Best way for a clock to add the second time control?

Running Chess Tournaments
1

Different chess clocks have different ways of adding the second time control. Which do you think is the best approach and why?

  1. Add the second time control for each player once they have reached the prescribed number of moves in the first time control.

  2. Add the second time control for each player once their time in the first time control runs out and only if the prescribed number of moves in the first time control have been met.

  3. Add the second time control for both players simultaneously once one of the players time in the first time control runs out, regardless of if the prescribed number of moves have been met or not.

  4. Add the second time control independently for each player once they run out of time in the first time control, regardless of if the prescribed number of moves have been met or not.

2
  1. Have the clock count up to a time control. No addition of time to the clock is necessary.
3

I don’t like move counters and never use them if given the choice, so that rules out #1 and #2. I can live with either of the other two (as long as the clock gives some indication (like the bar up in the corner for Chronos) that we are now in the second control), but I prefer #4, since that’s what I’m more used to. Whether or not the prescribed number of moves has been made is a scoresheet function, not a clock function (and any claims need scoresheet verification anyway).

4

IMHO clocks should not automatically move to the next time control.

I see this leading to two choices:

  1. Have only one time control.
  2. Have a separate button on the clock that advances to the next time control, which should only be activated after both players agree the time control has been met.

There is a third option, having the TD reset the clock, which is what most of us did back in the analog clock days. But there are so many different types of digital clocks that I doubt most TDs know how to set them all. I know I don’t.

5

What is the problem with a clock advancing automatically as long as there is a mechanism to show that it has done so, like in Mr. Bachler’s proposal above or an analog clock with the secondary control being an integer number of hours?

Alex Relyea

6

In the book Chess Handbook – Book for Arbiters by IA Zoran Bojovic and IA Branislav Suhartovic, which I have touted previously on these forums (and which is available from USCF Sales, or is it U.S.Chess Sales yet?), the following FIDE rules are quoted, and the authors’ opinion is expressed:

6.3.1. … each player must complete a minimum number of moves or all moves in an allotted period of time …

6.4. Immediately after a flag falls, the requirements of 6.3.1 must be checked.

“From 6.4 unequivocally follows that only when the allotted time expires, the control of the number of moves played is performed.” That is, the arbiter figures out whether 40 moves have been played only after a player’s flag has fallen. Therefore, the authors argue, it is illegal for the clock to add time before the arbiter has confirmed the move count.

That sounds as though the authors are proposing that time should be added only when the first control time runs out. But then they go on to say, “this also isn’t completely in accordance with the Laws of Chess. … from 6.4 clearly follows (implicitly) that it first must be established whether the players have completed the prescribed number of moves, and only then the time for the second time control added.”

The authors then conclude that the only correct way to set the clocks is to set them for just one control (such as G/90) and then add time manually after it is determined that 40 moves have been played.

To me, all of this sounds so 20th century, backward, and ignorant of current technology. What arbiter in a large tournament wants to go around verifying 40 moves at each board, and adding time by hand?

Kevin Bachler’s suggestion to set the clock to count forward, starting at (say) 4:30:00 if the control is 40/90 SD/30, so that the first period expires at 6:00:00 and the second at 6:30:00, sounds like a decent idea, if the clock can do that and opponents don’t freak out. The Chronos clocks can do that, even with a (non-cumulative) delay, but can they do that with (cumulative) increment? Somebody who owns a newer Chronos please let me know.

Bill Smythe

7

Alex, what if the move counter is wrong, which, in my experience, it is about 25% of the time?

The question is, which is worse, a clock that adds the next time control too soon or one that adds it too late? Hobson’s choice?

8

I guess I don’t understand the question. Consider, for the sake of argument, a time control of 40/90, SD/30 increment irrelevant. At some point White’s clock says he has 29:59 in the second time control. It is up to the players (under US Chess rules) or the arbiter (under FIDE rules) to determine whether 40 moves have been made. This is regardless of whatever the clock-press counter indicates. It doesn’t really matter when the time is added as long as the players are keeping score, and that the clock doesn’t halt either player’s time when nonfinal controls expire.

Alex Relyea

9

Note that on a digital clock counting up, the secondary control wouldn’t have to be an integer, although that would be best as it is easier to recall.

We are used to the concept of a flag, or 000, but that’s not really necessary, a clock that has counted up past the possible time control has “flagged”.

I suppose that increment would become subtracting time from the clock, in the case of a clock counting up.

10

One advantage of counting down is that “flag fall” is at the same point from tournament to tournament. Players wouldn’t have to remember if flags fell at 30 this week or 75 (more likely 30 or 25).

Yes, it would be cumulative or non-cumulative decrement.

Alex Relyea

11

Doesn’t counting down also avoid confusing issues (with decrement) of negative time? (For instance, if it’s G/60 +30 and I make my first few moves in ten seconds each, wouldn’t my clock have to be at negative one minute at that point?)

12

I was thinking more of a retro trip to the analog days. Set the count-forward clock so that the first control expires at 6:00:00. For example, with 40/90 SD/30, set it initially to 4:30:00, so that the first control expires at 6:00:00 and the second at 6:30:00.

That way, increment would never take you into negative-land.

Bill Smythe

13

Clocks that do not display seconds until under 10 or 20 minutes would leave people wondering how many seconds they had left to make moves.

14

Leaving aside the question of how to deal with increments, it seems to me that a “count-up” clock inherently shows how much time you’ve used, whereas a “count-down” clock shows how much time you have left. Although either of those can be fairly easily derived from the other, I think it makes more sense to directly display the time remaining, as this is the player’s primary concern. Therefore, I strongly prefer the count-down method (which, as far as I know, is how virtually all current digital clocks actually operate).

The “retro analog” method is in my opinion worst of all. It directly shows neither time used nor time remaining – it just shows “a time”. You would have to mentally compute the information you want by arithmetic involving the current time vs. the starting time and/or the finishing time. This would actually be easier with a real analog clock, due to the graphic nature of the display (a clock face vs. just numbers).

15

On further thought, I would like to revise/amend the above response. I no longer think that #3 and #4 are equally good. With #3, a player who has not yet made the prescribed number of moves may have to calculate how much time he has left to do so, rather than being able to see it directly. I would prefer that he be able to see it directly. Therefore #4 is preferable to #3, and indeed, preferable to all of the alternatives (including Kevin’s #5 – for reasons explained in my previous post).

16

I dislike how clocks don’t show the seconds. It can cause unnecessary confusion, although small. As an aside, I once had a situation where a player with no opponent had a DGT NA in a 90 minute game. When the hour was up he paused the clock and called me over. There was 00:30 showing on the clock. I told him time wasn’t up yet. You need to wait until it says 00:29. I even showed him how there were still 50 seconds left. Then, during that remaining minute his opponent finally showed up and the game continued. If it wasn’t for prematurely pausing the clock, it probably would’ve been a forfeit.

17

I agree that a standard flag is desirable; obviously analog didn’t have that, the best it could do was an hour’s end, and then many/most adopted the custom of using 6:00 for the first time control.

Perhaps there is an argument to be made that the standard should be the “anti-countup” method. That is, assume a clock is counting up, that there is a cumulative decrement, etc. Subtract that from the target time control per period (as if an analog clock is running.) The delta should be the amount displayed on a digital clock counting down with increment and multiple time controls.

18

Option 1, automatically executed by the clock, when the move counter reaches the prescribed number of moves.

This procedure is accurate 100 percent of the time when the players operate the clock properly, and we should presume players will do so and charge them with responsibility for fixing when they don’t do so.

19

I agree

20

As a player, I also prefer this method. I have never had a problem with an inaccurate move count adding the time prematurely or post-maturely.

It requires, though, that at least one player keep an eye on clock presses. If my opponent forgets to press his clock, I might remind him, at least the first 2 or 3 times he forgets. After that, I sort of feel he deserves to lose out on some of his time. When that happens, and I finally make my move anyway without waiting any further, I will make my move, then press first his clock and then mine, in rapid succession. This restores both the move count and the increment. Nobody has ever complained when I do this, probably because, as a habitual clock-press forgetter, they don’t even notice. And if they did complain, they probably would have no reasonable grounds anyway.

Now if two habitual clock-press forgetters were to be paired against each other, that could be a problem, I suppose.

Bill Smythe