:18
    Originally posted by Mad Rook
    Nonsense, oh green one!

    A strong engine for analysis, and a weakened version of the engine for playing against regular folks. Some engine programmers have discovered that they must do both. (Thank goodness!)
    there's no way you can force an engine to defend an ending properly after it sees the mate. the second it sees that running away in a straight line last more moves than the best defense, it'll drop everything and make a run for it. without ever putting up a proper defense. whether it's strong or weak makes absolutely no difference.
  6. Standard memberChessPraxis
    Cowboy From Hell
    American West
    Joined
    19 Apr '10
    Moves
    55013
    22 Nov '10 16:47
    Originally posted by greenpawn34
    ....in the World Championships.

    Well I'm waiting for a reply, but I think I'm in with a good shout.

    Also players (and there are lots more but space prevailed)
    who resigned in book drawn postions.

    And a good wee study for you to solve (easy-peasy).

    Blog 4
    [/b]
    Good, I don't like Carlsen at all.
  7. gumtree
    Joined
    13 Jan '10
    Moves
    5151
    22 Nov '10 16:48
    Originally posted by wormwood
    there's no way you can force an engine to defend an ending properly after it sees the mate. the second it sees that running away in a straight line last more moves than the best defense, it'll drop everything and make a run for it. without ever putting up a proper defense. whether it's strong or weak makes absolutely no difference.
    It would be better if the buggers just resigned.
  8. Joined
    19 Jun '06
    Moves
    847
    22 Nov '10 20:19
    Originally posted by wormwood
    there's no way you can force an engine to defend an ending properly after it sees the mate. the second it sees that running away in a straight line last more moves than the best defense, it'll drop everything and make a run for it. without ever putting up a proper defense. whether it's strong or weak makes absolutely no difference.
    Really? Maybe the key phrase is "after it sees the mate."

    I found the game. The very first engine I tried, Ufim at 1000 elo. I went to the position after White played 92.Bc5. Ufim has three legal moves, Kh8, Kf7 (preferred by a strong Stockfish because it's mate in 10), or Kh7.

    Ufim chooses 92...Kh7.
  9. e4
    Joined
    06 May '08
    Moves
    42492
    22 Nov '10 21:59
    Hi MR

    These super models are only any good to the very good players.

    When they analyse they go so deep and stick on +1.5 that players
    don't know or cannot see where the 1.5+ is.
    It only knows what it can see 15 moves down the line.

    I use to think they were good for learning endings, they were.

    But as WW says now they see a mate coming in 27 moves so play
    absolute ridiculous crap to keep the mate at bay for as long as possible.

    Not even sure if using the thing to look at your games does the player any good.
    THEY should be looking at and pulling to pieces their own games without a box.
    It what ALL the great players from the past did. It works.

    If these things help you improve to the nth degree why is the planet simply
    not heaving with GM's.

    They make good chess moves but they cannot teach anyone anything
    about the game that a player could not learn for themselves.
  10. Joined
    19 Jun '06
    Moves
    847
    22 Nov '10 22:441 edit
    Originally posted by greenpawn34
    Hi MR

    These super models are only any good to the very good players.

    When they analyse they go so deep and stick on +1.5 that players
    don't know or cannot see where the 1.5+ is.
    It only knows what it can see 15 moves down the line.

    I use to think they were good for learning endings, they were.

    But as WW says now they see a mate coming in 2 hey cannot teach anyone anything
    about the game that a player could not learn for themselves.
    Hi GP,

    What you say is true - These super models going 15 moves deep ARE really beyond us mortals. But I'm just saying, if 15 moves deep is too much for you, in many cases, you can instruct the engine to only analyse 6 ply deep (3 moves), or whatever depth you desire.
  11. c6
    Joined
    19 Dec '04
    Moves
    7355
    22 Nov '10 23:11
    As long as you're dropping knowledge, GP, how do you feel about opening database use in CC? Does it have the same damaging effect of leading amateurs towards correct but cryptic moves?
  12. Joined
    06 May '08
    Moves
    1908
    22 Nov '10 23:44
    Originally posted by greenpawn34
    These super models are only any good to the very good players.
    I may not be a "very good player", but I think a super model wouldn't go amiss...
  13. e4
    Joined
    06 May '08
    Moves
    42492
    23 Nov '10 00:04
    Hi MR

    But dumbing it down to my level is doing me what good?

    I think I'm good enough to see who does who does not have a plus.
    If there is a combination in there (espicially if it's a mate) I'll find it.

    My weakness is I am looking for that jaw dropping game and I
    shun the so called quiet postions.

    What I say and how I play are often two different people.

    Now how are these things going to help me and players like me and you...us.
    Our cheif weakness is ourselves and to change that we have to have
    a complete personality make over.

    And that's it.
    There is more too chess than knowing the finer points of the game.
    At the GM level the knowledge is shared, it's the same.
    Then you are into the character, the presence, the forcing of wills.
    Things you cannot teach and certainly not get from any computer.
  14. e4
    Joined
    06 May '08
    Moves
    42492
    23 Nov '10 00:332 edits
    Hi EinZweiDrei

    The opening should be used to get you to a middle game YOU want to play.
    Taking a move from an opening DB without knowing why it was played and
    what is it preventing or threatening is dodgy.

    If you get involved in a tactical line then be very careful about the
    move that leaves the DB. DB's are main line moves, they do not show
    all (or any) of the traps (some of which can be very deep and do not need
    to end in mate or a win of the Queen - there is such a thing as a postional opening trap).

    These bases are made up from GM games. They won't play opening tricks v
    each other in the opening, but at the lower level these things work.

    On here the Blackburne Shilling Gambit has claimed 104 victims.

    1. e4 e5 2. Nf3 Nc6 3. Bc4 Nd4 4. Nxe5 Qg5 5. Nxf7 Qxg2 6. Rf1 Qxe4+


    The Shilling Trap is not even what I would call a heavy tactcial line.
    I'm just using it as an example of how moves no GM woul play v another
    GM win at a weaker level,

    So blindly taking moves from a DB made up of GM games is OK provided
    the other lad is doing the same.

    In short you are both making moves which will make sense in the coming
    Middle game

    But the chances are you won't see the middle game position the opening
    moves in the DB were designed for.

    You will instead be looking at a pawn or piece sac, a premature attack,
    a basic one move threat or even a sensible developing move that has
    been deemed too passive at GM level so therfore unplayed.
  15. THORNINYOURSIDE
    Joined
    04 Sep '04
    Moves
    245624
    23 Nov '10 00:401 edit
    Originally posted by greenpawn34
    I'm Replacing Carlsen.....

    Why did you move him in the first place? 😛
Back to Top

Cookies help us deliver our Services. By using our Services or clicking I agree, you agree to our use of cookies. Learn More.I Agree