KGS/SGF treatment of time

[jgr314]

Sorry that this is a technical question. Can anyone clarify how KGS-generated sgf files deal with playing time?

Some components seem clear:
(1) main time remaining (after?) a move is played are indicated by something like BL[1668.753] indicating "black has 1668.753 seconds left." For white, the equivalent is WL[#]
(2) byo-yomi remaining is indicated by something like WL[30]OW[3] indicating white has three 30second periods remaining. I assume similar for black is BL[30]OB[3]
(3) result determined by a player running out of time is indicated near the start of the file with RE[W+Time]

Here's my confusion: in a recent game, approximately the last 10 moves don't have any time indicated, for either black or white. here's an example from the end of the file:
B[af]BL[304.362]
;W[hq]WL[30]OW[1]
;B[cg]BL[300.299]
;W[fo]WL[30]OW[1]
;B[en]BL[167.222]
;W[ep]
;B[eq]
;W[fp]
;B[gp]
;W[go]
;B[ho]
;W[hp]
;B[gq]
;W[gn]
;B[hr]
;W[ir]
;B[jr]

For what it is worth, black ended the game with about 2 minutes main time remaining (120 seconds).

My questions:
(a) is the amount of byo-yomi time used on each move recorded in some way? If so, I don't see it.
(b) why didn't KGS record the remaining time for the moves at the end of the game? I'd note that these moves weren't played in review after the end of the game, they were normal game moves while the clock was ticking.

Thanks!

[jeromie]

a) No, it is not.

b) I don't know the answer for this. Do you have a link to the whole sgf?

[Kirby]

You might be interested in the time section of the SGf spec, which shows the standard properties: http://www.red-bean.com/sgf/properties.html#BL

[EdLee]

Hi jgr,

CGoban also has bugs.
( Not that this case is definitely a bug, but it's a possibility. )