No, no. Please do not deprive the calculators their use of all their time and then playing on the delay. Ever seen GMs Walter Browne or Alexander Ivanov play a dozen moves on the delay and deliver mate? And a number of lesser players can do the same. It is all part of playing well and players should not be pampered by increasing the delay or using increment.
Note that this is not a problem with clocks that follow FIDE rules for delay. Under FIDE rules, the display shows the time available to complete the player’s next move, i.e., the sum of base time plus remaining delay time. So, for example, if the time control is G/30 with a 5 second delay, the clock initially displays 30:05, and operates in Bronstein mode. That is, it starts counting down when started, and when you move it stops and adds the minimum of 5 seconds or the time you took back to the displayed time.
This is exactly equivalent to starting the clock at 30:00 and using a simple 5 second delay, in the sense that after any sequence of move times, you will have the same time remaining until flag fall in both systems.
The only difference is that under the FIDE approach the clock display is actually useful at all times and understandable at a glance.
What is the rationale for encouraging simple delay in the US instead? Are there many tournaments using delay that would not allow a clock that follows the FIDE rules?
–Tim Smith
It may be useful to note that the DGT NA clock was developed precisely because Bronstein != delay. They are closely related, but not exactly the same.
I’ve heard that argument before, but never heard it satisfactorily explained. Can you please explain how they are different?
Alex Relyea
I believe that Bronstein and delay are exactly equivalent provided the clock (in Bronstein mode) adds the delay time to both sides at the start of the game. (In the example cited of G/30 d/5, the clock should be set for 0 hours 30 minutes 0 seconds for each side, and when White’s clock is started, both sides should start at 00:30:05.) The Chronos (“adagio” mode, if I remember correctly) implements this correctly but has a curious bug with multiple time controls. When the time in the first control runs out, the clock starts the time from the second control and incorrectly also adds the delay time again. (I have not verified whether the clock does this in later controls if there are more than two controls.)
The explanation is straightforward. Bronstein mode adds time to the base time at the completion of each move. Delay mode never increases the base time. If you own a DGT NA, you should be able to compare and contrast the two on the same clock.
Carol Jarecki has taken part in a number of clock tests for DGT. She told me a few years back that she eventually convinced Albert Vasse to buy a Chronos and compare it to the DGT 2010 or DGT XL in Bronstein mode so he could see the difference. The result was that Albert developed the DGT NA, which is really just the DGT 2010 with delay modes added.
Mathematically, Bronstein and delay are equivalent: a properly set Bronstein clock will flag at the same time as a properly set delay clock, ceteris paribus.
Psychologically, they are not, and I much prefer the separate delay timer for a variety of reasons. Your mileage may vary.
As far as I can tell Bronstein delay is functionally the same as simple delay, with one slight difference.
Example: time control is one minute with a five second delay.
With 1:00 remaining, I wait one minute and one second to make a move and then hit my clock.
Under simple delay, during the first :05 of that 1:01, my time would not move, so I would have four seconds left when I hit my clock.
Under Bronstein delay, my time starts ticking down immediately when it is my move, and since I only have one minute, I would lose on time before I hit my clock.
So with Bronstein, each player has whatever the delay is (5 seconds, 10 seconds, etc.) LESS than they would have under simple delay. As far as I can tell, that is the only functional difference.
It is also important to remember that Bronstein delay is not the same as Fischer time (increment), which is much different from the other two.
Please review this post (and following) for an explanation of how Bronstein and delay are mathematically equivalent.
The difference is psychological only.
Some players prefer that a single display tell them exactly how much time is left. These players prefer Bronstein.
Others prefer to know exactly how much delay time they have left, so they can press their clocks just before the delay counts down to zero, to optimize their use of time. These players prefer straight delay.
It’s a personal preference, nothing more.
It seems likely that a player who grew up on increment might prefer Bronstein over delay, while someone who grew up on analog clocks might prefer delay over Bronstein.
Bill Smythe
I am near-certain that the DGT NA supports delay but not Bronstein. (while the DGT 2010 is the opposite) I no longer have a DGT NA on hand, but as I recall the addition of straight delay capability was the main reason the clock came to be, as noted elsewhere in this thread. I would guess the Design of the Clock-maker was “delay for USA” (NA) and Bronstein for everywhere else. (2010)
As soon as Mike Atkins finds this thread, you will hear plenty. The debate is over what happens when a player has less than five seconds of main clock time. Some say that straight delay allows players a full five seconds added to the main clock time—for a minimum of six seconds no matter what—whilst Bronstein as implemented by some clocks does not allow it.
That is the biggie. The addition of five seconds at the first clock press is also needed to make Bronstein exactly equal to straight delay, but to me is not a huge deal for Regular/Classical games.
I knew a guy who gave up using his Saitek Pro (or the version of the silver Saitek prior to the Pro, maybe) for the reason Mike Atkins preaches. My friend swore he did not get a full x seconds at the end of a tense game when in dire time pressure, while in Bronstein mode.
Might be interesting to check all digital clocks to see how they handle it. Also check archives here for prior threads on this topic, maybe.
I didn’t discuss mathematical or functional equivalence for a reason. Two things can be functionally or mathematically equivalent, but still not be identical.
The first sentence above is correct. DGT NA (“NA” = “North American”) scrapped the Bronstein modes, along with a few other timer modes for go and other games. I was in a bit of a hurry with that last post.
That reminds me of one area where a simple delay would have a substantial advantage over Bronstein with pre-added time or over increment: it would be reasonably simple and straightforward to build analog clocks that support simple delay, at least for battery powered analog clocks where clock start/stop can be done by switching power to the individual clock units.
I’d welcome an analog delay clock. I’ve downloaded and read the manuals for all major digital clocks, and after seeing how badly written most of them are, I have pretty much no confidence if using a clock other than mine that my opponent has actually set the thing right. And even if he has set it right, there are so many differences between the various clock displays that I have no confidence I’ll actually know how to read the time during all phases of all time periods, and know how to interpret flags or other indicators.
Analog clocks were at least simple. You could tell they were set right and understand the flag even if you had never seen that model before. (Oops…have to cut this short. Gotta chase some kids of the lawn…)
–Tim Smith
Which will never happen in Bronstein, because in order for a clock to show five seconds or less after a move, it would have to have reached zero on the prior move.
Agreed that the delay time has to be added before move 1 to make Bronstein mathematically equivalent to delay. And agreed that even then, there’s a significant psychological inequivalence.
It’s been thought of. See figure one of US Patent 5,420,830.
Eric should have just e-mailed me but, to avoid feeling like a puppet I’ll just provide one example on how they differ and mention the experience of using the two formats:
Experience - This is the psychological perception of time. Both formats result in the same thing, since Bronstein adds and Delay waits and this works until you have 6 seconds left in Bronstein mode (see example below). From this perspective it really doesn’t matter that 1 or 2 seconds were added to the Bronstein clock at the start. A player isn’t going to be thinking “Hmmm, all is well, I have an extra second that was given to me 4-5-6 hours ago!” No, he is going to experience having 5 fewer seconds with which to think. As soon as the player uses more than 5 seconds, the two formats are not equal when there is six seconds left on the clock from a perception and experience of time. Look at the example and tell me if they are REALLY the same from a perceptual experience, without using TD Speak to say they are equivalent…
Example. You are playing a game and have 13 seconds left on your clock, losing 1 second for each step of the example. You use six seconds before you make your move and this is what you have left after making the move, Delay on left, Bronstein on right:
Delay Bronstein
13 12 12
12 11 11
11 10 10
10 9 9
9 8 8
8 7 7
7 6 6
6 5 Forfeit
5 4 Forfeit (but this will never exist in Bronstein)
4 3 Forfeit (but this will never exist in Bronstein)
3 2 Forfeit (but this will never exist in Bronstein)
2 1 Forfeit (but this will never exist in Bronstein)
1 Forfeit (but this will never exist in Bronstein)
Why would there be one or two seconds added to the clock at the start? Shouldn’t there be five extra seconds (in your example) added at the start? In Bronstein mode the clock will display the exact amount of time remaining for the player to make his (next) move. With a delay, that is counted down somewhere separately. This seems to be a lot more confusing to me.
If your clock shows six seconds, that means something different depending on whether you are using delay or Bronstein. If you are using Bronstein it means that you have six seconds to move. If you are using delay it means that you have six seconds PLUS some indeterminable time until your clock will start. If it is your opponent’s move it is fairly certain that that time is the standard delay time, but if it is your move, it is anyone’s guess.
Alex Relyea
Don’t you think that six seconds plus “some indeterminable time” (5 or 10 seconds) is more time than having only six seconds period on the clock? I was using adding 1 second to the Bronstein because it was mentioned earlier as cure for later time differences. I don’t think it matters if you add 5 seconds or 5 minutes as that time will disappear during the game. When you are down to six seconds you are down to six seconds. The two formats are very different in how the game ends, no matter what is done earlier to compensate for the difference. So they are equivalent but not the same, and the psychological perception and experience as the game ends seems more important than any supposed mathematical equivalence. But, in the end, we are talking opinions which is why we can differ and disagree.
Mike
Side by side comparison video of Bronstein vs. Delay