Your network blocks the Lichess assets!

lichess.org
Donate

Scientific analysis of lag compensation on lichess and chess-dot-com.

One factor I didn't see was that chess-com allows stacked premoves. So a player might make 5 pawn premoves, promote, then a 7-move ladder-mate, all premoved. This would take 1.2 seconds on the clock. In Lichess it would be very challenging to complete this sequence in 1.2 seconds, even with zero-time premoves, since only one premove is active (additional premove attempts replace any existing premove).

But the whole issue of lag compensation wasn't one I was familiar with, so thank you for documenting this. I feel better about playing bullet on Lichess now, despite it's slower ping times from the US.

One factor I didn't see was that chess-com allows stacked premoves. So a player might make 5 pawn premoves, promote, then a 7-move ladder-mate, all premoved. This would take 1.2 seconds on the clock. In Lichess it would be very challenging to complete this sequence in 1.2 seconds, even with zero-time premoves, since only one premove is active (additional premove attempts replace any existing premove). But the whole issue of lag compensation wasn't one I was familiar with, so thank you for documenting this. I feel better about playing bullet on Lichess now, despite it's slower ping times from the US.

@djconnel said ^

One factor I didn't see was that chess-com allows stacked premoves. So a player might make 5 pawn premoves, promote, then a 7-move ladder-mate, all premoved. This would take 1.2 seconds on the clock. In Lichess it would be very challenging to complete this sequence in 1.2 seconds, even with zero-time premoves, since only one premove is active (additional premove attempts replace any existing premove).

But the whole issue of lag compensation wasn't one I was familiar with, so thank you for documenting this. I feel better about playing bullet on Lichess now, despite it's slower ping times from the US.

This to me is what makes chess.com feel better in shorter time controls. Although the site itself is a cesspool, so I never use it anymore. I've just learned to accept that I'll lose some bullet games to it here and there on Lichess

@djconnel said [^](/forum/redirect/post/wNKLE8vL) > One factor I didn't see was that chess-com allows stacked premoves. So a player might make 5 pawn premoves, promote, then a 7-move ladder-mate, all premoved. This would take 1.2 seconds on the clock. In Lichess it would be very challenging to complete this sequence in 1.2 seconds, even with zero-time premoves, since only one premove is active (additional premove attempts replace any existing premove). > > But the whole issue of lag compensation wasn't one I was familiar with, so thank you for documenting this. I feel better about playing bullet on Lichess now, despite it's slower ping times from the US. This to me is what makes chess.com feel better in shorter time controls. Although the site itself is a cesspool, so I never use it anymore. I've just learned to accept that I'll lose some bullet games to it here and there on Lichess

Interesting, however, i use an adblocker
For the programming beginners can you show the python script that you used for the mouse clicks and stuff

Interesting, however, i use an adblocker For the programming beginners can you show the python script that you used for the mouse clicks and stuff

@cxherdge said ^

Interesting, however, i use an adblocker
For the programming beginners can you show the python script that you used for the mouse clicks and stuff

I took a module called 'mouse' and used mouse.is_pressed() in an infinite While loop.
It was in fact very buggy and not user friendly, and a better programmer sure would have found a proper solution, but it worked

@cxherdge said [^](/forum/redirect/post/xxt8LEvf) > Interesting, however, i use an adblocker > For the programming beginners can you show the python script that you used for the mouse clicks and stuff I took a module called 'mouse' and used mouse.is_pressed() in an infinite While loop. It was in fact very buggy and not user friendly, and a better programmer sure would have found a proper solution, but it worked

A few remarks and questions:

  1. "speed of light being at a stable 300.000 km/s" this is in vacuum, in optical fiber this is about 1.5 times slower due to the refractive index of the glass
  2. Which browser did you use?
  3. Do you think behavior is different for Chrome, Firefox, Opera, Proctor, Brave, ...?
  4. Most annoying for me was always the instantaneos lag spikes that occured. Once I had 1st round a game where opponents move did not arrive, reloading the page took 30s and when I could move 1:20 of 3 minutes were gone. Several time lag spikes divided a single move like Be2 to a6 into a part and a premove (1. Be2-c4, 2. Bc4-a6). Unfortunately pieces got capture after the first move ... Btw, I have premium at chesscom, so ads cannot be the reason. These experiences are from the pre-Proctor era.
A few remarks and questions: 1) "speed of light being at a stable 300.000 km/s" this is in vacuum, in optical fiber this is about 1.5 times slower due to the refractive index of the glass 2) Which browser did you use? 3) Do you think behavior is different for Chrome, Firefox, Opera, Proctor, Brave, ...? 4) Most annoying for me was always the instantaneos lag spikes that occured. Once I had 1st round a game where opponents move did not arrive, reloading the page took 30s and when I could move 1:20 of 3 minutes were gone. Several time lag spikes divided a single move like Be2 to a6 into a part and a premove (1. Be2-c4, 2. Bc4-a6). Unfortunately pieces got capture after the first move ... Btw, I have premium at chesscom, so ads cannot be the reason. These experiences are from the pre-Proctor era.

Also I have just lost the ability to see how many bars people have when facing them or even when they are just playing. Is this just a me issue? Or does everyone have this issue.

Also I have just lost the ability to see how many bars people have when facing them or even when they are just playing. Is this just a me issue? Or does everyone have this issue.

@DaBassie said ^

Interesting, however, i use an adblocker
For the programming beginners can you show the python script that you used for the mouse clicks and stuff

I took a module called 'mouse' and used mouse.is_pressed() in an infinite While loop.
It was in fact very buggy and not user friendly, and a better programmer sure would have found a proper solution, but it worked

Oh, okay

@DaBassie said [^](/forum/redirect/post/coFS58KU) > > Interesting, however, i use an adblocker > > For the programming beginners can you show the python script that you used for the mouse clicks and stuff > > I took a module called 'mouse' and used mouse.is_pressed() in an infinite While loop. > It was in fact very buggy and not user friendly, and a better programmer sure would have found a proper solution, but it worked Oh, okay

Also another interesting topic I would like investigated is new device comp. Because at least from my experience. Whenever I use a new device that lichess has never seen my time adv at the end of bullet games is always 10-15 seconds greater for the first day or so. It is quite a strange phenomenon that I think should be repaired as it is exploitable in money tournaments.

Also another interesting topic I would like investigated is new device comp. Because at least from my experience. Whenever I use a new device that lichess has never seen my time adv at the end of bullet games is always 10-15 seconds greater for the first day or so. It is quite a strange phenomenon that I think should be repaired as it is exploitable in money tournaments.

I think the research focused a lot in the lag compensation, which is fair, since in the title itself it says it's about the «analysis of lag compensation». However, the questions left by ElyneLee were more broad, and lag compensation was just a very technical aspec of the difference between the two platforms.

Certainly the lag compensation plays a role (especially for bullet games), but I think the main difference between Lichess and chess-dot-com is the social aspect, the emotional engagement, and the visuals of each platform. Even for cold, calculating chess players. That's why ElyneLee found a big difference in the playing styles and in the openings between both platforms. I think the difference of this aspect is so big, that makes almost negligible the effect of how lag compensation affects the game.

Many people just play in the platform "where their friends play". This creates two different networks of people, and each of them creates a different environment from the human aspect alone. Finding out what this means might mean doing another research though!

I think the research focused a lot in the lag compensation, which is fair, since in the title itself it says it's about the «analysis of lag compensation». However, the questions left by ElyneLee were more broad, and lag compensation was just a very technical aspec of the difference between the two platforms. Certainly the lag compensation plays a role (especially for bullet games), but I think the main difference between Lichess and chess-dot-com is the social aspect, the emotional engagement, and the visuals of each platform. Even for cold, calculating chess players. That's why ElyneLee found a big difference in the playing styles and in the openings between both platforms. I think the difference of this aspect is so big, that makes almost negligible the effect of how lag compensation affects the game. Many people just play in the platform "where their friends play". This creates two different networks of people, and each of them creates a different environment from the human aspect alone. Finding out what this means might mean doing another research though!