I am on AnkiDroid. I will be a century late to the party
You can request access to the iOS beta, if that’s the one you’re referring to.
I still don’t know whether that is an intended change, especially since we are moving in the opposite direction in terms of making the first interval close enough to the median stability. Or is this going to change in a future update? @L.M.Sherlock Sorry for pinging you.
The damping improves RMSE by 1% (relatively, not in absolute terms). The change to how the initial D is calculated also improves RMSE by 1%, relatively. Overall, without a huge dataset we would not be able to detect the difference.
It only improves RMSE by 1%, but it also improves sorting by Difficulty a lot because we’re not getting huge bunches of cards tied at a D of 10.
Have you or anyone did the simulations again? Actually, it might be worth doing it once again after 24.10 releases.
I’m a little frightened by what I see here. I hope I misunderstand what this change will do.
I absolutely need very short intervals, less than a minute, for what I use Anki for.
If I am memorizing how to write a kanji, I will never finish if I have to wait minutes before I see it again.
I have experimented. I’m one of those micro steps guys.
I realize that different people use Anki for different things and that very short steps are highly inappropriate for them.
But they are for me.
Please reassure me that I am worrying about nothing.
You may be responding to the wrong comment. That comment about difficulty has little to nothing to do with same-day reviews.
If you want short learning steps, just, well, have short learning steps.
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.