Page 25 of 53

Re: EGF vs AGA pros win-and-continue match

Posted: Wed May 15, 2019 2:46 am
by Javaness2
It would be interesting to know if both teams flat out refused a rematch. Such an impasse might explain the final change. Certainly, I don't see anything else that does.

Re: EGF vs AGA pros win-and-continue match

Posted: Wed May 15, 2019 2:56 am
by Bojanic
Javaness2 wrote:It would be interesting to know if both teams flat out refused a rematch. Such an impasse might explain the final change. Certainly, I don't see anything else that does.
But rematch would be a stupidest idea in this stupidity fair.
If they played it under same rules, what would change?

Re: EGF vs AGA pros win-and-continue match

Posted: Wed May 15, 2019 3:02 am
by Javaness2
Bojanic wrote:
Javaness2 wrote:It would be interesting to know if both teams flat out refused a rematch. Such an impasse might explain the final change. Certainly, I don't see anything else that does.
But rematch would be a stupidest idea in this stupidity fair.
If they played it under same rules, what would change?
It's not that stupid if you preface it with "We apologize for not having proper rules in place at the start of this event. From now on, these are the rules that we will use." I think it is quite diplomatic that way.

Re: EGF vs AGA pros win-and-continue match

Posted: Wed May 15, 2019 3:31 am
by k0n0
yakcyll wrote:
Renter wrote:Let this sink in: The tournament organizers are basically saying that if you do not after EACH AND EVERY lag spike inform the TOs of it, they will use that fact against you.
That's a strawman. Has anybody aside from Mateusz known that he had connectivity issues in previous games? Do you really think nobody would react and help him out if he had mentioned it the first time around, before a loss occurred because of it?
How could they react and help him? Do you think they would leave KGS and choose another server because of a 3-second lag? Did they do it now, after a 10-second lag?
It looks no measures that would cope with lags were taken now, even after one game was lost due to a 10-second lag.

(Moreover, if Mateusz reclaimed that loss of a period due to a 3-second lag, the referee would ask him whether he had another lag before. If Mateusz acknowledged he had some half-second lags before, it might end for him by another game lost by a refereee decision. ;-) )

So far it looks the message from the referee is
  • in case of lag problems you must inform us so that we can react
  • now we are informed, and we won't react

But I may be mistaken, and they will play next game with real boards and clocks, and proctors will send moves to KGS?

edit: clock->clocks

Re: EGF vs AGA pros win-and-continue match

Posted: Wed May 15, 2019 4:13 am
by Bojanic
k0n0 wrote: But I may be mistaken, and they will play next game with real boards and clock, and proctors will send moves to KGS?
Finally a good idea.

Re: EGF vs AGA pros win-and-continue match

Posted: Wed May 15, 2019 4:14 am
by Bojanic
While trying to send previous reply:
This site can’t be reached www.lifein19x19.com took too long to respond.
Try:

Checking the connection
Checking the proxy and the firewall
Running Windows Network Diagnostics
ERR_CONNECTION_TIMED_OUT
Lag appears in topic about lag, showing off now, obviously.

Re: EGF vs AGA pros win-and-continue match

Posted: Wed May 15, 2019 4:15 am
by tneva82
Bill Spight wrote:
Bojanic wrote:PS for european side, I would demand that next year match is to be played at KGS (Kazan Go Server), so other team could experience same difficulties.
As someone on neither side, I would recommend having time kept locally, offline.

That would be either impractical or open even bigger can of worms. You either have smart client on players computer open for tinkering and fooling the server(there's reason why dumb clients are default...Smart clients can be tinkered and there's then no way for server to verify. In effect player could give himself more byo-yomi than assigned time and server would get moves correct times) OR you have to go for heavy duty. Private custom client with own servers installed on (laptop) computers which are delivered for the game by 3rd party for players and never left alone with players to ensure they don't install any hacks. Preferably with firewall preventing access to anything but relevant addresses for game.

Do you think that's feasible? But as it is only way to play for big prize pools online clearly. As is this is not appropriate for prize pools when server lag players can't control can decide fate of that money.

Re: EGF vs AGA pros win-and-continue match

Posted: Wed May 15, 2019 4:21 am
by tneva82
Kirby wrote: Presumably, when Mateusz was contesting the repeal, it became clear that the repeated lag on his end was something that he knew about before it was game deciding - and maybe if it had been brought up prior to game #4, they could have worked together to come to a way to mitigate the lag problem (or maybe clarify what should happen in the case of lag).

I should be clear, I know nothing about the reasoning behind Myungwan's decision - I'm simply reading what was given on the FB page.

Of course only way they could do something to that would be fly them to same place and play it live rather than online.

Online go seems to be failure for big real money tournaments and more of for fame and fun.

Re: EGF vs AGA pros win-and-continue match

Posted: Wed May 15, 2019 4:24 am
by tneva82
jlt wrote:
- Mateusz was aware of the technical issue for some time, but never reported the issue to the tournament organizers
Now the organizers are aware of a lag problem in Kazan. What will happen if another player (e.g. Ilya Shikshin) plays from there?
Kazan? Howabout finland on various sites. Happens to me as well. It's not just Kazan. It's INTERNET. Lag is quaranteed period. Whole TCP protocol is built on assumptions packets WILL be lost and built so that they will get through no matter how long it takes. Because delays and packet losses are 100% quaranteed part of internet. There is no way to ensure those will never happen.

Re: EGF vs AGA pros win-and-continue match

Posted: Wed May 15, 2019 4:32 am
by jlt
One way to reduce the costs could be:
  • Play a few rounds online, but each side uses a real board and a real clock, the proctor relays moves on KGS.
  • For the final rounds, use an international event (e.g. IMSA) that most players attend to anyway, and stay 2-3 more days.
tneva82 wrote:Kazan? Howabout finland on various sites. Happens to me as well. It's not just Kazan. It's INTERNET.
I know, I should have said "at least in Kazan there is a serious lag problem which may cause delays of more than 10 seconds".

Re: EGF vs AGA pros win-and-continue match

Posted: Wed May 15, 2019 4:58 am
by Bill Spight
Javaness2 wrote:
What does his ruling hold for future games, what does it mean?
Players should call the referee or TD when an irregularity has occurred, to protect themselves.

Re: EGF vs AGA pros win-and-continue match

Posted: Wed May 15, 2019 5:01 am
by Bill Spight
jlt wrote:
tneva82 wrote:Kazan? Howabout finland on various sites. Happens to me as well. It's not just Kazan. It's INTERNET.
I know, I should have said "at least in Kazan there is a serious lag problem which may cause delays of more than 10 seconds".
I live in a major metropolitan area in the US, and I regularly get a message saying, "You are not connected to the internet." It can take minutes to reconnect.

Re: EGF vs AGA pros win-and-continue match

Posted: Wed May 15, 2019 5:26 am
by k0n0
Bill Spight wrote:Players should call the referee or TD when an irregularity has occurred, to protect themselves.
The question is: what is an irregularity?

The text "to protect themselves" indicates "irregularities" might be regarded from the point of view of players.
From the point of view of Mateusz a 3s-lag wasn't irregular, and Mateusz got accustomed with it, by playing at least 10s in advance.
Only the 10s-lag was irregular, and he reported it.

Still the verdict is blaming Mateusz.
So it looks in the verdict they speak rather about general KGS/internet irregularities. But lags are quite regular with both internet and KGS.

Re: EGF vs AGA pros win-and-continue match

Posted: Wed May 15, 2019 5:26 am
by Bill Spight
tneva82 wrote:
Bill Spight wrote:
Bojanic wrote:PS for european side, I would demand that next year match is to be played at KGS (Kazan Go Server), so other team could experience same difficulties.
As someone on neither side, I would recommend having time kept locally, offline.

That would be either impractical or open even bigger can of worms. You either have smart client on players computer open for tinkering and fooling the server(there's reason why dumb clients are default...Smart clients can be tinkered and there's then no way for server to verify. In effect player could give himself more byo-yomi than assigned time and server would get moves correct times) OR you have to go for heavy duty. Private custom client with own servers installed on (laptop) computers which are delivered for the game by 3rd party for players and never left alone with players to ensure they don't install any hacks. Preferably with firewall preventing access to anything but relevant addresses for game.

Do you think that's feasible? But as it is only way to play for big prize pools online clearly. As is this is not appropriate for prize pools when server lag players can't control can decide fate of that money.
For this particular tournament, with one player playing locally, you can have the player using a real board with a clock, while a proctor relays the plays online, plays the opponent's stones, and keeps the time record offline. The proctor may also count byoyomi, just like in the old days. ;)

As for large tournaments with several players in one location at any time, perhaps they could play on a local network, with the local server keeping time and relaying plays to and from the internet. The organizers would be responsible for the integrity and security of the local server.

Re: EGF vs AGA pros win-and-continue match

Posted: Wed May 15, 2019 5:42 am
by Bill Spight
k0n0 wrote:
Bill Spight wrote:Players should call the referee or TD when an irregularity has occurred, to protect themselves.
The question is: what is an irregularity?

The text "to protect themselves" indicates "irregularities" might be regarded from the point of view of players.
Better safe than sorry.
From the point of view of Mateusz a 3s-lag wasn't irregular, and Mateusz got accustomed with it, by playing at least 10s in advance.
Only the 10s-lag was irregular, and he reported it.
I sympathize. I do. IMO the netlag problem was known by the organizers and not his responsibility to report. But any net lag is, in terms of keeping an accurate record of time usage, an irregularity. Certainly a 3 second lag is significant.
Still the verdict is blaming Mateusz.
Yes. I blame the organizers for using KGS to keep the time record.