Hey, thanks for the amazing website.
I've been using the analysis tool today and it was displaying normally but after getting a bit greedy and requesting analysis on 2 games at once, the formatting in the window that displays the moves and analysis is all messed up. Each move is displayed on an individual line rather than a set of moves running on one or two lines until a mistake is made, so the whole box only displays about 6 moves.
The board now won't show the green/orange highlights for where the move was made and where the better move should have been. Anyone know if this is punishment for lying when I was 17, as well as being greedy today?
Yes. Did you seriously thought you would get away with that lie?
The analysis page has been changed, and I believe it's better now. The movelist is clearer, the board is finally coherent with all other lichess boards (including 3D), and it's snappier. And I finally got rid of the pgn4web library, which was the oldest piece of JS of the entire site.
I dropped the best move highlight for 2 reasons. First it was confusing for many people. Having 4 highlighted squares on the board did look weird - and sometimes only 3 when actual move and best move were overlapping. Plus, I like to try and find the best move myself. The highlight was giving it away, but now one has to explicitely click on the variation to see it.
Oh hi, thanks for making this awesome website.
But I completely disagree.
I want to quickly analyse my own games after I play them, not spend days trying to work out the best move - I could do that outside of analysis mode. I now have to wade through swaths of irrelevant moves before I come up to something interesting, and for a rookie like me it takes a while to visualise where E3 is, and doing that for FOUR squares simultaneously is exponentially more difficult to picture.
I think it would be great if there was the option to revert back to the old one.
Thanks again
I almost thought Thibault's first sentence (Yes. Did you seriously think you would get away with that lie?) was serious. :O
The main line character is big, and is missing the evaluation change for ex. -0.10 -> -0.60. Still no option for adding more variation.
I' am impress how quick this site is develop. The new "stockfish" is coming fast !!!
Keep up the good work !!!
Thank you very much.
Sorry but I have to agree with the others. It used to be such a nice streamlined way to review your game to simply scroll through the moves without having to take your eyes off the board until you notice a highlight mark alerting you to an inaccuracy/mistake/blunder and then you could look to the right to get more info about it. I mean that was really wonderful and you get used to the multiple highlighted squares fast especially since they are different colors :)
I know that the philosophy of this website is to keep it simple and not offer too many settings options, but perhaps this is one that genuinely deserves to be one.
Get used to it, it's better. Force this change thibault! People won't accept change unless thrusted upon them.
I can honestly say at a glance that anyone who is serious about chess (i.e. anyone who actively has been playing long enough to actually benefit from using computer analysis...that could be 2 months or 20 years, depending on the player!) would consider this a serious improvement! No one wants to have to scroll around the screen to check the graph to see when the advantage is slowly sliding or rising, and they want to see the mistakes in context of the game, not jump quickly to some random point and see which highlight square would have "saved" their position. If I make a huge blunder, I want to quickly run through the whole variation that is listed rather than quickly see a highlighted "best move" and waste time pondering what the computer could have been thinking when it's easier to just go look.
There is a time and place for visualizing moves...trying to understand computer logic is not that time or place, so the "best move highlight" some players are begging should return is actually a feature that gives beginning players bad habits trying to visualize why the computer's "perfect" move is accurate, when they should instead be trying to figure out what led them to the place that they began to slide downhill to begin with, and what might have worked better. I can honestly say, as a player of nearly 20 years, and having been engaged in a much more serious study the last 4 or 5 years, I only understand the computer's "perfect" move sequence about 1 in 10 times...I understand a Grandmaster's move sequence far more regularly. So maybe it'd be better to go analyze some Grandmaster games to improve your games rather than relying on a computer engine (and a limited one at that!) to tell you what the best move was when you blunder. Maybe then you'll come to realize why you blundered, which would be of far more benefit!
This new format is significantly better for analysis, both for quick post-mortem, as well as in depth study. Kudos.
The formatting of the text in the boxes is slightly better now, so thanks for that.
But, and I hate to repeat myself, I just find it funny that literally the same day that I was showing a friend how good the analysis was (UncrownedRoyalty described perfectly why it was so fluid), it's now essentially rendered impractical for me, because the highlighted squares are gone. I can't see how it's expected that the people who couldn't figure out what the highlighted squares mean are supposed to then be intelligent enough to visualise it instead. Also, the squares are still highlighted in the non-analysis game page, so why the discrepancy? Seems really draconian not to just give us the option.
I think you're missing the point.
If all you are doing is scrolling through the game and looking for a highlighted square to jump at you to indicate where you made a mistake, then you aren't actually doing analysis. If you need the computer to tell you "oh, that was the move that really lost the game for you" then you aren't really thinking about the game as it was, you're thinking about the game in the "perfect" computer land. It's better to check the evaluation of each move, and see where maybe you began to slide positionally, long before you blundered that piece, or what brilliant tactic you missed from your opponent.
The computer might show the "best" way to have played accurately, but better is to find those points where you feel you can improve, not simply jump around and expect the computer to flash at you and say "oh, hello master, this is where your mistake was!" It's lazy and it isn't really analysis.
