lichess.org
Donate

[Feature request ] ⅛+0 time control

@Chesstroll_Berserk For what it's worth, prior to introducing Ultrabullet, Lichess staff disagreed on its merits.

I agree: "Also it will demonstrate how accurately Lichess lag compensation works."
Fantastic idea... Can't wait for the maniacs who berserk every game, (Bumper and Berserk only) to tilt everyone (((:
I would love to see ½ and ¼ second time increments! :-)

I think it's quite obvious that these could be very useful in fast and ultra fast games, plus one more thing:

We already have ¼+0, so if we were allowed both to have ⅛ time control and factorial time increments, we could make it even more flexible!

Having ¼ second increment per move gives 7.5 seconds per 30 moves.

So, ⅛+¼ time control game that lasts 30 moves is at least as long as ⅛+0, while it can last even longer, if players keep moving fast enough.

And having +½ seconds increment would be really useful for even more people, both for ultrabullet players and for people that like to play bullet, but not necessarily ultrabullet :-)
I agree with the idea of having more flexibility in the time control.
An argument against it would be that players would need to wait longer to get an opponent, but as long as some time controls are standard (the "quick game" buttons) and some are "semi-standard" (available from the sliders), most players would use them. The odd time control games can be accepted from the lobby.
I propose adding a third tab labelled "non-standard TC" for options like 7/8+23 as well as delay, hourglass, time for X moves, etcetera (those would need to be fitted into a speed category).
Lastly, when issuing a challenge to a specific player, I see no reason why you should not be able to choose as strange a time control as you want.
Playing on other websites or in real world, we usually see these three different time controls or their combinations:

--> Time for a whole player's game

--> Time increment per move

--> Time buffer for a move.

About the third one, there are two variations, either you are forced to move within allocated time (like 1 minute per move versus AlphaZero?) or where the time for a game doesn't start to run, unless you go over the time buffer, like in 4-players chess on www.chess.com (X seconds for a move, and only after this the time for a game starts to run away).

Maybe lichess could have all 3 types of time control too?

Time buffer is very nice and useful, as you can always move within these few seconds, without worrying about losing any time. It's fundamentally different from time increments, as this time is independent and it doesn't stack, so you could play for example ¼+0 or ⅛+0 with 1 second per move buffer, without losing any time at all, unless you go over this 1 second!

I think that it could be really interesting option, especially for a bullet, which could also be used for other purposes, like as recompensation for a lag or for long time controls to reduce time increments and make games shorter, while still allowing for a quick moves in the endgame.

This way you could even play 0+0 games, with only condition being that every move would must to me made within the time allocated, for example within this 1 or 2 seconds :-)

There is another reason to introduce delay time control:
Many over-the-board tournaments use a delay based time control (30+5d in the ones I play in) so introducing delays would allow players to practice under the same conditions as their tournament games, which could bring more players to lichess and provide them more benefit.
<Comment deleted by user>
The system currently matches games with exact time controls, but maybe in the future, similar times could be also matched?

Like if I want to play a 1/4 + 1 second increment, and another wants a 1/4 + 1.5 sec, maybe we could be paired if waited for a long time?
And at what time control would you play at after you are paired?
Most people expect the game to have the exact time control they selected. If someone chooses 1/4+1 and gets 1/4+1.5 when the game starts...they should at least be able to opt out.

This topic has been archived and can no longer be replied to.