Talk:Shogi notation

SFEN needs more detail
The Shogi notation section needs a better example, showing that the holdings field uses counts if multiple pieces are held instead of having a letter per piece held. For example, this famous naked king tsumeshogi problem would be represented: `8k/9/9/9/9/9/9/9/9 b R2GSr2b3g2s4n4l18p` An example showing promoted pieces might also be helpful.

Also, I notice that some SFEN generators such as http://sfenreader.appspot.com/en/create_board.html also append a number, presumably a count of half-moves from the beginning of the game. It appears to be optional.

In addition, you might add a section for the similar Extended Position Description (EPD) notation: `8k/9/9/9/9/9/9/9/9[RGGSrbbgggssnnnnllllpppppppppppppppppp] w` This is the position format used by XBoard for shogi position import/export, but I don't know if it is widely used otherwise.

--IanOsgood (talk) 21:44, 15 December 2017 (UTC)


 * Good points. Thanks. I'll eventually get around to fixing it. (Feel free to do so yourself!) – ishwar  (speak)  19:27, 15 January 2018 (UTC)


 * Oh, for the move number suffixes: well, i guess there isnt an official SFEN specification, but i think that suffix is something added by that website and not really a part of the SFEN string itself. But, i need to go and read some stuff to make sure about. Adding that suffix is not a good idea because these previous game states have nothing to do with what follows computationally. Unfortunately, i see that the shogi game record database is using those suffixes. And, a popular computer shogi engine format (やねうら王 YaneuraOu) uses the suffixes in its opening book file, which means that a position encountered once could be found by the book file but the same position occurring after or before would be missed! And, i've even found a case where the book file (for the 2017 WCSC tournament version of the Tanuki engine) had some moves misnumbered so that entire lines from the book file were simply ignored by the engine! Ridiculous. I have no idea why anyone would think is a good idea.


 * But, at any rate, these move numbers should be mentioned. – ishwar  (speak)  19:37, 15 January 2018 (UTC)

Ambiguity resolution: Movement description
Could someone please add the Japanese pronunciation for these movements (in the Latin alphabet) to the table? Thank you in advance. 2A00:23C5:D00C:ED00:7F:FF89:988:31FE (talk) 23:58, 1 May 2020 (UTC)

Alternate kanji for Dragon
Probably worth mentioning '竜', the alternate character for the Dragon (promoted Rook) that I often seen in books, diagrams, and shogi pieces. --IanOsgood (talk) 23:36, 23 September 2020 (UTC)


 * good point, too. I'm going to work on this article sometime in the (hopefully) near future. – ishwar  (speak)  00:19, 24 September 2020 (UTC)