Ok.. maybe this topic is floating around somewhere but I haven't been able to find an answer specifically for my question... I'm wondering if it's necessary to buy the RTH books if I've completed RTK already. I know there is a massive spreadsheet available in the forum which will be very helpful but can anyone tell me if the number of different characters between chinese and japanese warrants the necessity to buy and work through RTH? Can someone give me the number of different characters in total? Are there a lot of new primitives in which I will need the book for?

Thank you

This is the wrong forum, so the thread might get moved...

IME, no, there's no point in getting the RTH traditional books, since the overlap is very high. I suspect the same is true for the simplified versions. I was using a simplified Hanzi text and saw no need for using the RTH books, although I did buy the first one.

I think it's nice to have them for reference but it's not essential, and reviewing both sets with an SRS can be really confusing because of the many conflicting keywords and primitive names (I'll give some examples below.) About the numbers you asked:

1. In case you finished both RTK1 and RTK3 and want to know what's different in RTH1 and RTH2 (by T, I mean the traditional set):

The good:
• 2016 characters are the same (the keyword might be different, see below);

The unnecessary:
• 0766 characters are covered in RTK1&3 but don't appear in RTH1&2;

The work:
• 0772 characters are not covered at all in RTK1&3, so you'd need to study them;
• 0247 characters are covered in RTK1&3 but are different from the ones in RTH1&2, so you'd need to study them too;

Total additional characters: 1019.


2. Now the keywords stats, which are independent from the character ones presented above:

• 0852 characters use the exact same keywords (even if the character is different, such as labor 勞–労 or encourage 勵–励);
• 1411 characters use different keywords (can be something small such as 翼 wings–wing, medium such as 助 assist–help, or big such as 拐 turn–kidnap).
• the remaining keywords don't have matching characters.

These "1411 different keywords" is what causes the most problems when studying both RTK and RTH, especially when they are on the "same characters" list. More examples of significant differences (the left one is the RTH keyword, the right one is RTK):

括 include – fasten
販 peddler – marketing
瞬 instant – wink
優 excellent – tenderness
帳 tent – notebook
爾 you (literary) – let it be
憑 proof – possessed
... and so on.

Source: This new RSH1&2+RTH1&2+RTK1&3 spreadsheet that I made, combining the ones which were posted before on the hanzi forum, plus some small filtering and coding with Excel 2010. The character comparisons use their Unicode codepoints, so it's possible that they might have the same codepoint and look slightly differently depending on the font, or have different codepoints but look the same.

Update:
Here are the numbers for those who finished only RTK1 5th. Ed. + characters 3008–3030 in this website, which are part of the supplement; the explanation and examples about the stats are the same as above so I won't repeat them:

Characters:
• 1589 characters are the same in RTK1 and RTH1&2 (keyword might be different);
• 1225 characters are new in RTH1&2;
• 0221 characters are different;
• 0255 characters are in RTK1 but not in RTH1&2.

Keywords:
• 0759 are the same in RTK1 and RTK1&2;
• 1051 are different;
• the remaining keywords don't have matching characters.

gadaxman,

Great summary. If it can help, Katsuo's list (and the one I link on my "beginner's guide") includes a kanji meaning group. That can help narrow down which RTH keywords need extra attention.

Gdaxeman, awesome! Exactly what I was looking for and more Thank you! Of course you would have been the one to know the answer Thanks for putting together the spreadsheet also!

is it necessary to learn the different keywords? or can i use the RTK keywords to learn the rest of the hanzi that need to be learned?

You don't need to learn different keywords and primitives for the RTH characters that were covered in RTK; for those, the RTK ones are just fine. Perhaps it's even better if you don't, to minimize duplicate/unnecessary work, saving a lot of time (months) and energy that you can use for other things. But I don't know for sure — it's just what I feel now that I'm more advanced in the game, with a big dose of hindsight and confirmation bias.

And yes, you can use the building blocks you learned in RTK for all the remaining hanzi; the only work is identifying the parts that compose the characters and finding new unique keywords for them in case you want to review them with an SRS. I suggest you to study RTH1&2 fully only if RTK was not cemented in your mind – if you "don't remember anything", so to speak. The advantage of doing so is that then you won't have to choose and deal with clashing keywords and primitive names, because Heisig and Richardson have done all the work already, but this advantage would be completely lost if you had to replace things that are already working well in your memory.

The way I see it, Heisig's keywords are there more to remember character forms than meanings, which is why any keyword works, no matter if it's related to the character's real meaning or not (better if it is, not that important if it isn't), so no need to worry about changing them if they're working. The next stage – following a textbook/Chinesepod/lessons, lots of conscious exposure to the language, dictionaries, and the help of an SRS to accelerate the process (the improved Cloze Delete in Anki 2 is great for this), is what will give you the meanings you need.

I agree that learning the new character keywords isn't very useful. That was my experience, anyway. Just start reading some Chinese! It's more fun and you'll learn more, anyway. I use this book:

http://www.amazon.com/A-Practical-Chines...se+grammar

But I don't know much about the alternatives. I picked it because it goes with the "New Practical Reader" series fairly well, although it was written for the older series. The book does have some typos, according to my Mandarin teacher.

gdaxeman Wrote:The work:
• 0772 characters are not covered at all in RTK1&3, so you'd need to study them;
• 0247 characters are covered in RTK1&3 but are different from the ones in RTH1&2, so you'd need to study them too;

Total additional characters: 1019.

Is it possible to sort the spreadsheet so that it will show the characters that need to be learned?

Thanks!

Miyumera Wrote:Is it possible to sort the spreadsheet so that it will show the characters that need to be learned?

The quick and dirty way is to simply sort using two columns at the same time, with the 'K and H check' as the main sort and 'RTH#' as the secondary one (you can also use the 'K and TH keywords' as the second level sort, prior to sorting by RTH#, to group the characters that have the same/missing/different keywords). Then use the Filter in the 'K and TH keywords are...' column to remove, for example, characters that don't have Heisig keywords.

One thing to add: in the summary I provided you before, I didn't include simplified characters, but remember that hundreds of kanji are simplified hanzi, so the numbers are even better in some ways. Also, I updated the spreadsheet since the last time I mentioned it, so if you got it at that time, download it again because that one had some unwanted errors.

Thanks, yeah I'm not interested in the simplified hanzi anyway. I'm not quite sure I follow how to sort two columns at once.. (i'm not that computer savvy..).. however, I couldn't really try it out because after downloading it and opening it, microsoft excel tells me there's a problem with the file and opens it in protected view in which the last three columns NOP are all blank but flagged red... is there a reason for this?

That's because you have Protected View enabled, which I believe was introduced in Office 2010 and is activated when you open files from the Internet or from your browser's temporary files folder; just click on the message and press the "Edit Anyway" button in there. The three last columns have a small code that compares the character in other columns, which are necessary for the functionality you want.

To sort using two or more columns in Excel 2007 and 2010, just select the entire table, press Sort & Filter in the Home tab, Custom Sort, check 'My data has headers' (otherwise the header will be part of the sort), then press 'Add level', and select the columns you want to sort, which are the ones I mentioned earlier.

Thanks gdaxeman... a little late reply, but i tried to figure it out a few times....

I couldn't really follow your instructions..and I don't think I understand the abbreviations in the chart.

1. K<>H : When you say this under k and h check, this doesn't necessarily mean that the characters itself are different right? just that some aspect of the comparisons are different ..ie. just the keyword but not the character?
and I guess your '<>' means NOT EQUAL TO?

2. I'm not quite clear why you asked me to sort k and H with RTK# as secondary sort, I don't really get what this brings up......
I think the k and th check as a secondary sort as you suggested was more useful..

I just want to know how do i get the:

• 0772 characters are not covered at all in RTK1&3, so you'd need to study them;

I use K and H check w/ K and TH check as 2ndary sort... right?


• 0247 characters are covered in RTK1&3 but are different from the ones in RTH1&2, so you'd need to study them too;

How do I get this one??

Thanks..

First and foremost, I updated the spreadsheet a couple of days ago, so if you didn't get the update from the Hanzi forum, get the "original, unconverted Excel 2010 spreadsheet" now:
→ RevTH forum: Spreadsheet with RSH + RTH + RTK!

I suggest the unconverted one because it's already formatted as a table and ready to use, with all the right fonts, filters enabled and so on. There were also many reliability updates so it's definitely worth it. The statistics I gave you before are still similar but not exactly the same anymore.

About what you've asked:

1. K<>H → Kanji <not equal to> Hanzi. The <> is a 'different' operator in some programming languages and actually what I was taught to use in math classes here (now that I think about it, I could use != in the spreadsheet but I don't know if that would be clearer or not.) These are the characters that are aligned in some way but are not the same; in the old spreadsheets there were some odd aligments with keywords but in the new one just means that the kanji are Japanese variants of the hanzi (and some Japanese variants are archaic versions of hanzi, things that were used before in China but not anymore. These are all indicated in the new K and H Visual Check column [another non-descriptive column name, I know].)

2. The sort order I suggested was just that, a suggestion; you can do anything you want if you think it's better. Using the RTH# (not RTK#) as the secondary sort was just so that you could get the RTH characters you have not seen in RTK in crescent order, as they are presented in the book.

Now, in the new spreadsheet, you can do some neat tricks such as these:

1/2. Hide the characters that were not covered in the Traditional Hanzi books by using a filter (the down arrow you see in the screenshot there) in the K and TH keywords column and uncheck the RTH key missing and RTK key & RTH key missing boxes (you don't need to do this, but if you don't then you'll get many characters that don't have keywords.)

2/2. Hide the kanji you've already seen in RTK by using a filter in the K and H check column where you unselect the K = TH = SH and K = TH boxes.

These are simply filters to hide what you've "already seen", to make the spreadsheet cleaner; you can disable the filters if/when you want to see the hidden characters for some reason or another, such as to check the new keywords in case they are different or something like that.

Then just sort things in any way you want — I believe what you want to study are probably the K <> H, K = SH and K missing characters defined in the K and H check column. That's all, no need to differentiate them between "what was covered but it's different" and "what was not covered", it just complicates things. K = SH are the Kanji that are simplified hanzi, that is, you know the simplified variant but need to study the traditional one. By "know" I mean that Excel says their codepoints are the same, which often means they look the same but there's a very small number of them that look different depending on which fonts you use.

I know these column names and values are a little hard to get at first but it was necessary to make the columns smaller and tidier, reducing the horizontal scroll and white space; the first versions, back in 2011, had full column names and values but that just didn't work out very well.