QUADRO rules — how to play
The complete rules of QUADRO in one page: drafting tokens from attention nodes and the buffer, context lines, hallucination overload, how memory settles, and when the game ends.
QUADRO is a game of complete information. Five attention nodes, a central buffer, and two consciousness boards — everything you can see is everything there is. There are no hidden hands and no hidden bag order.
The interface
Each round begins with five attention nodes holding four tokens each, and an empty buffer. Tokens come in five cognitive frequencies: blue (sorrow), yellow (sunlight), red (passion), green (stillness) and white (clarity). Every token also carries its initial letter, so you never have to rely on color alone to read the board.
Your neural interface sits on the left, your AI opponent's on the right. Each board has three parts:
- Five context lines on the left, holding 1, 2, 3, 4 and 5 tokens from top to bottom. This is staging — tokens are held and aligned here until the round settles.
- The memory on the right, a 5×5 grid. This is where permanent identity and points come from. Every color has a fixed column on every row.
- The hallucination line underneath, which holds excess tokens you could not place. It represents cognitive overload and costs you points.
A turn is two clicks
Every turn, without exception:
- Take. Focus on one color from one source. You take every token of that color from that source — never one token, never a mix. The source can be an attention node or the buffer.
- Place. Send the tokens you just took to one context line, or directly to hallucination.
When you take from an attention node, every token you did not take slides into the buffer. An attention node empties completely the moment either consciousness draws from it, so taking a single token off a full attention node hands the other three to your opponent via the buffer. That is often the real cost of a move.
The buffer is a sixth source and it grows all round as both sides spill unattended tokens into it.
The initiative marker
The first player to draw from the buffer in a round also takes the initiative ("1") marker. It costs one slot on your hallucination line — inflicting a −1 point neural penalty — but in exchange you lead the next round. That initiative is vital when a crucial memory token is running short.
Context lines
A context line can only ever hold one color at a time, and once filled, it settles into exactly one square in your permanent memory — the square in that row's designated column for that color. Two fundamental rules follow:
- A context line that already holds red refuses any other color.
- If your memory already has red in that row, that row will not accept red at all.
A partly-filled line is not penalized at the end of a round. It simply waits, keeping its tokens into the next cycle, and only that color may be added until it settles. Wide rows take several rounds to fill — and your AI opponent can see exactly what you are waiting for.
Overflow and hallucination
You must always take every token of the color you picked, even when the context line you want has fewer free slots than you have tokens. The extra tokens drop onto the hallucination line, and each one costs you points.
This is the trade you make all game: a completed context line crystallizes into points, while overflow triggers negative hallucination penalties. Taking two tokens for a one-slot line is often still correct — a completed line that scores 4 points easily outweighs a −1 hallucination. The hallucination line clears at the end of every round, so a penalty never follows you past the round you took it in.
How a round settles
The round ends when the last token leaves the attention nodes and the buffer. Then, in order:
- Every completed context line moves one token into permanent memory, in its color's column, and scores immediately (see how scoring works). The rest of that line is discarded.
- Incomplete context lines stay exactly as they are, carrying over into the next round.
- The hallucination line is penalized and cleared.
- New attention nodes are dealt, and whoever holds the initiative marker leads.
How the game ends
The game ends the moment either player completes a full horizontal row in their memory grid. That round is played out and settled as normal, then the end-game convergence bonuses land:
- +2 for each complete horizontal row in memory
- +7 for each complete vertical column in memory
- +10 for each color frequency you have all five of
A single column bonus is worth more than three completed rows of drafting, which is why the endgame is usually a race to set up bonuses rather than to fill lines quickly. Note that the player who ends the game is not necessarily the player who wins it.
What to do next
- The Story: The Mirror Link Protocol — read the lore behind the consciousness duel.
- Learn to play walks through one real round, move by move, in about two minutes.
- How scoring works explains how memory clustering multiplies your score.
- Practice is untimed and unrecorded, against any AI opponent.