Qi / QuaiController

A dual-token controller balancing two rewards on one physical anchor. Move the miners and watch it correct.

The simulator

Move the miners, watch the controller respond.

Miner preference drives everything. When miners lean toward Qi, the controller raises kQuai to make the Quai election more attractive; when they lean toward Quai, it lowers kQuai. Every other slider changes the environment that correction happens inside.

Direction
kQuai
Qi reward
Quai reward
Qi / Quai rate

The mechanism

One equation, applied every block.

kQuaii = kQuaii−1 · [ 1 ± r ]
01 · Qi reward

Proportional to difficulty

Qi emission tracks difficulty-adjusted work directly, so a Qi reward is a claim on the energy actually spent. That is what makes the Qi curve a real rate rather than a nominal one.

02 · Quai reward

Logarithmic in difficulty

The Quai reward scales with log₂ of difficulty, not difficulty itself. The two rewards therefore diverge as the network grows, and the controller exists to manage that divergence.

03 · The correction

Asymmetric by design

Pushing kQuai up when miners favour Qi is three times stronger than pushing it down. The asymmetry expresses a view: drifting toward the energy token is the failure mode worth correcting hardest.

Open questions
Not settled

No paper exists on this controller yet

The mechanism is currently code rather than a written specification, and no proof exists that a downward feedback loop cannot occur. External modelling work is underway specifically to test for it.

The deferral mechanism — miners taking 3, 6, or 12 month locks at 10%, 15%, and 25% — interacts with this controller and is not yet jointly modelled. Two controllers acting on the same variable without shared state is its own risk, independent of anything in the price mechanism.

We publish it in this state deliberately. It is the next best evolution we know how to build, and it is open to forking and refinement.