DEV Community: Chapri

I generate cinematic video ads from a Bun script with the Runway API

Chapri — Thu, 30 Jul 2026 06:53:58 +0000

I needed cinematic 5-second video clips for a chess app's ads. No camera, no actors, no budget for either. So I wired up Runway's API and generated them from a Bun script. Here's the whole thing, including the part the docs gloss over.

The shape of the API

Runway's video generation is async. You don't get a video back from one call. You start a task, then poll it until it's done, then download the result. Three steps, and the polling is where people trip.

The models that matter for text-to-video right now:

gen4.5 at 12 credits/sec. My default. Good enough that a still frame passes for stock footage.
veo3.1 at 20-40 credits/sec. Google's model. Better motion, costs more.
gen4_turbo at 5 credits/sec, but it needs an input image.

At 12 credits/sec a 5-second clip is 60 credits. Credits run about a penny each, so ~60 cents a clip. I generated a dozen clips iterating on prompts and it cost me less than a sandwich.

Auth, and the one thing not to do

The key lives in an env var, RUNWAYML_API_SECRET. Do not pass it as a CLI flag. It ends up in your shell history and in ps output for anyone on the box. Env var or a .env file, nothing else.

Check your balance before you batch anything:

curl -s https://api.dev.runwayml.com/v1/organization \
  -H "Authorization: Bearer $RUNWAYML_API_SECRET" \
  -H "X-Runway-Version: 2024-11-06" | jq .creditBalance

That X-Runway-Version header is required. Leave it off and you get a confusing error that has nothing to do with the actual problem.

What actually worked

For portrait reels I set the ratio to 720:1280. This matters more than the prompt. Generate at 1280:720 and then crop to vertical and you throw away half your pixels and the framing is wrong. Ask for the aspect ratio you're shipping.

The prompts that gave me usable footage were boring and specific:

Cinematic vertical shot of a young woman studying a wooden chessboard by warm window light, focused expression, photorealistic, shallow depth of field, soft film grain

Every word there is doing work. "Cinematic" and "shallow depth of field" get you the blurred background. "Soft film grain" kills the plastic AI sheen. "Photorealistic" keeps it from drifting into illustration. Drop any of them and quality drops with it.

Vague prompts gave me vague, floaty, obviously-generated garbage. The model rewards you for describing a real shot a real DP would set up.

The polling gotcha

The task sits in RUNNING for 60 to 120 seconds for a 5-second gen4.5 clip. Your script has to poll and wait, not fire-and-forget. And gen4.5 has a concurrency limit of 1 on my tier, so if you're generating a batch you run them in series, not with Promise.all. I learned that by watching the second call fail while the first was still cooking.

Rough loop:

POST to start the task, get back a task id
GET the task every few seconds
When status flips to SUCCEEDED, pull the output URL and download it
When it's FAILED, read the reason. SAFETY.INPUT means your prompt tripped moderation, reword it. ASSET.INVALID means a bad input image.

Was it worth it

Yeah. The clips came out looking like real ads. I dropped captions and a voiceover on top with ffmpeg and shipped them as reels. A frustrated player under a desk lamp, a kid in a suit making a move, a woman by a window. None of them are real people. All of them look real enough that nobody scrolling past would guess.

The whole thing runs from a script. I describe a scene, wait two minutes, get a clip. That's a wild place to be compared to a year ago.

If you're building anything that needs video and you don't have a film crew, this is the cheat code. It's the app I made the ads for: aichess.guru, an AI coach that tells you why you lost. Free to start.

Building video pipelines on Runway or Veo or Kling? I want to know what prompts you found that consistently work. Mine still feel like guesswork half the time.

Why LLMs are terrible at chess (and what I did about it)

Chapri — Thu, 30 Jul 2026 06:49:21 +0000

Ask GPT-4 or Claude to play a real game of chess and it will, with total confidence, try to move a knight like a bishop. Or capture its own queen. Or castle out of check. I've watched a frontier model announce "checkmate" on a board where the king had four legal escape squares.

This is the whole problem I ran into building aichess.guru, an AI coach that tells you why your move was bad in plain English. Turns out the "plain English" part is easy. The "not lying to the user about the position" part is where everything breaks.

LLMs don't have a board

Here's the thing people miss. A language model doesn't see a chessboard. It sees a string like rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR. It has read millions of games in that notation, so it can autocomplete plausible-looking moves. Plausible is not legal. Plausible is not correct.

The failure mode is nasty because it's confident. The model won't say "I'm not sure where the rook is." It'll tell you the rook on a1 forks your queen and king when there is no rook on a1. A beginner reading that has no way to know it's nonsense. That's worse than saying nothing.

So rule one for me became: the LLM is never allowed to decide anything about chess.

Split the job in two

Every chess claim in the product comes from real software. Not the model.

Legality, checks, checkmate, what piece is where: chess.js. It's a boring, deterministic library. game.move("Nf3") returns null if the move is illegal. game.isCheckmate() is a fact, not a vibe.
Evaluation, best move, "how bad was this blunder": a real engine (Stockfish). It gives you a number. -3.4 means you're down a piece and change. It doesn't hallucinate the number.

The LLM only ever gets handed verified facts and asked to phrase them like a patient human coach. "You dropped your knight on f5 for free" instead of "eval swung from +0.2 to -3.1." The model is a translation layer, nothing more.

I have a hard rule in the codebase, written in a comment so I don't forget it at 2am: AI generates feelings, never chess. If a number appears in the output, an engine produced it. If a board appears, chess.js rendered it from a legal move sequence. The model gets zero authority over the game state.

The part that surprised me

I assumed the hard bit would be prompt engineering the explanations. It wasn't. Give a decent model the real eval, the real best move, and the real refutation line, and it writes a genuinely good explanation on the first try. Models are great at "explain this fact simply." They're terrible at "produce this fact."

The hard bit was FEN plumbing and trusting nothing. Every position that gets shown to a user is computed from a starting position plus a list of moves, each one validated. I don't hand-write FENs anymore. I did once, shipped a smothered-mate example with the rook and king swapped, and it rendered a perfectly legal-looking board that was the wrong position. Both parse. Only one is the mate you're talking about. That bug is why the "compute, never type" rule exists.

Why this matters beyond chess

Chess is just a clean test case for a pattern that shows up everywhere you point an LLM at a domain with actual rules. Law. Medicine. Accounting. Anything with a source of truth.

The move that works: let the deterministic system own correctness, let the model own communication. Don't ask the model to be the calculator. Ask it to be the friend who explains what the calculator said.

People keep trying to make the LLM do the whole job because it looks like it can. It'll happily produce an answer. It just won't tell you when the answer is invented.

Try it

The product is live and free to start: aichess.guru. Play a game, blunder something, and it'll tell you exactly where it went wrong in words a human understands, with the engine backing every claim.

If you've built LLM stuff on top of a rules engine and hit different walls, I want to hear about it. What did you let the model own, and what did you rip out of its hands?

Stuck at 1200? Try This Chess Training Routine for Beginners (No Burnout)

Chapri — Thu, 30 Jul 2026 06:36:07 +0000

You want to get better at chess. You've tried grinding rapid games, watching YouTube, and maybe even solving a few puzzles. But your rating is stuck. You feel like you're spinning your wheels. The problem isn't you. It's your training routine. Most beginners do too much random stuff and not enough focused work. I'm going to give you a simple weekly plan that actually works. It's realistic, it won't burn you out, and it targets the three things that matter most for your level: tactics, understanding your own games, and basic endgames.

Spent 2 hours on puzzles… Still blundered a rook in 5 moves

Why Most Training Plans Fail Beginners

The biggest mistake is doing too much. You try to study openings, endgames, tactics, and strategy all at once. You end up overwhelmed and quit after two weeks. Or you do the opposite: you only play blitz and wonder why you don't improve. Playing without analysis is like practicing free throws blindfolded. You need a plan that fits your life and your brain.

Another trap is using random puzzles without a system. You solve 50 puzzles a day but never review the ones you got wrong. The pattern doesn't stick. Or you watch a 30-minute video on the Catalan and forget everything by the next game. Your brain needs repetition and structure. That's what this plan gives you.

Your Weekly Chess Training Routine

Here is the plan. It takes about 30-45 minutes on weekdays and a bit longer on weekends. Monday through Friday: do 15-20 minutes of tactics. Use a puzzle app or website that gives you a mix of motifs. Don't just guess. Take your time. Try to calculate before you move. If you get it wrong, understand why. That's the key.

On Saturday, play one slow game. I mean at least 15+10 or longer. No blitz. No bullet. Play like you mean it. After the game, analyze it without an engine first. Find your mistakes. Then turn on the engine to check. Write down one thing you learned. On Sunday, study endgames for 30 minutes. Focus on basic checkmates, king and pawn endgames, and rook endgames. That's it.

Tactics: The Daily Bread (15-20 Minutes)

Tactics are the most important thing for beginners. Most games under 1500 are decided by a simple fork, pin, or skewer. If you can't see those, you'll lose. But here's the thing: quality over quantity. Doing 10 puzzles with deep thought is better than 50 rushed ones. I want you to try to calculate every line before you move. If you get it wrong, figure out why you missed it.

Set a timer for 15 minutes. Solve as many as you can, but don't sacrifice accuracy for speed. Use a site that shows the solution after each puzzle. If you got it right, great. If not, replay the winning idea in your head. Over time, you'll start seeing patterns automatically. That's when your rating jumps.

The One Slow Game: Your Personal Case Study

Once a week, you play one serious game. No distractions. After the game, you analyze it. This is where most beginners fail. They play a game, see the result, and move on. That's like taking a test and never checking your answers. You need to find your own mistakes. Start by looking at the moves where you felt unsure or where the evaluation changed.

Don't use the engine right away. Try to explain why you made a bad move. Was it a tactical oversight? Did you miss a plan? Then turn on the engine to confirm. Write down one concrete lesson. For example: "I hung my queen because I didn't check for checks." Next time, you'll remember. That one habit will fix more rating points than any opening book.

The Real Cause: What Nobody Tells You About Endgames

Here is the truth: most beginners ignore endgames. They think endgames are boring or too hard. But endgames are where you turn a winning position into a win. You can't just hope your opponent blunders. You need to know how to convert. And the good news is, you only need a few basics. King and pawn vs king. Rook and king vs king. The opposition. That's 80% of what you'll face.

The real cause of your rating plateau is not that you're bad at chess. It's that you don't have a system. You're doing a little bit of everything, but nothing consistently. This plan gives you structure. It's designed for your brain. Tactics daily build pattern recognition. One slow game gives you deep learning. Endgames once a week fill the biggest gap. Stick with it for a month and you'll see the difference.

How the Guru Watches Your Position and Explains in Plain English

This is where I come in. The Chess Guru is a tool that watches your game in real time. It doesn't just tell you the best move. It explains why in plain English. For example, if you're about to hang a piece, it says "Your bishop is undefended. Watch out for a fork." That's the kind of feedback you need to learn fast. And it's free to start.

Imagine playing your slow game on Saturday with the Guru watching. After the game, it shows you where you went wrong and gives you a simple lesson. No jargon. No engine lines you don't understand. Just clear, direct advice. That's how you break through the plateau. You get personalized coaching every game. Try it. Your rating will thank you.

Guru catches my blunder… I actually learn this time

This guide was first published on aichess.guru, where the Chess Guru explains your moves in plain English while you play. It is free to start.

En Passant Explained: Why It Feels Illegal and How to Use It Correctly

Chapri — Thu, 30 Jul 2026 06:36:06 +0000

You move your pawn two squares forward, thinking you're safe. Then your opponent takes it as if it only moved one. That's en passant. It feels illegal. It looks like cheating. But it's a real rule, and you need to know it.

When opponent pushes pawn two squares… I just take it anyway

What Is En Passant in Chess?

En passant is French for "in passing." It lets you capture an enemy pawn that just moved two squares forward, as if it had only moved one. This only applies when your pawn is on the fifth rank (for White) or fourth rank (for Black), and the enemy pawn is directly next to yours on an adjacent file. The capture is optional, but if you don't do it immediately, you lose the chance forever.

Think of it as a special pawn capture that stops pawns from dodging conflict. Without it, a pawn could skip past an enemy pawn by moving two squares, avoiding capture. En passant keeps the game fair. It's not a trick, it's a rule designed to maintain the integrity of pawn battles.

When Does En Passant Apply?

En passant only works in one specific situation: your pawn must be on the fifth rank (if you're White) or fourth rank (if you're Black). The enemy pawn must be on an adjacent file and must have just moved two squares forward from its starting square. That's the only time you can capture en passant.

For example, if you're White and your pawn is on e5, and Black moves their pawn from d7 to d5, you can capture it en passant by moving your pawn to d6 and removing Black's pawn. The same logic applies in reverse for Black. It's a one-turn window, so you have to act fast.

Why Does En Passant Exist?

En passant exists to prevent pawns from bypassing enemy pawns without risk. Before the two-square pawn move was introduced, pawns only moved one square per turn. When the two-square option was added, pawns could suddenly skip past an opposing pawn, avoiding capture. En passant was created to restore the original balance.

Without en passant, pawns could freely advance two squares and ignore threats. That would change the game fundamentally. It's not a punishment, it's a correction. Every chess player, from beginner to grandmaster, uses this rule. It's part of the game's history and strategy.

The One Turn You Can Use It

You only get one chance to capture en passant: immediately after the enemy pawn moves two squares. If you make any other move first, the opportunity is gone forever. You cannot capture en passant on a later turn, even if the pawn stays next to yours. This is the strictest timing rule in chess.

Many beginners miss en passant because they don't look for it. They see the pawn move two squares and assume it's safe. But if you have a pawn on the right rank, you can capture. Train yourself to check after every opponent pawn move from the starting square. That split-second awareness can win you material.

What Nobody Tells You About En Passant

Here's the real truth: en passant is not just a weird rule, it's a weapon. Beginners often ignore it, but advanced players use it to create threats. If you know when it's possible, you can force your opponent to avoid moving certain pawns. You can also sacrifice a pawn to lure an enemy pawn into an en passant capture that weakens their structure.

Another hidden fact: en passant can change the entire pawn structure. Capturing en passant removes a pawn from the center or opens a file for your rook. It's not a gimmick, it's a strategic tool. The best players calculate en passant lines just like any other capture. Don't be the player who says "I forgot about en passant." Learn it, use it, and win.

How the Chess Guru Helps You Master En Passant

Reading about en passant is one thing. Spotting it in your games is another. That's where the Chess Guru comes in. Our tool watches your position in real time and highlights every en passant opportunity. You'll never miss a capture again. It explains the rule in plain English while you play, so you learn by doing.

Best of all, it's free to start. You don't need to memorize every rule. The Guru points them out as they happen. You get instant feedback, clear explanations, and a better game. Stop losing to confusion. Start capturing with confidence. Try the Chess Guru today and see how easy chess can be.

Finally used en passant correctly… Opponent rage quit instantly

This guide was first published on aichess.guru, where the Chess Guru explains your moves in plain English while you play. It is free to start.

Stuck on Castling? The 5 Rules You Keep Missing (And How to Fix It)

Chapri — Thu, 30 Jul 2026 06:31:00 +0000

You set up your pieces perfectly. You move your king two squares toward the rook. Then your opponent smirks. "Illegal move." You feel stupid. You're not alone. Castling rules trip up beginners more than any other rule. They're simple on paper but sneaky in practice. Let me fix that for you.

Me trying to castle… Opponent says 'illegal' every time

The Five Conditions for Legal Castling

Castling is the only move where you move two pieces at once. The king and the rook. But you can only do it when five things are all true. First, neither the king nor the rook has moved yet. Not even one square. If you touched either piece earlier, you lose the right for that side. Second, the king cannot be in check. You can't castle out of check. Third, the king cannot pass through a square that is attacked by an enemy piece. Fourth, the king cannot land on a square that is attacked. Fifth, all squares between the king and rook must be empty. That's it. Five conditions. All must be met.

Most beginners remember the first two. They know the king and rook must stay home. They know they can't castle out of check. But they forget the king can't jump over an attacked square. Or they think the rook can be attacked (it can). Or they try to castle queenside when the bishop's square is occupied. Check each condition before you touch the king. It becomes automatic after a few games.

Kingside vs Queenside: What's Different?

Kingside castling is short. The king moves two squares to the right (for White) or left (for Black). The rook hops over the king and lands next to him. The squares between them are f1 and g1 for White, f8 and g8 for Black. Only those two squares must be empty. The king ends up on g1 or g8. The rook on f1 or f8. Kingside is popular because it's fast and the king is tucked away.

Queenside castling is longer. The king moves two squares left (for White) or right (for Black). The rook jumps three squares. The squares between them are b1, c1, d1 for White (b8, c8, d8 for Black). All three must be empty. The king lands on c1 or c8. The rook on d1 or d8. Queenside takes one extra move to clear the bishop and queen. But it often activates the rook faster. Many beginners forget the b1 square. They think only c1 and d1 matter. That's a common mistake.

The Hidden Trap Nobody Warns You About

Here is the real cause of most castling errors. It's not the five conditions. It's the timing. Beginners try to castle too early or too late. Too early means you haven't developed pieces. You castle into a cramped position. Your king is safe but your pieces are stuck. Too late means your opponent has already opened lines. Your king gets caught in the crossfire. The best time to castle is after you've developed your knights and bishops, but before the center opens up.

What nobody tells you is that castling is not always good. Sometimes you should keep your king in the center. If the center is closed and your opponent can't attack, you might not need to castle. Or if castling puts your king on the same file as an enemy rook, think twice. The rule is: castle when you need safety, not because you can. Many beginners castle automatically. That's a mistake. Ask yourself: does my king need to move? If yes, castle. If not, maybe wait.

When Castling Is Quietly Forbidden

Sometimes you can't castle even though all five conditions are met. Wait, that sounds wrong. Actually, if all five conditions are met, you can castle. But there are situations where you think they're met but they're not. For example, you might have moved your king earlier and forgot. Or you moved your rook to capture and then back. That counts as a move. You can't castle with that rook anymore. Another trap: the king passes through a square attacked by a pawn. Pawns attack diagonally. Even if the pawn is blocked, the square is still attacked.

Here's a tricky one. Your opponent's bishop attacks the square your king must pass through. You think it's fine because the bishop is far away. But it's not fine. The king can't jump over an attacked square even if the piece is far. Also, if your king was in check earlier but you moved out of it, you can still castle later. As long as the king hasn't moved. Many beginners think once in check, you lose castling rights. That's false. Only moving the king or rook loses the right.

How Chess Guru Watches Your Position and Explains in Plain English

You don't have to memorize all this alone. When you play on Chess Guru, I watch your position in real time. If you try to castle illegally, I stop you and tell you why. Not in chess notation. In plain English. "Your king would pass through a square attacked by the bishop." Or "You moved your rook earlier, so castling is not allowed." I don't judge. I just explain. You learn by doing, not by reading.

I also help you decide when to castle. I look at your position and suggest the best plan. "Your king is safe here. Develop your knight first, then castle kingside." Or "The center is opening. Castle now before it's too late." You get personalized coaching every move. And it's free to start. No credit card. No catch. Just better chess.

Me after Guru explains castling… Finally castling like a pro

This guide was first published on aichess.guru, where the Chess Guru explains your moves in plain English while you play. It is free to start.

What Is Stalemate in Chess and How to Stop It From Stealing Your Wins

Chapri — Thu, 30 Jul 2026 06:30:59 +0000

You have a queen, a rook, and three pawns. Your opponent has only a lonely king. You are about to deliver checkmate. Then you move your queen, and the game suddenly ends in a draw. What just happened? You fell for the cruelest trick in chess: stalemate. That half point might as well be a loss when you were winning.

Me: I'm about to win… Stalemate: Surprise, it's a draw

What Is Stalemate, Exactly?

Stalemate is a situation where the player whose turn it is has no legal moves, but their king is not in check. The game ends immediately in a draw. It is not a win for the side that delivered it. It is not a loss for the side that escaped. It is a half point for both players.

Many beginners think stalemate is a rare fluke. It is not. It happens all the time when one side has a huge material advantage but does not know how to finish. You chase the king, block its escape, and accidentally take away its last square. That is stalemate.

Why Does Stalemate Happen to Beginners?

You get excited when you are winning. You want to checkmate fast. So you push your queen close to the enemy king, hoping for a quick kill. But the king runs to the edge of the board. You follow. You put your queen right next to it, forgetting that the queen also takes away squares.

Your opponent's king ends up with no squares to move to, no pieces to capture, and no pawns to block. But it is not in check. That is the trap. You gave check every move before, but this time you did not. The king is stuck, and you lose your win.

The Real Cause Nobody Talks About

Here is the truth: stalemate is not bad luck. It is a sign that you do not know how to checkmate with your material. If you have a queen and king against a lone king, there is a specific technique. You must use your king to help, not just chase with the queen.

Beginners think more pieces mean easier checkmate. That is wrong. More pieces actually create more stalemate risks if you do not coordinate them. The real cause is rushing. You stop thinking about the opponent's moves and only focus on your own attack. That is when you blunder into a draw.

How to Avoid Stalemate When You Are Crushing

First, always give check. Every move that is not check gives the opponent a chance to escape or get stuck. If you are not sure, give check again. But do not just check randomly. Use your king to shrink the box around the enemy king. Push it to the edge, then bring your king up.

Second, leave the enemy king one or two escape squares. Do not crowd it completely until you are ready to deliver checkmate. A common pattern: put your queen a knight's move away from the enemy king, with your king supporting. That gives the enemy king two squares, and you can checkmate next move.

What to Do If You Accidentally Stalemate

It happens. Even strong players stalemate sometimes in time trouble. The key is to learn from it. After the game, review the position. Ask yourself: where did I take away the last square? Could I have given check instead? Write down the pattern so you recognize it next time.

Do not blame your opponent. Do not blame luck. Stalemate is a draw because the rules say so. You cannot change the rules. You can only change your play. Practice the basic checkmates with a queen and king, two rooks, and rook and king. Drill them until they are automatic.

How the Chess Guru Helps You Stop Stalemates

At aichess.guru, I watch your games in real time. When you are about to stalemate, I send you a plain English warning: "Hey, that move leaves the king with no squares. Give check instead." No confusing engine lines. Just clear advice while you play. You learn by doing.

I also show you the exact stalemate patterns you fall for most. Over time, you stop making those mistakes. You turn those half-point steals into full points. And the best part? It is free to start. No credit card. No hidden fees. Just a coach that helps you win.

Me after the Guru fix… Now I give check every time

This guide was first published on aichess.guru, where the Chess Guru explains your moves in plain English while you play. It is free to start.

Check vs. Checkmate: Why Beginners Confuse Them and How to Fix It

Chapri — Thu, 30 Jul 2026 06:25:53 +0000

You're playing chess and suddenly your opponent says 'check.' Panic sets in. You move your king randomly, lose immediately, and wonder what happened. Here's the truth: check and checkmate are not the same thing. One is a warning. The other is the end. Understanding the difference is the first step to winning more games.

When opponent says check… I panic and blunder my queen

What is a Check? A Warning, Not a Death Sentence

A check means your king is under attack. It is like a fire alarm. The game does not end. You must respond immediately, but you have options. Beginners often freeze or make a bad move because they think the game is over. It is not. You can escape.

Think of a check as your opponent saying 'I could capture your king next move.' But you are allowed to stop that. The only rule is you cannot ignore a check. If you do, that is illegal. So take a breath. Look at your three ways out.

What is Checkmate? The End of the Road

Checkmate is different. It means your king is in check AND there is no legal move to escape. You cannot block, capture the attacker, or move the king to a safe square. The game ends right there. Your opponent wins. No second chances.

Many beginners think check and checkmate are the same because they sound similar. But one is a problem you can solve. The other is the solution to the problem. If you learn to tell them apart, you will stop resigning too early and start fighting back.

The Three Legal Answers to a Check

When your king is in check, you have exactly three legal options. First, move the king to a safe square. This is the most common response. Look for a square that is not attacked by any enemy piece. If you find one, move there. Simple.

Second, block the check. Put a piece between your king and the attacking piece. This only works if the attacker is a queen, rook, or bishop (not a knight or pawn). Third, capture the checking piece. If you can take the piece that is giving check, you remove the threat. These are your only choices.

Why Beginners Miss These Options (The Real Cause)

Here is what nobody tells you. Beginners miss the three responses because they panic. They see 'check' and think 'danger' without pausing to analyze. They also forget that blocking or capturing are legal. They only think about moving the king. That narrow thinking costs them games.

Another reason: beginners do not check if their king is already in checkmate. They might have a legal escape but assume they are lost. Or they try to move the king when blocking would be better. The real cause is lack of practice under pressure. You need to train your brain to scan for all three options every time.

How to Practice Escaping Checks

The best way to get better is to practice with real positions. Set up your board and ask a friend to give you random checks. Force yourself to find all three responses before moving. Do this until it becomes automatic. You will stop panicking and start seeing solutions.

Another tip: when you are in check, say out loud 'move, block, capture.' Then look for each one. This simple checklist will save you from blunders. Over time, you will spot checkmate threats too. You will know when you are truly lost versus when you can escape.

How the Chess Guru Fixes This for Good

You do not have to practice alone. The Chess Guru watches your games in real time. When you get a check, it explains your options in plain English. It shows you the safe squares, the blocks, and the captures. No jargon. No confusion. Just clear help while you play.

And it is free to start. You get instant feedback on your moves. You will never miss a check escape again. The Guru turns panic into confidence. Try it now and see the difference between check and checkmate disappear from your mistakes.

When Guru shows me the block… I finally escape checkmate

This guide was first published on aichess.guru, where the Chess Guru explains your moves in plain English while you play. It is free to start.

Why Your Pawn Is Worth More Than You Think (And When the Points Lie)

Chapri — Thu, 30 Jul 2026 06:25:52 +0000

You probably learned the point values right away. Pawn is 1. Knight and bishop are 3. Rook is 5. Queen is 9. Simple math, right? Trade a knight for a pawn and you lose 2 points. But here is the ugly truth: the numbers are a guideline, not a rule. Beginners lose because they follow the numbers blindly. They trade a bishop for a knight because it's "equal" and then wonder why their position collapses. The points are a starting point, not the final word.

Me trading a bishop for a pawn… Because the computer said +0.3

The Pawn Is Not a Loser

A pawn is worth 1 point, but that number is a lie when you look at the whole board. Pawns are the only pieces that can become a queen. That pawn that just marched to the seventh rank is worth way more than 1 point. It threatens to promote. Your opponent will sacrifice a rook or even a queen to stop it. Suddenly your little pawn is worth 5 or 9 points in practical value.

Consider a pawn chain. Those pawns protect each other and control key squares. Losing a pawn in the center can collapse your whole structure. The point value doesn't capture that. When you trade a pawn for a piece, ask yourself: is that pawn doing something special? If it's part of an attack or blocking a file, it might be worth more than 1.

Knights and Bishops Are Not Twins

Knights and bishops are both worth 3 points, but they are completely different animals. A knight can jump over pieces. It is great in closed positions where pawns block everything. A bishop is a long-range piece that needs open diagonals. In a closed position, a knight can be worth 4 or more. In an open position, a bishop can dominate.

Beginners often trade a knight for a bishop without thinking. But if you have a bishop pair and your opponent has a knight and bishop, that bishop pair is a real advantage. Two bishops working together control both colors. They can be worth 3.5 each in practice. Do not trade them lightly. The point values are averages, not exact measurements.

The Rook Is a Late Bloomer

A rook is worth 5 points, but in the opening it is often a liability. Rooks need open files to be powerful. In the first ten moves, your rooks are usually stuck on the back rank. They are worth maybe 3 or 4 points in practice. That is why you should not rush to activate rooks. Focus on developing minor pieces first.

In the endgame, rooks become monsters. A rook on the seventh rank can eat pawns and checkmate the king. That same rook is suddenly worth 6 or 7 points. The point value of 5 is an average across the whole game. When you calculate trades, think about the stage of the game. A rook in the endgame is much stronger than a rook in the opening.

The Queen Is a Target

The queen is worth 9 points, but she is also a magnet for attacks. Beginners love to bring out the queen early. They think they are attacking, but they are really giving the opponent free development. Every time your queen moves, you waste a tempo. And if your queen gets chased around, you fall behind in development.

A queen is powerful when the board is open and the enemy king is exposed. But in a closed position, a queen can be clumsy. She can be trapped by minor pieces. The point value of 9 assumes she can use her full power. If your queen is stuck behind pawns or blocked by knights, she might be worth only 7 or 8. Do not overestimate her.

What Nobody Tells You: The Real Cause of Bad Trades

Here is the secret: the point values are just a rule of thumb. The real cause of bad trades is not knowing the activity of the pieces. A piece that is doing nothing is not worth its full value. A knight on the rim is grim. A bishop that is stuck behind its own pawns is a big pawn. You need to evaluate each piece based on its role in the position.

The other lie is that material is everything. Sometimes you should sacrifice material for an attack. A pawn sacrifice to open lines can be worth more than a point. A rook sacrifice to deliver checkmate is priceless. The point values are a tool, not a law. Use them as a guide, but always ask: what is this piece actually doing? If it is doing nothing, it is worth less.

How the Guru Watches Your Position (And Saves You From Point Blindness)

This is where I come in. The Chess Guru watches your games live and points out exactly when the point values are lying. You make a trade that looks equal on the surface, but I show you why your bishop was worth more than their knight. I explain in plain English, no jargon. You get instant feedback while you play, so you learn in real time.

Me after the Guru explained my trade… Why did I trade my active bishop for a sleeping knight

Best part? It is free to start. You do not need to memorize tables or read a hundred books. Just play, and I will help you see the board the way a strong player does. The point values are a starting point. Let me help you go deeper. Sign up and try it. Your rating will thank you.

This guide was first published on aichess.guru, where the Chess Guru explains your moves in plain English while you play. It is free to start.

Why Does the Knight Move Like an L? (And Why Beginners Keep Hanging It)

Chapri — Thu, 30 Jul 2026 06:20:45 +0000

You know the knight moves in an L-shape. Two squares one way, then one square perpendicular. Simple, right? Then why do you keep forgetting it during games? Why does your knight end up trapped or captured for free? And why do your opponents' knights seem to fork your queen and rook every single time?The answer isn't just the L-shape. It's about how the knight jumps. It's the only piece that can leap over others. That makes it powerful but also confusing. Beginners treat it like a slow horse when it's actually a ninja. Let's fix that.

Me after moving my knight… Wait, where did it go?

The L-Shape Is Not the Hard Part

Every beginner learns the L-shape fast. Two steps forward, one to the side. You can draw it on a board in seconds. The problem isn't remembering the shape. It's seeing the knight's reach in a real game. When pieces are everywhere, you lose track of where your knight can go next.You think you know the shape, but then you move your knight to a square that looks safe. Suddenly it's attacked by a pawn you forgot was there. Or you try to fork two pieces, but the knight lands on a square where it gets captured immediately. The L-shape is just the start.

Why the Knight 'Jumps' Matters More

The knight jumps over pieces. That's its superpower. It can ignore pawns and other pieces blocking the way. But beginners often forget this works both ways. Your knight can jump out of a crowded position, but it can also jump into a trap. Just because it can land on a square doesn't mean it's safe.Here's what nobody tells you: the knight's jump means it changes color every move. It goes from a light square to a dark square, then back. That's why knights are great for attacking bishops (which stay on one color). But beginners don't track this. They move randomly and wonder why the knight feels clumsy.

The Real Cause: Beginners Don't See Knight Forks

Let's be honest. You hang knights because you don't see the fork coming. A knight fork attacks two pieces at once. Your queen and rook are both on the same color, and a knight can jump in and attack both. You move your queen, then your rook gets captured. You feel stupid. It's not your fault, you just didn't train your eyes.The real cause is that beginners only look at direct attacks. They see a pawn threatening their knight, so they move it. But they don't scan for knight forks. They don't ask: 'If I put my queen here, can a knight attack it and something else?' That split-second check saves games. You have to build the habit.

How to Stop Hanging Your Knight (Simple Fix)

Stop moving your knight to the center without support. Beginners love putting knights on e5 or d5 early. But if a pawn can chase it away, you waste time. Knights are strongest when they have a safe square and can't be attacked by pawns. Keep your knight behind your pawns until the center opens up.Another fix: count defenders. Before you move your knight, check if the target square is attacked by more enemy pieces than you have defenders. If it's one vs one, you might lose the knight. If it's zero defenders, don't go there unless it's a fork that wins material immediately. Simple math saves knights.

What Nobody Tells You About Knights

Here's the truth: knights are the easiest piece to blunder because they look harmless. A bishop points at you. A rook controls a whole line. But a knight just sits there, looking cute. Beginners underestimate its power and overestimate its safety. They think 'it's just a knight' until it forks their queen and king.The other secret: knights are terrible against pawns. A single pawn can chase a knight all over the board. But beginners don't use pawns to trap knights. They let knights dance around. If you have a knight, keep it away from enemy pawns. If you face a knight, push pawns to restrict its squares. That's the real fix.

How the Chess Guru Fixes Your Knight Blunders

You don't have to figure this out alone. The Chess Guru watches your games in real time. When you move your knight to a bad square, it flags the danger. It says 'That knight is undefended' or 'Watch out for a fork.' No jargon. Just plain English while you play. You learn by doing, not by reading a book.And it's free to start. You get immediate feedback on your knight moves. You'll stop hanging them in a week. You'll start spotting forks yourself. The Guru doesn't lecture you, it catches your mistakes and explains them on the spot. That's how you build the habit. Try it today and see your knight play transform.

Me after using Chess Guru… I actually saw the fork coming

This guide was first published on aichess.guru, where the Chess Guru explains your moves in plain English while you play. It is free to start.

What Is the Fifty Move Rule in Chess? (And How It Can Save Your Lost Game)

Chapri — Thu, 30 Jul 2026 06:20:44 +0000

You're down a rook. Your opponent has all the time in the world. You feel like resigning. But there's a rule that might save you: the fifty move rule.

When you're down a queen… But remember the 50 move rule

If 50 moves pass without a pawn move or a capture, either player can claim a draw. That's the rule. It stops games from going on forever. But for you, it's a lifeline. Here is what nobody tells you.

What Is the Fifty Move Rule Exactly?

The fifty move rule says that if 50 consecutive moves by both sides go by with no pawn move and no capture, either player can stop the game and claim a draw. Each move counts as one move for White and one for Black. So 50 moves means 100 half-moves.

You don't need to count yourself. But you need to know when to claim. The rule exists to prevent players from dragging out a dead position. It keeps the game moving. Without it, someone could shuffle pieces for hundreds of moves just to annoy you.

Why Does the Fifty Move Rule Exist?

Imagine a king and bishop against a king. That's a draw, but a stubborn player might try to win by forcing a mistake. Without a rule, they could shuffle for 200 moves. The fifty move rule puts a limit on that nonsense.

It also helps endgames that are theoretically drawn but require perfect play. If you are down material but have no pawn moves left, your opponent might not know how to checkmate you. The rule gives you a safety net. After 50 moves of no progress, you get the draw.

When Can the Fifty Move Rule Save a Lost Game?

You are down a rook and a pawn. Your opponent has a king and rook versus your king. Normally that's a win for them. But if you can avoid checks and keep your king in the center, they might not find the mate. If 50 moves pass with no capture or pawn move, you claim a draw.

Another classic example: you have a king and bishop versus a king. That's a dead draw. But your opponent doesn't know that. They keep trying to checkmate you. After 50 moves, you can stop the game. You get half a point from a losing position.

What Nobody Tells You About the Fifty Move Rule

Here is the real cause of confusion: the rule is not automatic. You have to claim it. If you don't say anything, the game continues. Many beginners lose because they don't know they can stop the game. They let their opponent shuffle until they blunder.

Also, the count resets after every pawn move or capture. So if you push a pawn on move 49, the count goes back to zero. That can be good or bad. If you are losing, avoid pawn moves. Keep the count high. If you are winning, push a pawn to reset the count and keep playing.

How to Use the Fifty Move Rule in Your Games

First, know when to claim. If you are in a losing endgame and your opponent is taking forever, start counting. After 50 moves with no pawn move or capture, say 'I claim a draw under the fifty move rule.' Be polite but firm.

Second, avoid unnecessary pawn moves when you are losing. Each pawn move resets the count. If you are the one trying to win, use pawn moves and captures to reset the count and keep the game going. The rule is a tool, not a trap.

How the Chess Guru Helps You Master This Rule

At aichess.guru, I watch your position in real time. I see when the fifty move count is getting high. I tell you in plain English: 'You can claim a draw now' or 'Don't push that pawn, it resets the count.' I am your coach, not a robot.

You get this help for free. No subscriptions, no hidden fees. Just honest chess advice while you play. You will never miss a draw again. You will turn losses into half points. Start playing with the Guru today.

Me claiming a draw… With the fifty move rule

This guide was first published on aichess.guru, where the Chess Guru explains your moves in plain English while you play. It is free to start.

Stuck in a Loop? How to Force (or Dodge) the Threefold Repetition Rule

Chapri — Thu, 30 Jul 2026 06:15:37 +0000

You're winning. You have an extra queen. Then you repeat the same position three times, and your opponent claims a draw. You lose half a point. It stings. The threefold repetition rule is one of the most misunderstood rules in chess. Beginners either forget it exists or panic when they see it. Let me explain it clearly, show you how to use it to save lost games, and how to stop it from ruining your wins.

When you blunder into a draw… by repeating moves on autopilot

What Exactly Is the Threefold Repetition Rule?

The rule says: if the same position appears on the board three times, with the same player to move, either player can claim a draw. The positions don't have to be consecutive. They just have to be identical in every way: same piece placement, same castling rights, same en passant possibilities. You don't have to claim it immediately. You can wait until the third repetition happens, then stop the clock and tell the arbiter.

Most beginners think it only counts if the same exact moves are repeated. That's wrong. You can reach the same position through different move orders. For example, you could move your knight back and forth, but the opponent's responses change the board. As long as the final arrangement matches a previous position, it counts. The key is to keep track of the position, not the move sequence.

How to Force a Draw When You Are Losing

Imagine you are down a rook. Your opponent is about to checkmate you. But you have a knight and a few pawns. You can start checking the king repeatedly. If the king has only one safe square, you can chase it back and forth. Each time the king moves to the same square, the position repeats. After three times, you claim a draw. This is called a perpetual check.

Another example: you have a rook and king against a king. That's a draw anyway. But if you have a pawn about to promote, and your opponent keeps attacking your pawn with their rook, you can move the pawn back and forth to the same square. If they don't vary their attack, you repeat the position. This can save you from losing. Always look for a way to repeat when you are in trouble.

The REAL Cause: Why Beginners Accidentally Repeat

Here is what nobody tells you. Most beginners repeat positions because they play on autopilot. They see a good move and play it without thinking about the previous positions. They don't keep a mental note of where the pieces were. Then they wonder why the game ended in a draw. The real cause is a lack of awareness. You need to actively remember the last few positions.

Another hidden cause: you are too focused on your own plan. You don't notice that your opponent is deliberately repeating moves to force a draw. They might be losing, so they try to trick you into repeating. If you don't pay attention, you fall for it. The fix is simple. After every move, pause and ask: "Has this position happened before?" Train this habit in every game.

How to Dodge a Draw When You Are Winning

You are up a queen. Your opponent keeps checking you with their rook. You move your king back and forth. After two repetitions, you realize the third is coming. What do you do? You need to break the pattern. Move your king to a different square, even if it seems less safe. Block the check with a piece instead. Or move your queen to cover the checking square. Any change that alters the position works.

Sometimes you can't avoid the repetition because every move leads back to the same position. In that case, you have to accept the draw. But usually you can dodge it by playing a different move earlier. For example, if you see your opponent setting up a perpetual check, don't let them start. Trade off their attacking pieces or create a safe haven for your king. Prevention is better than cure.

Practical Tips for Keeping Track

You don't need to memorize every position. Just remember the last few key ones. If you notice the same piece configuration twice, write it down mentally. In online chess, the interface often shows a repetition warning. Use it! In over-the-board games, you can ask the arbiter to keep track. But the best method is to develop a habit of scanning the board after each move.

Another tip: if you are losing, actively try to repeat. Move your pieces to the same squares over and over. Your opponent might get frustrated and make a mistake. If you are winning, avoid moving the same piece back and forth unless you have a plan. Always have a threat. If your opponent offers a draw via repetition, you can refuse by playing a different move. But you must do it before the third repetition.

How aichess.guru Helps You Master This Rule

At aichess.guru, our AI watches every move you play. It knows when a position is about to repeat. It will alert you in plain English: "Watch out, this position has occurred twice. Change your move to avoid a draw." Or if you are losing, it will say: "Try to force a repetition here. Move your queen back to the same square." No confusing notation. Just clear advice while you play.

You get a real-time coach that sees the board like a grandmaster. It helps you spot patterns you would miss. And the best part? It is free to start. You can play a game right now and get instant feedback. No subscriptions, no hidden costs. Click here to try it. Stop losing draws you should have won. Start turning losses into draws. Your chess will never be the same.

When the AI saves your game… and you finally win that drawn position

This guide was first published on aichess.guru, where the Chess Guru explains your moves in plain English while you play. It is free to start.

What Is a Fork in Chess? The One-Move Attack That Wins Games

Chapri — Thu, 30 Jul 2026 06:15:36 +0000

You've got a nice position. Everything is defended. Then your opponent plays a knight move that attacks your queen and rook at the same time. You lose material. That sinking feeling, you just got forked. A fork is a single piece attacking two or more enemy pieces at once. It's one of the most common tactical patterns in chess. The fix is knowing the shapes and training your eyes to see them before they happen.

When you finally get a fork… But it's your own pieces

1. Thinking only knights can fork

Knights are the fork champions. Their weird L-shape lets them attack pieces that seem safe. But pawns, bishops, rooks, and queens can fork too. A pawn fork happens when a pawn attacks two pieces diagonally. A bishop fork uses long diagonals. A rook fork uses ranks and files. The queen can fork in any direction. The fix: treat every piece as a potential forker. Ask yourself: 'If I move here, can any enemy piece attack two of my pieces at once?'

2. Placing pieces on the same color as your opponent's bishop

A bishop only attacks squares of its own color. If you put two valuable pieces on the same color squares, and that bishop is aimed at them, you're asking for a fork. For example, a light-squared bishop can fork your queen on g4 and rook on d1 if both are on light squares. The fix: when your opponent has a bishop, avoid putting two high-value pieces on its color. Spread them out, put one on light, one on dark.

3. Forgetting the knight's 'family fork' pattern

The most devastating knight fork is the 'family fork', attacking the king, queen, and rook all at once. It usually happens when the king is on its starting square and the queen and rook are nearby. The knight lands on a square like f7 or c7. The fix: memorize the key squares. If your king is on e1, a knight on f7 forks king on e8 and queen on d8? No, the pattern is: knight on f7 attacks e5, g5, d6, d8, h6, h8, g5? Wait, the exact squares: from f7, knight attacks d6, d8, e5, g5, h6, h8. So if king is on e8, queen on d8, rook on h8, bam. The fix: before castling, check if the opponent can land a knight on f7 or f2.

4. Moving your queen out too early

Beginners love to bring the queen out early. It attacks everything. But it also becomes a fat target. If your queen is on d5 and your rook is on a1, a knight on c3 forks them. Or if your queen is on f3 and king on e1, a knight on g4 forks queen and king. The fix: keep your queen back until the board is less crowded. Don't develop her before knights and bishops. If you must bring her out, make sure she's not on a square that can be forked with another piece.

5. Not seeing the 'quiet' fork

Not all forks are loud. A quiet fork happens when a piece moves to a square that attacks two pieces, but neither piece is under immediate attack elsewhere. For example, a knight moves to e4, attacking your queen on f6 and rook on c3. You don't notice because you're looking at other threats. The fix: after every opponent move, scan the board for pieces that can attack two of yours. Use your peripheral vision. Train with puzzles focusing on double attacks.

6. Ignoring pawn forks in the opening

Pawn forks are common in the opening. You play d4, then c4, attacking the knight on d5 and bishop on e6? No, a pawn fork like e4-e5 attacks f6 knight and d6 pawn? Actually a classic: after 1.e4 d5 2.exd5 Qxd5 3.Nc3, Black's queen is attacked. Black might play Qa5, then White plays d4, and if Black plays Nf6, White's e5 pawn forks the knight on f6 and the bishop on c5? Wait, e5 attacks f6 and d6? No, e5 attacks f6 and d6 only if pieces are there. The fix: be aware of pawn advances that attack two pieces. In the opening, watch for pawn moves like e5, d5, or c5 that can fork your knights and bishops.

Puzzle break — Black to play (rated 1191).

An endgame position. One accurate move decides it.

FEN: 6k1/p2n2pp/6r1/1p1qp3/Pp2P3/6P1/1PP2P1P/R2QK2R b KQ - 0 19

Show the answer

Black plays Qxe4+. The best try is Qe2, and the attack keeps going.

Solve one like this with the Guru watching →

7. The Guru: Watch your back rank

The back rank is a fork magnet. Your king, rooks, and queen often sit there. A knight on f7 or f2 creates chaos. A bishop on the long diagonal can fork a rook on a1 and queen on d4. The fix: before you castle, make sure your king has an escape square. Keep a pawn moved in front of the king. If you have a rook on f1 and queen on e1, a knight on g3 forks them. The Guru's advice: always look at the squares your opponent's pieces can reach. If a knight can jump to a square that attacks two of your pieces, don't let it happen. Move one piece or block the square.

Me after learning fork patterns… Suddenly I see them everywhere

The Guru watches your actual position. He doesn't give generic advice. He looks at your board and says: 'Your knight on f3 is eyeing g5 and e5. Your opponent's bishop on c1 is on a dark square. Put your queen on a dark square too? No, that's a fork waiting to happen.' He explains in plain English while you play. He is free to start. Just open a game and ask him: 'Am I about to get forked?' He'll tell you the truth.

This guide was first published on aichess.guru, where the Chess Guru explains your moves in plain English while you play. It is free to start.