Hello everyone =).

After doing some research and getting some awesome replies on my Core 2k/6k/10k thread, I came to the conclusion that for someone who's aiming at full proficiency or even at passing N1 eventually, the Core material won't be enough.

So, would you guys recommend any sentence decks that cover N2 and N1 with images and all the nice stuff that Core decks usually have?

Thank you in advance again.

You could try subs2srs. I don't know that there are any big, public, premade subs2srs decks around though, so you'd probably have to roll your own.
It does take some effort to get your cards set up, but you get audio and images. Plus it's native material, which is nice (I've discovered that the apparently dreaded 'とことん' appears three times in my deck of 4694 cards).

Well, pmnox made a post-10k supplement deck here. It doesn't have audio though.

I'll try these out =).

Thank you so much guys!

After core10k, you will need to stop adding things to anki, and focus on reading (a lot) in order to close the gap to N1.
speaking as someone who has taken the test... you need a high reading speed, and a comfort with MANY frequently used idioms/turns of phrase that would be super inefficient to anki.

dtcamero Wrote:After core10k, you will need to stop adding things to anki, and focus on reading (a lot) in order to close the gap to N1.

I don't know about a lot, I've only read about 7.5* VNs (most of them before doing Core or anything) and I can already read fast enough for the N1; not fast enough to have time left over to check my answers one last time, but fast enough to comfortably reread the passages that give me the most trouble.

*Then again, that actually works out to hundreds of hours so I guess you could call it "a lot" if you wanted.

my point is just that a post-core10k anki deck is kind of missing the point.

I do understand that by only studying the decks, it's possible I won't be fully prepared to take N1. But I'm frequently consuming all kinds of japanese media, which makes me believe I won't have big issues with the test if I review everything thoroughly.

Thank you for all the replies =)!

Shuugo I think you are still missing the point so I'll summarize them.

1) You need a solid vocabulary (10k+), kanji (2k+) and grammar foundation.
2) You need to be able to apply that foundation to native level reading and speaking situations with a high degree of speed and accuracy.
3) You will need some N1/JLPT specific test taking skills and strategies.

You're original post asked, "I already have done tons of #1 should I do more #1?" The answer is no, you'll need to do a massive amount of #2 and also a small amount of #3. Furthermore your post above sounds like you were going to ignore the fact that N1 is somewhat focused on scholarly articles and not "Japanese media". Mass media, regardless of the language, is often closer to middle school level so that it is easily and quickly understood by everyone.

[RawToast]

dtcamero Wrote:After core10k, you will need to stop adding things to anki, and focus on reading (a lot) in order to close the gap to N1.

From reading previous threads, I would agree. If i were at this point I would only review anki and add interesting words from reading using Rikaisama and the odd tricky sentence.

If N1/2 are your goals then make sure you aren't neglecting your listening skills. Those listening sections can be a pain.

dtcamero Wrote:After core10k, you will need to stop adding things to anki, and focus on reading (a lot) in order to close the gap to N1.

Are you sure? Personally I still like to learn the words I encounter in the wild using Anki although the further I get the more I raise my easy bonus rate. Granted I am ignoring the last 4k of 10k and just making a deck of my own after the Core6k.

I just don't like not recognizing so many (10-20k frequency) words all the time when reading.

Betelgeuzah Wrote:

dtcamero Wrote:After core10k, you will need to stop adding things to anki, and focus on reading (a lot) in order to close the gap to N1.

Are you sure? Personally I still like to learn the words I encounter in the wild using Anki although the further I get the more I raise my easy bonus rate. Granted I am ignoring the last 4k of 10k and just making a deck of my own after the Core6k.

I just don't like not recognizing so many (10-20k frequency) words all the time when reading.

Now I appreciate this sentiment... but you have to understand that Anki'ing has diminishing returns. You put a certain amount of time/energy in per word (during entering and reviews). You get high comprehension but for a relatively small data set. Reading for the same amount of time/energy gives you relatively lower comprehension but for a much greater data set. After a certain point it becomes more efficient timewise to learn new vocab from simply reading. There is a varying concensus on where this point is, but on the polyglots vs. polynots thread this seemed to come out somewhere between 6 and 10k...

Moreover reading is like srs'ing. things you have seen before keep popping up... and assuming you read >10 pages a day your vocabulary will skyrocket within one year. I forgot where the link was for that thread but someone actually did the math and 10 pages a day was the point at which returns really pile up.

The things you would be practically anki'ing after 10k would appear so infrequently, and would require relatively a lot of time to review. By comparison that same amount of time provides much greater returns for overall language comprehension via reading. Perhaps with a shallower understanding per-word, but applied to thousands and thousands of new words. Over a year this becomes very powerful.

You may forget but this is probably how you learned your native language as a child, well before 10k words...

In the case of Japanese I actually prefer the high comprehension approach because looking words up is necessary for three reasons: I forgot the meaning, reading, or both. This is double the amount of interruptions compared to roman alphabet languages. Obviously reading issues will diminish over time but exceptions are still abundant and new kanjis keep being introduced.

My point is that at least for me, until I move away from knowing readings on a per word basis to a per kanji basis I feel that there is much more value for me in taking the time to learn the words I come across because each word has two components for me to learn. It's more taxing to do on the fly than with any other language.

I notice this when I come across names. The readings just don't come to me effortlessly even though I may know many words with the kanjis in them.

It's hard to explain in all honesty. It just feels like learning in anki continues to benefit me because I am learning and reinforcing two kinds of information.

you're gonna have to take the training wheels off eventually. It's never going to feel easy when you do either, no matter how much one furiously anki's in preparation.

Yes you can also dive in without using training wheels at all.

But when it comes down to what is the most efficient way to do something I'm not buying that 6-10k argument (esp. 6k) which I agree should be true for other languages.

As long as the word includes kanji one is still unfamiliar with reading-wise or has an uncommon reading I see significant value in adding it to Anki, surpassing the value of SRS'ing roman language vocab for example.

dtcamero Wrote:

Betelgeuzah Wrote:

dtcamero Wrote:After core10k, you will need to stop adding things to anki, and focus on reading (a lot) in order to close the gap to N1.

Are you sure? Personally I still like to learn the words I encounter in the wild using Anki although the further I get the more I raise my easy bonus rate. Granted I am ignoring the last 4k of 10k and just making a deck of my own after the Core6k.

I just don't like not recognizing so many (10-20k frequency) words all the time when reading.

Now I appreciate this sentiment... but you have to understand that Anki'ing has diminishing returns. You put a certain amount of time/energy in per word (during entering and reviews). You get high comprehension but for a relatively small data set. Reading for the same amount of time/energy gives you relatively lower comprehension but for a much greater data set. After a certain point it becomes more efficient timewise to learn new vocab from simply reading. There is a varying concensus on where this point is, but on the polyglots vs. polynots thread this seemed to come out somewhere between 6 and 10k...

Moreover reading is like srs'ing. things you have seen before keep popping up... and assuming you read >10 pages a day your vocabulary will skyrocket within one year. I forgot where the link was for that thread but someone actually did the math and 10 pages a day was the point at which returns really pile up.

The things you would be practically anki'ing after 10k would appear so infrequently, and would require relatively a lot of time to review. By comparison that same amount of time provides much greater returns for overall language comprehension via reading.

Going to have to agree with this. Reading is like an elixir.

If he wants to keep prioritizing his SRS that's fine though. Anything is better than nothing.

anritsi Wrote:Well, pmnox made a post-10k supplement deck here. It doesn't have audio though.

Can anyone please reup that.
It is already deleted from Anki site.

ryuudou Wrote:If he wants to keep prioritizing his SRS that's fine though. Anything is better than nothing.

I don't think he's talking about prioritizing using Anki over reading. I still use Anki too even though I read all the time. It doesn't take much time or effort and I still notice a benefit from doing it.

Back to the main topic though, I agree that the topic creator should be reading a lot.

EDIT: Full disclosure, I'm also dyslexic so I don't know how much that affects my learning methods.