When starting to learn a new vocabulary list, I like to learn in batches. Study the first 20, re-test on the ones that I got wrong, then study the next 20, etc. With the Random card selection system I can do this by setting "Max number of cards" to 20. But this setting is not available when using Repetition-spaced card selection. The closest corresponding option I can find is Limit new cards, which the manual describes as "a very useful option that lets you limit the pool of non-“learned” cards to a finite number; this keeps you from being overwhelmed by more new words than you can manage at a time, so if you add 8000 new flashcards to your database you won’t suddenly be expected to start memorizing all 8000 at the same time." Indeed, that sounds good. But I can't quite make it work. Suppose I enabled Limit new cards by maximum unlearned with count set to 20. Then: Starting with a fresh scorefile, I will get 20 cards in my first test. Suppose I get 15 right and 5 wrong. Since I enabled the Repeat incorrect cards I will be retested on those 5, and let's say I get them right now. So far this is perfect, this is what I was looking for. After this first test Pleco now considers me to have 5 "Unlearned" cards -- best I can figure out, a card is considered "Unlearned" if it has been reviewed at least once but is not yet considered learned. Now on my next test I only get 15 new cards. Suppose again I get 5 wrong. Now I have 10 "Unlearned" cards. Next test: 10 cards; if I again get 5 wrong, there are now 15 "Unlearned" cards Next test: 5 cards. If I get all of them wrong, there are now 20 "Unlearned" cards. So far this is reasonable-ish. It's not quite batches of 20 because the "batch size" is reduced by the number of currently unlearned cards, but it's okay. But now I have a problem. Suppose I want to learn another batch. Since Pleco now considers me to have reached my maximum of unlearned cards, it will say "Session complete" when I try to start a new test. It does give me an option to continue regardless, but if I do, I will then be asked to review all cards I have reviewed before. In the example above, that is 20 + 15 + 10 + 5 = 50 cards. If during this review I will get some of the cards right that I got wrong in the previous session, they are no longer considered Unlearned (depending on the If review early setting) and so I can start to study some new cards again. But this isn't very convenient. I don't want to review all new cards -- in fact, I'd rather not review any cards at this point, I just want to learn some new. Especially when I have been working with this vocabulary list for a while, reviewing all previously studied cards would be far far too many. The cards that I got right before I'd like to review when they are up for review, depending on the regular SRS settings. Moreover, the same problem arises a few days later when the cards that I did study are up for review. Now when I start a test I will also get all cards that are up for review, rather than study them in batches. (This is somewhat less of an issue because when reviewing previously learned cards it's not quite so important to focus on smaller groups at a time, but I think it's the same problem from a technical point of view.) Am I using the SRS system in the wrong way? For now I can see two workarounds. The first workaround (the one I am using now) is to use Simple card selection for learning new cards (with a card filter that removes any cards with score higher than some threshold), and use SRS only to review (using the same scorefile). The other workaround is to split the vocabulary list into smaller segments; although Pleco makes this splitting easy enough, it's still somewhat of a pain when studying when the segments are as small as blocks of 20, say. Any suggestions?