[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums

Error symbols

Posted By: Nack Ballard
Date: Wednesday, 16 February 2011, at 11:22 p.m.

In Response To: Committing to Memory: what about a bg informator? (higonefive)

You have some interesting ideas here. Personally, I study from Nactation alone for early game positions. It's not that I prefer dry strings of characters and don't care about the diagrams. Rather, it's that I "see" the diagrams in my head, so I would much rather not to have to read or scroll past the diagram to get to the next nugget of information -- it slows me way down.

OTOH, I'm sure that many people either need or imprint better seeing the position on the page. I'm just pointing out that both type of people exist, at least to some extent. Perhaps it would be prudent to have one section without diagrams and in another (much larger) section have the exact same information repeated with diagrams; anyway, something to consider.

You suggested the use of chess-like symbols to describe error categories. This is similar to though perhaps an improvement on the notion of using a small-case letter (instead of the margin in thousandths) after each character in the nacbracs (such as t for tied, v for very close, c for close, etc., or perhaps the simpler a, b, c...)

Some chess symbols refer to surprisingly strong moves. For example, ! refers to a good move that is difficult to see (and !! to a good move that is very difficult to see), it being possible that there are other moves almost as or as strong. And !? means interesting and probably good, and ?! means dubious. This is different from the concept of a wipeout or double wipeout in backgammon, where the best move is much better than the second best candidate. A double wipeout might be equivalent to recapturing a piece in chess when failing to do so will leave you that far behind in material without compensation, and you would certainly not use "!!" for such an obvious move.

With your backgammon informator approach, I think it would be better to limit your symbols to error sizes (and not to "brilliant" moves). First of all, there will be fewer symbols to learn. Secondly, there aren't many symbols left that Nactation and nacbracs aren't already using. Thirdly, it would seem more economical to use one character instead of sometimes two or three (e.g., "(?)"). Fourthly, based on your examples so far (though correct me if they're atypical or if you intend broader usage), it seems that your are only designating the errors (not the blowouts, etc.) anyway.

Here is one possible symbol scheme for error sizes:

    ...... Mnemonic ....... Error .... Descriptive
    ..... word/phrase ...... size ..... error size

    = .. (equivalent) .... .00–.01 ... Tied
    ; .. (gives pause) ... .01–.02 ... Very close / Barely wrong
    : .. (more pause) .... .02–.03 ... Close / Marginally wrong
    \ .. (downhill) ...... .03 + ..... Wrong
    ? .. (very iffy) ..... .06 + ..... Blunder
    { .. (rightward W) ... .10 + ..... Whopper
    } .. (leftward W) .... .20 + ..... Double whopper

    .... (...or double whopper could be {{ or 2, etc.)

For example, the last position of your prototype informator doc (according to bot aggregate value) could be summarized as

    31P-64 [R S; P:]

... which, with exact error sizes, is normally written as

    31P-64 [R S18 P27]

If the rollout or evaluation is done by a single bot, the information can be displayed after the bracket (<46 or whatever), as usual.

If you just want the error symbol to be part of a move string, it can inserted after the character, for example, adding a semi-colon (barely wrong) after the S:

    31P-64S;-62H ..... etc.

Then again, if what you are doggedly after is to as closely as possible resemble the format of the chess informant, you can use ! and !!, etc., but keep in mind this has to do with the cleverness of the move rather than its distance from the second best play. The chess position evaluation symbol strings such as +- (winning advantage for White) might not translate so well into backgammon, but you could modify them to refer to ranges of equities or winning chances adjusted for gammons.

Anyhow, that's food for thought.

Nack

Messages In This Thread

 

Post Response
Your Name:
Your E-Mail Address:
Subject:
Message:

If necessary, enter your password below:

Password:
 

 

[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums is maintained by Stick with WebBBS 5.12.