I think the way answer quality affects the timing of subsequent tests is not optimal. For example, a new card will be reviewed on day 1 and then, whether I barely remember it (and give my answer a quality of 4) or nail it instantly (and rate my answer 6), its score will go from 100 to 400 and I will thus see it again on day 5. On day 5, the quality of my answer at that time will be similarly irrelevant to the amount of time before I see it again.
It really ought to be the case that I see the quality-4 card for the second time before day 5 -- I had trouble remembering it and so clearly it needs to be reviewed more frequently.
I think all I want is for the difficulty to be adjusted based on the answer quality *before* the card's score is adjusted based on the difficulty. So if I rate my day-one answer 4, the resulting score will be less than 400 and I will thus see the card earlier.
Does that make sense? Is there a reason why I should want it to work the way it does?
It really ought to be the case that I see the quality-4 card for the second time before day 5 -- I had trouble remembering it and so clearly it needs to be reviewed more frequently.
I think all I want is for the difficulty to be adjusted based on the answer quality *before* the card's score is adjusted based on the difficulty. So if I rate my day-one answer 4, the resulting score will be less than 400 and I will thus see the card earlier.
Does that make sense? Is there a reason why I should want it to work the way it does?