Hello everyone. 2 years ago, I downloaded the Core 2k/6k Optimized Japanese vocabulary deck from the Nukemarine thread for beginners. So far I have done about 2600 cards ( yeah very slow x( ) . I found out that there's a new Core 2k/6k/10k Further Optimized deck and I would like to use it.

Is there a way I can sort the cards according to the Core 2k/6k deck and suspend the first 2600 cards and sort the remaining cards back to original setting? This deck also comes with production and recognition cards. How can I suspend all the production cards?
Thanks.

[RawToast]

It's possible, but could be a lot of hassle. You might be able to simply import the new deck and Anki will update the values.

Suspending all the production cards is rather simple. Open the browser and go to the deck then amend the search bar at the top from:

deck:<deck name>
to
deck:<deck name> card:<Name of production cards>

So something like:

deck:Core6k cardroduction

The further optimized deck I'm using comes with a lot of indices, you could see if one matches the one of your old deck (say, look up card 1000 in the browser, find the card in the new deck and see if any index says 1000). If so, sort by that index, suspend the first 2.6k cards, then resort by the optimized index (usually the first field).

If that doesn't work, you could create a new index via Excel, but it'll take some time, depending on your familiarity. Basically, find one identical field between both decks with no duplicates (maybe name of the audio file) and use VLOOKUP to pull the old 2k/6k index numbers from the first deck into the optimized one. Delete or suspend the old entries and done.

Personally, I'd just start from 0, increase your easy interval and just spend a day rushing through. That way, you can be sure you're not forgetting any old vocabulary and it doesn't take that much time, tbh. (Source: Screwed up my deck and did 5k cards in a week last month)

RawToast Wrote:It's possible, but could be a lot of hassle. You might be able to simply import the new deck and Anki will update the values.

Suspending all the production cards is rather simple. Open the browser and go to the deck then amend the search bar at the top from:

deck:<deck name>
to
deck:<deck name> card:<Name of production cards>

So something like:

deck:Core6k cardroduction

I already tried importing the deck but as I suspected, it created a new deck instead. The two decks are so different for Anki to update the values. Especially since I manually created deck from Anki 1.0 files instead of downloading it from the Anki website. Thanks for help on suspending production cards though!!

gstieglitz Wrote:Personally, I'd just start from 0, increase your easy interval and just spend a day rushing through. That way, you can be sure you're not forgetting any old vocabulary and it doesn't take that much time, tbh. (Source: Screwed up my deck and did 5k cards in a week last month)

I think I'll end up following your advice. I start from scratch and increase the interval as I discover new cards. 5k in a week!? I also screwed up my Core deck up to the extent that the cards do not have pictures any more x(. I think I can manage 2.6k cards in a week.

What do you mean by indices? Do you mean the various options you get when you click on the Field's button?

mkabness Wrote:What do you mean by indices? Do you mean the various options you get when you click on the Field's button?

Yeah, number fields that you can use to sort the deck by. Some decks don't have any, which can get really annoying if for some reason scheduling gets reset. Seems to happen sometimes when you change options. My Tae Kim deck ended up sorted by starting letter once, ugh.

I'm using 10k Further Opt, which comes with 4 index fields: Standard core, 2+4k optimized, 2k optimized and nukikame.
If(f) standard core is identical with the one of your former deck, you could select "sort by this field", resort everything, suspend, then select your new favorite sorting method (or just stick with the old one) and continue.

Still, I think the brute force method is probably preferable. I still forget some really basic vocab here and there and it's comforting to know that I'll see it every couple of months during review.

gstieglitz Wrote:

mkabness Wrote:What do you mean by indices? Do you mean the various options you get when you click on the Field's button?

Yeah, number fields that you can use to sort the deck by. Some decks don't have any, which can get really annoying if for some reason scheduling gets reset. Seems to happen sometimes when you change options. My Tae Kim deck ended up sorted by starting letter once, ugh.

I'm using 10k Further Opt, which comes with 4 index fields: Standard core, 2+4k optimized, 2k optimized and nukikame.
If(f) standard core is identical with the one of your former deck, you could select "sort by this field", resort everything, suspend, then select your new favorite sorting method (or just stick with the old one) and continue.

Still, I think the brute force method is probably preferable. I still forget some really basic vocab here and there and it's comforting to know that I'll see it every couple of months during review.

Oh seems like I'm using the same deck as yours. Mine is also 10k Further Opt and it has the index fields you mentioned. My former deck doesn't have these fields though. Seems I'll go the brute force way.

Which index field do you think is best for learning? Do you know the difference between CoreIndex, Optimized-Voc-Index-2k+4kDefault, Optimized-Voc-Index-2k and Nukimame-Core6k-Optimized-Voc-Index?

mkabness Wrote:Which index field do you think is best for learning? Do you know the difference between CoreIndex, Optimized-Voc-Index-2k+4kDefault, Optimized-Voc-Index-2k and Nukimame-Core6k-Optimized-Voc-Index?

Core seems to be the original iKnow order. Vocab is mostly correlated with frequency.
Voc 2k, 2+4k and nukikame all cluster similar kanji. So 一　and 一つ end up grouped together and so do お兄さんや兄や兄さん.

Difference between 2k and 2+4k is how much the algorithm is allowed to sort at a time. 2k works in increments of 2000 words, so whenever you finish a set of 2000, you know exactly the same cards someone using CoreIndex would know, you just learned them in different order.

2k+4k doubles the second and third increment, so after 2000 cards, you're now learning cards from the 2000-6000 range, not just 2000-4000. (Similarly, after 6000 cards, you're seeing the full 6k-10k)

Nukikame has no such restrictions and just uses the entire 6k catalog, but with a fair amount of restraint. I don't have much insight why the deck is sorted the way it is, but I'm guessing it's optimized for use in JLPT prep, since it includes odd stuff like 八百屋 very early.

"Best" is kinda subjective. I'm using 2k4k at the moment and it's fine. I don't think the differences between the last three are all that important. For you especially, since you're already a quarter into the deck.

I wouldn't recommend CoreIndex, especially for newbies. You'll see stuff like 事務所 pretty early, and without having the repeated 事 and 所 exposure of clustered sorting, it can be frustrating to try to memorize these.

Got it! Thanks a lot gstieglitz for the help and the thorough explanation.

You can actually import the new deck into the old deck and keep your learning data intact. It sounds like you may brute force it so I'll spare the details, but let me know if you want a step by step and I'll write one up.

gstieglitz Wrote:Nukikame has no such restrictions and just uses the entire 6k catalog, but with a fair amount of restraint. I don't have much insight why the deck is sorted the way it is, but I'm guessing it's optimized for use in JLPT prep, since it includes odd stuff like 八百屋 very early.

I believe nukemarine also sorted similar to your description of 2k+4k. It is also sorted by kanji rtk number so that you can study rtk and core simultaniously with low rtk numbers at the beginning of the vocab deck and high rtk numbers at the end.