search

lichess-org / lila

♞ lichess.org: the forever free, adless and open source chess server ♞
https://lichess.org
GNU Affero General Public License v3.0
15.09k stars 2.24k forks source link

Lichess study does not recognize time stamps of DGT boards during import. #14343

Open eskopp opened 8 months ago

eskopp commented 8 months ago

Exact URL of where the bug happened

https://lichess.org/study/nDKoIXr8/eKYon9wb

Steps to reproduce the bug

  1. Open a new study
  2. upload the PGN file
  3. Wait briefly until lichess is ready
  4. be sad that the time is not there ^^

What did you expect to happen?

I had expected the remaining reflection time to be similar to the following broadcast (https://lichess.org/broadcast/cracovia-2023-open-a/round-5/dwJJOHqX). The required data is contained in the game. An example can be found at the end of the issue

What happened instead?

Lichess ignores the TimeStamps

Operating system

Windows 11

Browser and version (or alternate access method)

Chrome 122.0.6182.0

Additional information

Chapter from the PGN-File:

[Event "20. Bad Koenigshofen Open 2023"]
[White "Skopp,Erik"]
[Black "Müller,Gerhard"]
[Site "Bad Koenigshofen, Germany"]
[Round "5.22"]
[Annotator ""]
[Result "1/2-1/2"]
[Date "2023.12.29"]
[WhiteElo "1505"]
[BlackElo "1744"]
[PlyCount "20"]
[TimeControl "60"]

1. d4 {[%emt 0:0:2]} d5 {[%emt 0:1:45]} 2. Bf4 {[%emt 0:0:10]} Nf6 {[%emt 0:1:33]} 3. Nf3 {[%emt 0:0:7]} Nc6 {[%emt 0:2:5]} 4. e3 {[%emt 0:0:15]} a6 {[%emt 0:0:10]} 5. c3 {[%emt 0:0:12]} Nh5 {[%emt 0:1:32]} 6. Bg3 {[%emt 0:1:19]} Nxg3 {[%emt 0:2:1]} 7. hxg3 {[%emt 0:0:3]} Bg4 {[%emt 0:1:17]} 8. Bd3 {[%emt 0:0:27]} e5 {[%emt 0:4:36]} 9. dxe5 {[%emt 0:3:14]} Nxe5 {[%emt 0:0:26]} 10. Be2 {[%emt 0:0:4]} Nc4 {[%emt 0:3:29]} 1/2-1/2
kraktus commented 8 months ago

related to https://github.com/lichess-org/lila/issues/13868, https://github.com/lichess-org/lila/issues/10142, which was closed with

%emt is the time one spent playing a move.

It's not enough to reconstruct the game clocks, because it doesn't account for client lag, network lag, server lag, lag compensation, and possible manual changes to the clock (such as giving time).

Clock inaccuracy would increase with every move, eventually resulting in nonsensical clock states, such as flagging with non-zero time on the clock.

Therefore it has little value and can't be used by lichess studies.

I would be in favour of revisiting this as

client lag, network lag, server lag, lag compensation

are not an issue for OTB game.

eskopp commented 8 months ago

Hello @kraktus If Lichess sees it that way, I will close the issue. Thank you for the information.

eskopp commented 8 months ago

Hey @kraktus. One question. If the data from the PGN is not enough to realize the TimeStamps, why did you open the issue. I mean if the data is not enough, it is not enough. I would be happy, but I trust your judgment.