FSRS: SM2 retention and random new cards

First, thanks to @L.M.Sherlock & Co. for FSRS. I finally enabled it two days ago and it really makes a difference.

There are a few minor things I still don’t understand and I haven’t found a clear explanation:[1]

  • SM2 Retention: This is what the tooltip says:

“If your actual retention before switching to FSRS was significantly different from 0.9, adjusting this value will allow Anki to better estimate your memory state when it encounters cards that are missing review logs. Since review logs typically won’t be missing unless you explicitly deleted them to free up space, most users will not need to adjust this.”

But everybody is focusing on “unless you explicitly deleted them to free up space, most users will not need to adjust this”. However, what if I did not delete any cards but my SM2 true retention is not close to 0.9? What should I do in this case? Also, which retention are we talking about here? Recent (1 month), all-time? I would love some clear recommendations or more guidance on this.

  • Treatment of new cards: What algorithm decides which new cards to learn? Is that changed at all by FSRS? I assume not, but I’m not sure. Also, while at this, what is the recommended way of learning new cards in random order? I previously had insertion order set to random but now there is a warning to rather adjust the new card gather order with the V3 scheduler. Is this even relevant when using FSRS? If yes, what are the settings to enable learning new cards in random order?

  1. I have read the FAQ, Github documentation, and watched the Anking video about FSRS. ↩︎

I’ll answer this part: FSRS doesn’t change the way new cards are handled AFAIK, so the recommended way to change the order is still the gather order option.

Unless you have manually removed old review logs, changing that setting won’t do anything.

Thanks. I have not specifically removed old review logs, but I have definitely removed (deleted) many cards that had reviews. Probably thousands of them.

However, I am still confused by the name of the parameter (SM2 true retention). Probably not everyone had the same SM2 retention, right? For most decks, I tuned the interval modifier to retain roughly 85 %, so I guess my question is:

Should I change the SM2 true retention from 0.90 to 0.85 for my FSRS settings if m true SM2 retention was actually 0.85?

No. You can ignore it.

1 Like

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.