Switching my learning process to SuperMemo

From SuperMemopedia
Revision as of 11:36, 7 May 2011 by SuperMemoHelp (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

From: Chrostopher


Before I started using SuperMemo I regularly used another flashcard program. This program, however, does not support long repetition intervals as they are regularly used in SuperMemo. The longest possible interval in this program is 1000 days. For this reason I would like to transfer the most important flashcards from that program to SuperMemo. However, as I have already worked with those flashcards, I want SuperMemo to start scheduling those items with a interval of e.g. 360, or sometimes 1000 days. Then, once I will have done my first repetition on those items in SuperMemo, SuperMemo should continue to schedule the repetitions on those items like it always does (according to my forgetting pattern). For example, if I answer the question wrong, it could continue with an interval of 6 days and if I answer it correct it could continue with an interval of 3000 days.

I thought about using the Remember function in SuperMemo and set the interval as long as the current interval in the other program is. However, I do not know if this will work the way I want it to. Moreover, I found the function Reschedule which obviously works quite similar. Is there a difference between those two functions? Is one of those a feasible solution or how do I best approach this problem?


  • If your collection is large, you can confuse SuperMemo by learning material that you already know pretty well. As a result you will get longer intervals that will get corrected only when you new material starts influencing the learning process. You could probably auto-memorize all the material, spread it equally over a period of time, and set intervals manually each time you know an item well.
  • A perhaps better idea would be to continue with the other application and transfer to SuperMemo only items that you forget? This would probably be less work, and less confusion between learning algorithms?