generally, i think this is a common view in the forum: http://forum.koohii.com/viewtopic.php?p … 90#p174390
the original core list is organized in order of word frequency as gathered from newspapers. the optimized deck organizes that same list in a way that is theoretically optimized for learning using (roughly) the i+1 method* (thank Nukemarine for this). If you are planning to learn all 6000 words anyway, then you may as well use the optimized deck.
i haven't finished all 6k so that's why i'm pointing to more experienced posters' opinions rather than espousing my own limited view. i did just switch from prearranged order to the optimized deck after doing about 1600 words, so maybe i'll let you know later after i have experience with both directions.

*http://en.wikipedia.org/wiki/Input_hypothesis

Edit: the chronology of the post i linked you to is such that it was written before there was a shared deck, so don't worry about the excel sheet stuff.

Last edited by tashippy (2013 February 16, 4:19 pm)

Are all 6000 cards 'optimized' in a different way than the original order was? I looked at Nukemarine's thread and after the first 3500 words or so he suggests simply unsuspending the next 500 words of the unoptimized deck up until the end.

I am asking because I'm at the 3500 point around now and would like to 'optimize' my learning of new cards but what's the point if the cards are in the same order regardless of the deck?

Betelgeuzah wrote:

Are all 6000 cards 'optimized' in a different way than the original order was? I looked at Nukemarine's thread and after the first 3500 words or so he suggests simply unsuspending the next 500 words of the unoptimized deck up until the end.

I am asking because I'm at the 3500 point around now and would like to 'optimize' my learning of new cards but what's the point if the cards are in the same order regardless of the deck?

The entire deck was optimized in two ways. There were the word frequency "chunks" which meant top 400 then 1600, then 2000 then the final 2000. These chunks were then sorted in the KO2001 kanji order which is based on kanji frequency plus grouped meanings. The word frequency grouping allows you to get bang for your buck while the kanji sorting allows both for an i+1 and easier to learn vocabulary since the next word in the list is related in some way to the previous word barring kana only words.

Note that I went with sorting the words and not the sentences. I decided to go with the idea the sentences are there to help you learn that particular word. So, no it's not the sentence method made popular by AJATT, but I think it works better for the purpose of learning beginner, basic and intermediate vocabulary in a structured manner.

Whatever you do I think you should use the new official Core 2K ordering. It's much better in my opinion.

So, I'm starting out core. Does anyone recommend the default setting on the Anki shared deck "Core 2k/6k"? It's clozed.

If not, what's the most optimal setup?

Last edited by UltraRunAllDay (2013 February 17, 1:19 am)

Nukemarine wrote:

ryuudou wrote:

Whatever you do I think you should use the new official Core 2K ordering. It's much better in my opinion.

iKnow offered a new sort that's not by dictionary order? I've not seen it so could you describe it or post a document list of it?

I think it was the middle of 2012, or maybe a bit earlier. You might've already seen it. I'm not sure if we're talking about the same thing, but I was referring to the new order that iKnow presents the Core words. I noticed that many many terms have been moved up to 500-1500 places higher or lower, but I only looked over Core2k.

Here's an official PDF of the first 1000: http://www.mediafire.com/view/?allr9z5ylvhq52e

A TSV file with all of the new ordering: http://www.mediafire.com/?bam52xun39rx68s

http://support.iknow.jp/entries/2166874 … ld-series-

Last edited by ryuudou (2013 February 17, 8:38 am)

^The front side of the card is usually the fact being tested, with an answer on the back. Unless you are studying specifically for translation purposes, going from Japanese on the front, with English on the back, is what you would be testing if you are testing Japanese comprehension.



For a very simple example:-

Front:

田中です。

Back:

[たなか]・です
(I) am Tanaka

tashippy wrote:

Daichi wrote:

If you want to further optimize your sentence order just go grab the morphology plugin. And while your at it go make some subs2srs cards for far more interesting sentences.

what morphology plugin? i see morph man 3, but that's it.

Yes, that would be the one.