Digital clocks that pass Smythe Dakota muster

Spinning off from a previous thread, I too have some desiderata, very different from those of FIDE. As far as I know, no current clock even comes close:

  • Each clock face should have three separate display areas: main time, delay time, and move count. All three should be visible simultaneously.
  • The main time should have a full 5-digit display: h:mm:ss. Leading zeroes in the hours and minutes positions should be suppressed, along with the colon between them. A leading zero in the seconds position should not be suppressed, nor should the colon.
  • Delay time should have a 3-digit display: m:ss. The minutes and colon should be suppressed when the delay goes below 1 minute, and the first digit of the seconds should be suppressed under 10 seconds. However, when the delay goes to 0, the zero should continue to be displayed.
  • Move count should have a 3-digit display: nnn. The hundreds digit, but not the tens digit, should be suppressed if zero.

I envision the display as consisting of a single computer-like screen (not segmented digits), about 7 inches across by 2 inches deep:

-[size=200]1:45:28 @@1:39:57 [/size]
[size=150] 5 @@@@ [/size][size=90] 17 @@@- 16 [/size][size=150] @@@@ 0 [/size]

That’s not a perfect rendering, but it’s the best I could do within phpBB limitations. Those @ symbols are supposed to be invisible – I just put them in for spacing. My apologies if the rendering is even less satisfactory on your browser than on mine.

The digits for the main time should be about 3/4 inch high, as on a Chronos or DGT North American. The delay digit(s) should be smaller, say 3/8 inch. The move-count digits should be smaller still, say 3/16 inch.

Imagine, also, a thick vertical line down the middle, top to bottom, to separate white’s side of the display from black’s. (No way I was going to figure out how to represent that in phpBB.)

The above sample display indicates that the players’ main times are 1:45:28 and 1:39:57. White has made 17 moves, black 16. White has a 5-second delay waiting for him at his next move, while black has used up his 5-second delay for his current move.

The clock should render each player’s main time as follows:

h:mm:ss if one hour or more;
mm:ss if between 10 minutes and 59:59 inclusive;
m:ss if between 1 minute and 9:59 inclusive;
:ss if under one minute.

The main time should be centered on each player’s display. When the time ticks down from 1:00:00 to 59:59, the hours digit and colon should disappear, and the remaining digits should jump to the left a bit so as to keep the display centered. Similar leftward jumps should occur when the time ticks from 10:00 to 9:59, and from 1:00 to :59.

The clock should render delay times as follows:

m:ss if 1 minute or more;
ss if between 10 and 59 seconds;
s if 9 seconds or less.

The delay time should appear at the outside edges of the display – at the left edge for the player on the left (usually white), at the right edge for the player on the right (usually black). When the delay time ticks down from 1 minute to 59 seconds, the minutes digit and colon should disappear, and the remaining digits should remain left-justified on the left display, right-justified on the right display. Thus, on the left side, the remaining digits should jump to the left, but on the right side, they should stay where they are. Similar re-justification should occur when the delay time ticks from 10 seconds to 9.

The clock should render the move count as follows:

nn if 99 moves or less;
nnn if 100 moves or more.

The move count should be displayed at the inside edges of the display – at the right edge for the player on the left, at the left edge for the player on the right. When the move count is below 10, it should nevertheless be displayed as two digits, with a leading zero. At move 100, a third digit should appear, with the move count remaining right-justified on the left display, left-justified on the right display.

I like the above arrangement because of the symmetry. And there is a natural mnemonic: The move counts are always within 1 of each other, so are close together on the display. The delay times are almost always far apart, so are at opposite edges of the display.

The different digit sizes, and diverse display conventions, should quickly foster an intuitive grasp of which digits serve which purposes. Main times are always large, and are displayed with a colon and at least two digits. Delay times are medium-size, and (if under 10 seconds) are displayed as a single digit with no colon. The move count is small, and is virtually always two digits, with a leading zero if necessary.

Other points:

  • If a delay is in effect, the delay digit should always be displayed, even if zero at the moment. If no delay is in effect, or if there is an increment, the area of the display normally used for delay seconds should remain blank. Any increment is simply added to the main time, as usual. This convention allows an arbiter or spectator to quickly determine whether the delay has been set.
  • The move-count display should have the following options: (a) move count not displayed; (b) move count displayed; (c) move count displayed and triggers the subsequent time control.

Bill Smythe

This is quite interesting, and in general I like the idea. But I’m not sure I see where you describe how you indicate a player has moved on to a second time control (e.g., 40/90 SD/30, after the player has used up the first 90 minutes).

I’m not entirely sure I like the idea of the position of the main time display shifting as hours and then leading digits for minutes drop out, though.

I think there should be a fourth display area on each clock face. It could be underneath the delay and move-count areas, and centered horizontally between them. It could read “1st”, “2nd”, etc.

Whether the 2nd control would be triggered by the expiration of time in the first control, or by the move count, would depend which option was chosen. It would be the former if option (a), “move count not displayed”, or option (b), “move count displayed”, had been chosen. It would be the latter if option (c), “move count displayed and triggers the subsequent time control”, had been chosen.

I wouldn’t want the clock to force players to use the move counter, or to rely on it, because many are opposed, and some players can never seem to keep the move counter accurate. On the other hand, some players like the move counter, and it’s nice to have the clock add the secondary time automatically at the appropriate move number.

It would happen only three times per player per game, at most. And it would keep the main time nicely centered in the display.

Various models of the Chronos already shift the main time, in some way, in some modes. For example, in the move-count modes, the display changes from h:mm to m:ss at the ten-minute mark.

Thanks for your interest, and your apparent approval of at least some of the ideas. :slight_smile:

Bill Smythe

I like the general format except for the move counter. I’ve become a strong believer that a move counter on a clock only causes problems. At best it matches what is on your score sheet. At worst it misleads you.
Mike

I agree that the move* counter is the cause of far more trouble than it solves. Exhibit A would be the last round of the USATE where the reliance on it may have decided the tournament winner.

  • more accurately clock press counter

Well, under my proposal there would be an option to shut off the move counter. Then the second control would be triggered by the expiration of the first control, regardless of the move count.

Bill Smythe

Bill,
But doesn’t having a clock press counter just ask for problems? Given what we know today would we really design a new chess clock with a clock press counter?
Mike

It solves problems, too. The extra time is added after move 40 (or whatever the first control is), rather than whenever the first control expires, which could be at move 45, 50, 55, 60, etc.

If you’re down to a few seconds when you play your 40th, it’s comforting to see the clock immediately add the extra time when you press your clock. You don’t have to worry about bogus time forfeit claims by your opponent, requiring a ruling by a TD who may or may not know what he’s doing, and who may make the wrong ruling.

The use of a move counter requires that at least one of the two players be observant enough to notice irregularities, such as the opponent forgetting to press his clock. If you’re not observant enough, and neither is your opponent, then I agree with you, you shouldn’t be using the move counter.

Bill Smythe