The optimizer

EON's spaced repetition algorithm comes with default parameters that work well for most people. But everyone remembers differently. The optimizer reads your review history and adjusts those parameters so the algorithm adapts to your own forgetting curve.

What it does exactly

When you run the optimizer, EON analyzes the session logs for that Deck Pack — how many times you correctly recalled each card, how many times you forgot it, and how long passed between reviews. With that information, it calculates internal weights that make the predicted intervals match your actual behavior as closely as possible.

The result is an algorithm tuned to you: cards you take longer to forget are spaced further out; cards you struggle with come back more often.

How many reviews you should have before using it

The optimizer needs enough data to calculate something meaningful. The app tells you: recommended at 200+ reviews. With fewer logs, the result may be less accurate than the default parameters. With many more, the personalization is more precise.

How to run it

1. Open the Review Preferences for the Deck Pack you want to optimize.
2. In the Algorithm tuning section, press Train with my history.
3. Wait a few seconds while it processes.

If the process completes successfully, you will see a summary with the result. If there is not enough data, the app tells you with a message.

What changes afterwards

The internal weights of the algorithm are updated for that Deck Pack. The upcoming review dates are recalculated progressively: they do not all change at once. The algorithm uses the new weights the next time it grades each card.

No history is lost. You can return to the account-wide parameters at any time with Reset to account-wide settings.

How often it makes sense to run it again

There is no fixed rule. As a practical guide: the first time you have enough reviews accumulated, and then when you have added many new cards or several months of consistent studying have passed. If your study habits and the type of material do not change much, the optimizer result stays valid for a long time.

The optimizer works Deck Pack by Deck Pack. If you have several, you can optimize each one separately based on the data available in each.

Frequently asked questions

Can I run the optimizer at any time? Yes. The app only warns you if there is not enough data. Otherwise, you can run it whenever you like.

Does the optimizer change desired retention or the maximum interval? No. Those settings you control manually. The optimizer only modifies the internal algorithm weights, not the parameters you configure from the interface.

What if the results are worse after optimizing? You can restore the account-wide parameters with Reset to account-wide settings in the Deck Pack's Review Preferences. No history is lost.

Next step

• The spaced repetition algorithm — understand the settings the optimizer complements.
• Review sessions — the ratings you collect are the raw material for the optimizer.
• Statistics — check your metrics to know when you have enough data.