Asbjørn Steinskog

3.1K posts

Asbjørn Steinskog banner
Asbjørn Steinskog

Asbjørn Steinskog

@Assios

Dev @TakeTakeTakeApp · mod & volunteer @lichess

Oslo, Norway Tham gia Temmuz 2009
552 Đang theo dõi1.4K Người theo dõi
Vidit Gujrathi
Vidit Gujrathi@viditchess·
Thinking of moving to Linux permanently... Really liking it so far.
English
166
22
1.8K
134.1K
Asbjørn Steinskog
Asbjørn Steinskog@Assios·
@DavidHowellGM Haven’t seen the UK version, but based on the Norwegian: The game design is actually pretty terrible! It punishes you for being right. I’d rather watch a show where celebrities just play Avalon
English
1
0
3
322
David Howell
David Howell@DavidHowellGM·
I love Traitors & all social deduction games. But UK series 4 (+ Celebrity) made it clear a revamp is needed. The Faithfuls too concerned with self-preservation to even try and figure things out. Maybe cast people who actually want to be Faithful? Too easy for the traitors atm 😪
English
8
2
30
7.2K
David Howell
David Howell@DavidHowellGM·
Commentator of the year!! 😊😊Thank you to everyone who made this possible. Especially all my fellow commentators who work insanely hard & do a fantastic job. They've taught me so much. 2025 was a crazy & busy year. I'll remember it for many reasons. This one's for you, Danya 🎙
David Howell tweet media
English
66
24
929
27.7K
Asbjørn Steinskog
Asbjørn Steinskog@Assios·
@viditchess And this is arguably even worse in the chess world! Chesscom is the king of dark patterns, and also shady tactics to suppress competition (like hidden exclusivity agreements to keep streamers off Lichess)
English
1
0
15
3.4K
Vidit Gujrathi
Vidit Gujrathi@viditchess·
Spotify has so many dark patterns. wild how strategically manipulative design choices still scale to billion dollar businesses. personally, once trust erodes, I stop engaging..
English
47
17
726
125.9K
Asbjørn Steinskog
Asbjørn Steinskog@Assios·
@viditchess Yeah, it's based on a Lichess rapid rating of 2300+, so it works well for that purpose. You could refit the eval -> win% curve on games from the world rapid pool, but then other sources of error like quick draws probably dominate anyway
English
1
0
6
1.1K
Vidit Gujrathi
Vidit Gujrathi@viditchess·
@Assios Yes that's a better way, although it has limitations too. The curve is based on "2300" rated players. So that's why even +2.5 or +3 is given roughly only about 71-75% winning chances.
English
1
0
30
3.1K
Vidit Gujrathi
Vidit Gujrathi@viditchess·
The problem with drawing statistics from only centipawns eval is that it leads to wrong conclusions. For eg: A shift from +3.5 to +2.5 doesn't change the game much, it is still clearly winning, but a swing from +0.5 to -0.5 changes the entire nature of the game and both scenarios are given the same weightage. Also, Artemiev played many quick 10move draws, hence his accuracy is going to be highest.
Mehmet Mars Seven 🐴@MehmetMars7

Massive analysis from World Rapid Championship: Stats from more than 150k moves in 1500 games! 🥇Top TPR: Carlsen 🥇Most accurate: Artemiev Most accurate White: Villagra Most accurate Black: Artemiev

English
19
24
749
117.6K
Anish Giri
Anish Giri@anishgiri·
@Assios @lichess @chesscom I have good compi and good wifi, I am almost certain it is some sort of a bug happening.🧐 Please try it out on different devices, maybe you encounter the same. It usually happens when you make a move that wasn't one of the preloaded lines. Similar issues in the app, by the way!
English
1
0
6
584
Anish Giri
Anish Giri@anishgiri·
On a serious note, I think the play zone websites should investigate how the server-side engine analysis and local hardware interfere with each other. Will be happy if they get it fixed, this is not really the main point of ChessMonitor lol cc @lichess @chesscom
Anish Giri tweet media
Anish Giri@anishgiri

Currently, chessmonitor.com is the only website where I can simply check my finished games with a bug-free engine.😂

English
3
4
80
29.5K
Gastón Franco
Gastón Franco@GatoFranco90·
First in the world to solve today's Advent of Chess! In a somewhat special day to me 😆
Gastón Franco tweet media
English
1
0
2
134
Asbjørn Steinskog
Asbjørn Steinskog@Assios·
Advent of Chess 2025 starts tomorrow! An annual chess problem-solving competition: 24 days of logic-based chess problems. Helpmates, selfmates, proof games, retro puzzles, and a few new categories revealed this year. Every solved problem awards a ticket in the drawing for a chessboard signed by @MagnusCarlsen Create an account and join for free at adventofchess.com
English
0
1
3
543
Pawn
Pawn@pawnona7·
@anishgiri @davidllada well it may be Nf3 e5 Ng1 e4 Nf3 d5 Ng1 which is a crazy opening so...
English
1
0
5
345
Asbjørn Steinskog
Asbjørn Steinskog@Assios·
@durarbayli What an incredibly stupid take. Lichess massively expanded the chess ecosystem, providing tools, APIs, innovation, all which makes it much easier than ever to build new chess projects. So many projects rely on Lichess, including yours.
English
0
1
20
436
Vasif Durarbayli
Vasif Durarbayli@durarbayli·
I know praising Lichess is as popular as ever. On chess Reddit, someone shows up every second Saturday to say Lichess is the best thing that ever happened to chess. I’ve actually contributed to Lichess myself as a streamer and blogger. I stopped because of their political stance – that’s a different story. I’m not pretending Lichess hasn’t done a lot of good. In fact, my own project, ChessEver, wouldn’t have progressed so fast without Lichess existing. Still, if you put the benefits aside for a moment, there’s something fundamentally wrong with the Lichess model: it destroys the chess market. Providing so many features for free makes it very hard for any chess business to charge for its services. If businesses can’t charge, they can’t pay professionals properly. If professionals can’t earn enough, they’re pushed to play in badly conditioned events. Conditions don’t improve for professionals, and that eventually hurts casual fans, too. That’s the vicious cycle. Nothing is truly free. Someone always pays — with money, time, or the opportunities that never get built because it stopped making economic sense to try.
English
203
12
226
191K
Asbjørn Steinskog đã retweet
Vidit Gujrathi
Vidit Gujrathi@viditchess·
Appreciation post for @lichess Such an amazing open source work. Absolute delight to use it. Now, the new mobile app makes it so easy to follow events! And all of this is run by donations and passion of chess lovers! Epic.
English
35
133
2.2K
171.6K
Asbjørn Steinskog
Asbjørn Steinskog@Assios·
@fins0905 Gonna buy this just because it’s not chessable/chesscom. Buying every course going forward that’s not feeding chesscom!
English
0
0
2
180
John Bartholomew
John Bartholomew@fins0905·
Pleased to announce that after 1,000+ hours of work, my course is officially HERE! ♟ The Comprehensive Scandinavian: A Complete Repertoire Based on 3...Qa5 by IM John Bartholomew It features: • 225,000+ words of instruction • 25 Model Games (another 59,000+ words!) • 10 "Repertoire Rumble" videos, where I play the course LIVE • 75 Key Positions • A 6.5+ hour Move Order Guide • Course Game practice vs. the Chessiverse bots This was a true labor of love, and I documented all the time I spent on this course going back to Apr. 2023 here on Twitter/X (x.com/fins0905/statu…). The Scandinavian is an excellent practical weapon against 1.e4, and I would be honored if you checked my course out. This is also the very first course we're releasing on Chessiverse. We've put enormous thought and effort into building innovative and useful features for chess improvers, and we hope to raise the standard for chess courses going forward. Here's a comprehensive overview of the course and the approach we're taking at Chessiverse.
English
20
8
222
21K
Tom Zahavy
Tom Zahavy@TZahavy·
I am excited to share a work we did in the Discovery team at @GoogleDeepMind using RL and generative models to discover creative chess puzzles 🔊♟️♟️ #neurips2025 🎨While strong chess players intuitively recognize the beauty of a position, articulating the precise elements that constitute creativity remains elusive. To address this, we pre-trained generative models on public datasets and then applied reinforcement learning, using novel rewards designed for uniqueness, counter-intuitiveness, realism, and novelty. This approach doubled the number of novel chess puzzles compared to the original training data, while successfully maintaining aesthetic diversity. Three distinguished experts—International Master of chess compositions Amatzia Avni (author of "Creative Chess"), Grandmaster Jonathan Levitt @JonathanLevitt7 (author of "Secrets of Spectacular Chess"), and Grandmaster Matthew Sadler @gmmds (author of "Game Changer")—evaluated and selected the puzzles they found most compelling. Their preference was for puzzles exhibiting original, paradoxical, surprising, and naturally occurring positions, with particular emphasis on those that integrated aesthetic themes in innovative ways and demonstrated exceptional over-the-board vision. 🧩Try to solve the puzzles @chesscom: chess.com/c/2wCTN7Uv2
Tom Zahavy tweet media
English
52
244
1.4K
397K
Anish Giri
Anish Giri@anishgiri·
I am acquiring @chessmonitor and together we are turning it into the one place you need for opening prep. You can already start exploring the platform. Launching an app soon. And then more. 👉chessmonitor.com
English
57
30
481
67.1K