I'd have to see some serious stats to believe whatever SRS algorithm you use is ...

williamsss · on Feb 10, 2025

It uses the same FSRS as Anki. You can see some of the logic in my open source version of the app here https://github.com/Amber-Williams/language-flashcard-app-gpt...

Anki uses this FSRS which is an open sourced algorithm https://github.com/open-spaced-repetition/fsrs4anki

jamager · on Feb 10, 2025

not OP. I have a competing product with an ad-hoc algorithm that I 100% believe is better than FSRS (tested it for months or real practice)

But what you ask is not feasible, neither I nor OP have 700 millions of reviews to stack up against FSRS.

I have my own experience and a bunch of simulations, and if FSRS were better I would have switched to it.

The reason in my case btw is that FSRS is generic, mine is tailor made for languages - big difference.

jarrett-ye · on Feb 10, 2025

You don't need to beat FSRS with 700 millions of reviews. You can collect thousands of reviews for months and compare the ad-hoc algorithm with FSRS on those reviews.

marcusbuffett · on Feb 10, 2025

Yeah as someone working on educational software, anything hand-rolling their own SRS is a pretty big red flag. Beating FSRS is going to be next to impossible, especially FSRS with parameters optimized from your users’ review history.

williamsss · on Feb 10, 2025

The app is using the same Anki FSRS algo its open source https://github.com/open-spaced-repetition/fsrs4anki

Agreed would be swimming against the tide rolling a custom one. Anki's is awesome! This app functions much the same just makes card generation MUCH faster than Anki.

jamager · on Feb 10, 2025

FSRS is so cargo-culted. It's just an algorithm, claiming that any algorithm can't be improved is ridiculous.

jarrett-ye · on Feb 10, 2025

Actually, we have found some algorithms which outperform FSRS[1]. Unfortunately, it's hard to deploy them in user's local device.

[1] https://github.com/open-spaced-repetition/srs-benchmark?tab=...