Seriously Messed Up Clocks

Sort:
Brother-Mouzon

I just got done playing a nice friendly game of 1/0 that seemed to go on forever.

I lost by about one second but when I looked at the move count it was 57.

How likely does that sound to you in a straight game of 1/0 among players rated below 1000?

When I searched I found an explanation that sounded like it was written by somebody in a heavily post modern English department at University and mentioned using some sort of algorithm to allow for "thinking time."

The point of bullet is to think and play fast.  

Has anybody else encountered the same problem in a game of fast chess?  This is my fourth game in two days out of about 25 in which this was obvious and I have never experienced it here before even though I have encountered my share of clock cheaters.  

I have also won one game which turned the same table in my direction.  It felt all wrong after well over one thousand games.

This is completely different from clock cheating.  

notmtwain
Brother-Mouzon wrote:

I just got done playing a nice friendly game of 1/0 that seemed to go on forever.

I lost by about one second but when I looked at the move count it was 57.

How likely does that sound to you in a straight game of 1/0 among players rated below 1000?

When I searched I found an explanation that sounded like it was written by somebody in a heavily post modern English department at University and mentioned using some sort of algorithm to allow for "thinking time."

The point of bullet is to think and play fast.  

Has anybody else encountered the same problem in a game of fast chess?  This is my fourth game in two days out of about 25 in which this was obvious and I have never experienced it here before even though I have encountered my share of clock cheaters.  

I have also won one game which turned the same table in my direction.  It felt all wrong after well over one thousand games.

This is completely different from clock cheating.  

I don't believe that anything has changed recently.

That explanation about how they adjust the clocks to account for lag has been up for years.  It always made sense to me. Do you want to have it explained to you again?

If you take one second on a move and then click your mouse and it takes three seconds to make its way to chess.com, that's a total of four seconds.

How many seconds should show on your clock for your move? You took only one second. Should you be charged for four?

 They say that they try to avoid this whole situation by measuring the lag when people request a game and they don't allow people who have lags over two seconds to play. In a game where two seconds is being subtracted from every move, a 60 move game could be  adjusted by up to two minutes per side. That would be a huge adjustment in a 1 0 game.

I would think that allowable lag would have to be set much lower than two seconds in a 1 0 bullet game (but I don't know what it is set at.)

notmtwain
surelyouchess wrote:
notmtwain wrote:
Brother-Mouzon wrote:

I just got done playing a nice friendly game of 1/0 that seemed to go on forever.

I lost by about one second but when I looked at the move count it was 57.

How likely does that sound to you in a straight game of 1/0 among players rated

When I searched I found an explanation that sounded like it was written by somebody in a heavily post modern English department at University and mentioned using some sort of algorithm to allow for "thinking time."

what? 

The original poster was confused by this explanation offered on the site's Help page.

Why did the clock times suddenly change?

The clocks seem broken!

 

Don't worry - the clocks are not broken. This jumping is caused either by Internet lag,

or because of bonus time. Here is how lag works on chess.com:
 

You make a move, the move gets sent to our server, & then your opponent's computer.

We adjust the clocks dynamically so that neither player is "charged" for communication time

- but rather, only for the time actually spent thinking.
 

A very extreme example
 

Imagine it takes 5 seconds for the move to hit our server, then another 5 seconds for the move

to hit your opponent's computer from there. Your opponent will not see this move until

10 seconds after you make it, but your opponents clock will usually start counting down

on your screen when the move hits our server. Then, let's say your opponent thinks for a

total of 10 seconds. After this, it takes another 5 seconds for the move to be transmitted

to our servers, and another 5 seconds for the move to now go to your computer.
 

So, you have been waiting for this move for roughly 30 seconds (10 seconds total travel time,

10 seconds thinking time, and another 10 seconds total travel time), and your opponents clock

has been counting down for roughly 25 seconds. But since your opponent only spent 10 seconds

thinking, that's all the time they get charged for. So, when their move finally hits your computer,

their clock display re adjusts for the 15 extra seconds that your opponent wasn't actually thinking.

Meanwhile, your clock display on your end will only count down while you are thinking.

There is no need for correction on your end (and remember, your clock is adjusting similarly

on your opponent's screen).
 

Fortunately, in reality, travel times are usually less than what is described above.  (In fact,

we only allow members to play if their detected lag is two seconds or less.) However, the travel times

are not consistent, so adjustments must be made dynamically. There is no way around it:

all real-time chess servers have to deal with lag, though other sites are a bit less transparent

about it.

 

/ I don't find it difficult to understand. Do you?

Keywords: clocks, skip, clock, jumping, time, cheat, delay, bonus

notmtwain
JollyBoyJohn wrote:

I've heard of clock hacking before, like people hacking into your clock and stealing time from you and giving it to themselves. Maybe that's what happened? It does sound suspicious. 

It's not clock hacking. It's lag.

Did you read the explanation from the Help page?

notmtwain
JollyBoyJohn wrote:

Yes, that could be possible. It could also be clock hacking. Worth submitting a ticket. Better safe than sorry.

Not funny 2Q