EGF vs AGA pros win-and-continue match

[Javaness2]

I know that your idea is good, but nobody is going to take the time and effort to implement some technological solution like that for a one off event.

[Uberdude]

A valid criticism of the double loss ruling is that Eric loses by fiat, not having done any of the things that are ordinarily required to lose:
- resign
- score game and have fewer points
- run out of time
- forfeit due to rule breach

So I propose a slight modification
1. Mateusz lost on time so next EGF pro plays
2. To decide if Eric lost play resumes between Mateusz and Eric from the last position pre KGS timeout loss
a. if Mateusz wins then Eric loses and next AGA pro plays
b. if Eric wins then he wins and plays next game for AGA.

This sets no precedent for future games, which must be played after clearer rules regarding lag in internet play are agreed.

This means no decision is made by the referees about the board state like 'Mateusz was winning by about 10 points'. The only effect the board state has is the 2nd-order effect of meaning the break in play is judged to offer neither side a significant advantage in being able to pause, collect their thoughts, ask a friend/bot.

[Bojanic]

I think there could be only 2 options for ruling:
- loss in the game due to timeout,
- continuation of game if lag argument is accepted.

Situation on the board is following:
White:
- 19 captures
- 6.5 komi
- upper group: 13+7 in best case scenario (black never plays first)
- lower group: 45 points
Total: 83 +7 in best case scenario

Black:
- 18 captures
- upper right: 15+1
- right: 2
- left and center: 60+3
Total: 95+4

Even if white takes basically all endgame points, he will loose by at least 5 points.
There was 6 places where it was possible to play, all worth less than 3 points.
It would probably end in some 10 points difference.
We can conclude that in this case game was basically decided, there was no chance to turnover, therefore there was no need for Black to try to manipulate game.

----

Finally, given all into account, it is embarassment for organizers of highest level intercontinental match with good prizes, to have such situation occured, and that game is to be decided not by players skill, but by technical unforeseen glitch or judges decision.

[Bill Spight]

A valid criticism of the double loss ruling is that Eric loses by fiat, not having done any of the things that are ordinarily required to lose:
- resign
- score game and have fewer points
- run out of time
- forfeit due to rule breach

So I propose a slight modification
1. Mateusz lost on time so next EGF pro plays
2. To decide if Eric lost play resumes between Mateusz and Eric from the last position pre KGS timeout loss
a. if Mateusz wins then Eric loses and next AGA pro plays
b. if Eric wins then he wins and plays next game for AGA.

This sets no precedent for future games, which must be played after clearer rules regarding lag in internet play are agreed.

There is a precedent, going back to a different case where the rules were unclear.

Eric won, but Mateusz did not lose.

[Bojanic]

I know that your idea is good, but nobody is going to take the time and effort to implement some technological solution like that for a one off event.

It is not for one off event, but for all future hi-profile games.

It is unacceptable that hi-profile games in 2019. should depend on quality of connection, or any other technical problem.
What would happen if ie there was short power shortage in byoyomi?
Those questions must be addressed.

[Bill Spight]

Finally, given all into account, it is embarassment for organizers of highest level intercontinental match with good prizes, to have such situation occured, and that game is to be decided not by players skill, but by technical unforeseen glitch or judges decision.

Agreed, except for the unforeseen part.

Edit: It may have been unforeseen, but it was definitely not unforeseeable. Netlag has been a problem for a long time.

[Uberdude]

Speaking of modified clients, I wonder if any players are using pasky's hacked KGS client. Unlike official KGS. it supports audible byo-yomi countdown (as most Asian servers and OGS do).

And on power cuts, does anyone know the PGETC rules for that? I know when South Africa joined the tournament that was an issue because they can suffer from rolling power outages.

[Bojanic]

With online games there is lot of things which can go wrong.

I am playing games on some 10 years old desktop computer, with Windows XP, integrated graphics card.

While playing PGETC game some years ago, during period when my opponent had to think a lot, I opened another window and took a look on my teammates games (what I usually do at live games as well). At that point I lost graphics in my game. I tried opening and closing window, applying different settings, and best I could do (without restarting computer) was to have stones in strange colors, on board without lines.
What to do in such situation?
Or if windows during long thinking decides to open screensaver, or run update, that could slow down game or make some other disturbances.
Etc, there is long list of unforeseen events that could go wrong.

[jlt]

On IGS, when you lose the connection (due to a power shortage, restarting the computer or anything else), you can reconnect within 5 minutes and resume the game.

[yakcyll]

Speaking of modified clients, I wonder if any players are using pasky's hacked KGS client. Unlike official KGS. it supports audible byo-yomi countdown (as most Asian servers and OGS do).

I was under an impression you could set the timer to flash, beep or both for the last n seconds of byo-yomi in the official client as well (3.5.22).

[Uberdude]

Speaking of modified clients, I wonder if any players are using pasky's hacked KGS client. Unlike official KGS. it supports audible byo-yomi countdown (as most Asian servers and OGS do).

I was under an impression you could set the timer to flash, beep or both for the last n seconds of byo-yomi in the official client as well (3.5.22).

Yes, just beep, but not say "ten, nine, eight...". You can even get it in chid0ri's (author of empty triangle comic) voice.

[Renter]

From a TO and EGF referee (though not for a while) perspective, I would like to sum up the situation:

• * One player was on their last byo-yomi period.
  * A game loss by time was decided by the server.
  * Player claims they had played their move while 10 seconds in the clock were remaining.
  * Proctor/observer in the room confirms this.

To me it seems obvious that the game must resume. No other option is fair or reasonable or in the spirit of the game.

In the future steps can be taken to mitigate these kinds of events, including:

• * Local time stamps, even if server time is used local time for each move SHOULD be recorded, for cases such as this at least.
  * Proctors/observers should be instructed to pay close attention when in last byo-yomi period (or last minute or last 1-3 stones in the case of Canadian byo-yomi or similar).
  * A camera recording the player could be used.

The raison d'etre of time rules is to keep the tournament at a reasonable length. They should never be about "HA! HA!" moments for the player with the worse Internet connection, especially in a game between professionals or any other serious game.

I have lost games with time. I have won games with time. I have used time-suji on occasion and had time-suji used against me. I have called wins and losses for time and resumptions on many occasion, including once where a player lost the same game 3 times with time in the european go congress. Time settings are an integral part of any tournament and they should be respected, but the rules and the rulings must also align with the spirit of the game and the tournament and outside interference (faulty clocks, natural disasters, fire alarms, internet lag) should never be a major factor. If it seems like it would be a major factor, the TOs must take action to prevent or mitigate such problems.

In the case of a fire alarm or a natural disaster, it is obvious that any games would be returned to a known good state and continued if it is possible to do so and both players are capable. In my opinion, if the tournament rules do not account for internet lag, it is outside interference and should not be a major factor in deciding any game result. Chess clocks break or get faulty buttons all the time, you don't call games based on that.

If the tournament organizers decide to call the game loss by time or other options except resuming, they are essentially washing their hands from the results of their own choices of venue/rules/observers, saying they are not responsible for KGS problems or Internet problems, KGS rulings are the word of Gord and that they do not trust their own proctors. Why have them in the first place, then? This is a game between professionals and even the idea that one player should be handicapped by being forced to play within 30 or 40 seconds when the other has 1 minute is absurd.

TL;DR If the chess clock breaks, you get a new one and carry on with the game.

[Kirby]

To me it seems obvious that the game must resume. No other option is fair or reasonable or in the spirit of the game.

Resuming the game is one option, but clearly not one that people universally find to be the best option. The length of this thread proves that.

[mhlepore]

We should roll a die to determine what happens:

1 = Eric wins right now
2 = Mateusz wins right now
3 = Resume game
4 = Rematch
5 = both are eliminated and next players are up
6 = nothing - roll again until get 1-5

I'm not even joking. Making it random will take away people's ability to complain about the organizers' rationale when they reach their decision. This is a crappy situation, and there is no consensus, so roll a die/flip a coin and move on with life.

[jlt]

That's what the referees have been doing for days.

dice.jpeg (7.04 KiB) Viewed 7822 times