Just watched Anking’s latest vid; if I got it correctly:
If I only ever use Good, or, (heaven forbid) Again, PLUS, I reset the Ease factor of the whole collection to 250% daily, then I needn’t really bother with FSRS, correct?
(I’m really happy with SM; hope it’ll always remain an option …)
I’m not familiar with FSRS. However I found some FAQs that may answer your worry.
You say that you only use Good and sometimes Again.
Q12: I only use “Again” and “Good”, will FSRS work fine?
A12: Yes. FSRS is about equally accurate for people who rarely use “Hard” and “Easy” and for people who use all 4 buttons a lot. However, this is not the final conclusion, and as we gather more data, this conclusion may change.
You say that you reset Easy factor to 250. It seems that FSRS doesn’t use Easy, Difficulty instead(my humble opinion).
The
graduating interval,easy interval,new interval,starting ease,interval modifier,easy bonus, andhard intervalbecome irrelevant.
AnKing’s video was generally fairly positive about FSRS, saying that it would give the same level of retention with about 20% or so fewer reviews.
But he did say that with FSRS, it’s especially important that you have to do the ratings honestly: for example, if you actually forgot something, then be sure to mark it Again and not simply Hard.
I was happy with SM-2, but I like the idea of improvement, so I started using FSRS as soon as I installed 23.10.
Do you have any worries in particular about FSRS? Is it doing anything you don’t like?
Of course, the algorithm plays an important role; but at some point it might become a question of diminishing returns …
Keep in mind that Ebbinghaus made his experiments with completely random syllables, that’s not what most of us are learning. So, in addition to the algorithm, there are other factors that influence the retention rate.
Plus, on top of the unavoidable rep-load, you’ll be adding decision fatigue: This answer was, well … Good, NO, WAIT, it was Easy … or was it Hard ?? — Oh no, it was … Again … 
Plus … I’d prefer an algorithm that’s reasonably easy to understand … 
addendum
Random encounters with the material to be learned outside of Anki reps will happen for some cards, but not for others, which, to some extent, diminishes the importance of a perfect algorithm …
Doesn’t FSRS account for and adapts to that?
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.
