I just switched from Anki 24.11 to 25.2 and noticed the FSRS simulator under deck options giving me pretty substantially different results now. For example, I have a deck with no more new cards, a solid retention > 80%, and currently ~150 reviews/day. With no new cards I was expecting my reviews to drop over time, and indeed the old simulator gave me this:
I did not re-optimize or change the fsrs parameters or any of the deck options. As you can see, reviews are no longer expected to decrease over time and now I don’t know which result to believe. Is there any reason for why results might be so different, i.e. were there any significant changes to the simulator? I kind of think that the old result looked more reasonable, as I was expecting (or hoping for) reviews to come down over time.
Yes, definitely – there were bugs fixed. And because the feature is still experimental, there may be more.
It might help us figure out what is happening for you if you include things like your parameters (as text, not a screenshot), your Desired Retention (DR), the settings you’re using in the simulator (for instance, do you have a low max interval?), and your Stats > True Retention table.
There have been other folks reporting similar results in the Simulator lately, so perhaps this is the sign of another issue that needs to be addressed. Or perhaps the advice to them might shed some light for you.
Thanks for your reply! OK, if there have been significant updates/fixes then maybe the new results are the better ones. I will keep an eye on how well it matches my actual reviews.
