These are my flashcard parameters. I'm feeling they work fine, but maybe I'm wrong and it's not so efficient. I'd be happy if you got any suggestion to improve them or if you would change something. Thank you!
I have to confess I can't help you all that much here, since I've always had my Flashcards set to Random card selection with split categories and card filters. I just feel that gives me the most flexibilty to study exactly the cards I want to. But if you know what the settings you chose mean, I suggest you write on paper what scores and easinesses your card will have after a few correct/incorrect repetitions with various different settings. That may help you picture which settings produce the most natural results for you. Or I think @pdwalker also has quite an elaborate system, I suggest you have a look at this thread:
You can also use excel formulas to forecast the progression of cards given different parameters. Figure out the kind of results you want then work backward to find the right parameters. I'll note that the correct scale score increase % parameter gave me some trouble in terms of predicted scores matching Pleco behavior, so it might be easier to set that to 100% if you want to target the other parameters systematically.
Heh. Thanks for the mention @Shun . I'm actually trying to simplify the entire process to make it easier.
If I were starting again - I would use stock SRS for everything, instead of the two stage testing I have now (and described in the mentioned thread)
The only changes I would make are:
- increase the Scoring, Scoring Options, Tweak Parameters, Score Limits: Maximum Score to 512000 to allow for cards that I really know well to disappear off my lists for a long time ("一" anyone?)
- increase the Scoring, Scoring Options, Tweak Parameters, Incorrect Score Decrease %, Incorrect to (33 or 50 - still deciding) instead of 0. Thus cards that I have "learned" but forgot "a little bit" don't get their scores reset to 0 on a mistake, but something lower so I'll see it again soon, but not too soon.
- oh, and finally always use the "repeat incorrect cards" option under more settings, commands, repeat incorrect (on), and set a time during the test