If you're seeing this message, it means we're having trouble loading external resources on our website.

If you're behind a web filter, please make sure that the domains *.kastatic.org and *.kasandbox.org are unblocked.

Main content

Why lattice multiplication works

Sal explains why lattice multiplication works. Created by Sal Khan.

Want to join the conversation?

  • marcimus pink style avatar for user mlyulkin56
    Why did they name this "lattice" multiplication?
    (157 votes)
    Default Khan Academy avatar avatar for user
  • blobby green style avatar for user seth
    Does lattice multiplication work with negative numbers?
    (76 votes)
    Default Khan Academy avatar avatar for user
    • old spice man green style avatar for user Andy Bauer
      Yes, I believe it would. When doing the lattice problem I would disregard any negative signs. When you arrive at your lattice "answer" I would then follow the rules around multiplying negative numbers.

      If only one of the numbers that you are multiplying is negative, then the answer will be a negative number. If both numbers that you are multiplying are negative, then the answer will be a positive amount since the negative signs cancel out.
      (77 votes)
  • leaf green style avatar for user Themis
    I just noticed that the sum of the number of the diagonals are always the rows you have plus the columns. So extending this pattern I get to the following, since I have 9999 x 99 witch is 4 digits + 2 digits, the result will always be less than 1.000.000.
    Another example is say I have 999999 x 999 witch is 6 + 3 digits = 9 that denotes that the result will always be less than 1.000.000.000, 9 zeros and 1.
    (55 votes)
    Default Khan Academy avatar avatar for user
    • leaf red style avatar for user Rae
      Good observation! Yes, the sum of the total number of digits being multiplied together is always going to be the maximum number of digits in the answer.

      Why do I say maximum and not exactly? Let's look at some examples:
      3x2=6
      3 (one digit) multiplied by 2 (one digit) equals 6 (one digit).
      9x9=81
      9 (one digit) multiplied by 9 (one digit) equals 81 (two digits).
      10x9=90
      10 (two digits) multiplied by 9 (one digit) equals 90 (two digits)
      99x9=891
      99 (two digits) multiplied by 9 (one digit) equals 891 (three digits)
      84x11=924
      84 (two digits) multiplied by 11 (two digits) equals 924 (three digits)
      99x99=9801
      99 (two digits) multiplied by 99 (two digits) equals 9801 (four digits)

      If you continue like this, you will see that the number of digits in your answer will always be somewhere between the number of digits in the larger number (the minimum number of digits you will end up with) and the total number of digits being multiplied (the maximum number of digits you will end up with).
      (46 votes)
  • leafers ultimate style avatar for user caleb rodgers
    This is cool I never have learned this in high school or college. my question is When did this type of multiplication came out or was discovered?
    (13 votes)
    Default Khan Academy avatar avatar for user
    • eggleston orange style avatar for user Jennifer J
      Actually, it looks as if the origin of lattice multiplication dates back further, and to outside Europe.
      "Lattice multiplication, also known as sieve multiplication or the jalousia (gelosia)
      method, dates back to 10th century India and was introduced into Europe by Fibonacci in the 14th century (Carroll & Porter, 1998)."
      SOURCE: An Introduction to Various Multiplication Strategies by Lynn West, University of Nebraska-Lincoln, July, 2011
      CITING THIS REFERENCE:
      Carroll, W. M., & Porter, D. (1998). Alternative algorithms for whole-number operations. In The
      National Council of Teachers of Mathematics, The teaching and learning of algorithms
      in school mathematics (pp. 106-114). Reston, VA: The National Council of Teachers of
      Mathematics, Inc.
      (12 votes)
  • female robot grace style avatar for user The Lonely Donut
    Can you use lattice with negative numbers?
    (11 votes)
    Default Khan Academy avatar avatar for user
  • male robot hal style avatar for user Shakour
    can you do this with division?
    (5 votes)
    Default Khan Academy avatar avatar for user
  • old spice man green style avatar for user 20weberd
    Does this work with decimals or percents
    (5 votes)
    Default Khan Academy avatar avatar for user
  • old spice man green style avatar for user mhetzel
    so which is better the traditional method or the lattice method
    (4 votes)
    Default Khan Academy avatar avatar for user
  • blobby green style avatar for user Evan Williams
    Hey everyone, this does work for decimal numbers, and you don't even have to ignore the decimal point!

    Write the numbers and draw the lattice as you usually would, but put the decimal point of the numbers being multiplied right on the row and column lines that separate the place value columns.

    Do everything like you normally would, but once you're done, look at the row and column lines "coming out of" the decimal points and go to where they intersect on the lattice. Follow the diagonal line coming out of that intersection point down, and that is where your decimal point is gonna go.

    I hope that made sense, sorry, it's a little difficult to explain with just text.
    (4 votes)
    Default Khan Academy avatar avatar for user
  • leafers seed style avatar for user shriya
    does lattice multiplication work with decimals and fractions?
    (4 votes)
    Default Khan Academy avatar avatar for user

Video transcript

In the last video we did a couple of lattice multiplication problems and we saw it was pretty straightforward. You got to do all your multiplication first and then do all of your addition. Well, let's try to understand why exactly it worked. It almost seemed like magic. And to see why it worked I'm going to redo this problem up here then I'll also try to explain what we did in the longer problems. So when we multiplied 27-- so you write your 2 and your 7 just like that-- times 48. I'm just doing exactly what we did in the previous video. We drew a lattice, gave the 2 a column and the 7 a column. Just like that. We gave the 4 a row and we gave the 8 a row. And then we drew our diagonal. And the key here is the diagonal as you can imagine, otherwise we wouldn't be drawing them. So you have your diagonals. Now the way to think about it is each of these diagonals are a number place. So for example, this diagonal right here, that is the 1's place. The next diagonal, I'll do it in this light green color. The next diagonal right here in the light green color, that is the 10's place. Now the next diagonal to the left or above that, depending on how you want to view it, I'll do in this little pink color right here. You could guess, that's going to be the 100's place. And then finally, we have this little diagonal there and I'll do it in this light blue collar. That is the 1,000's place. So whenever we multiply one digit times another digit, we just make sure we put it in the right bucket or in the right place. And you'll see what I mean in a second. So we did 7 times 4. Well, we know that 7 times 4 is 28. We just simply wrote a 2 and an 8 just like that. But what did we really do? And I guess the best way to think about it, this 7-- this is the 7 in 27. So it's just a regular 7. But this 4, it's the 4 in 48. So it's not just a regular 4, it's really a 40. 48 can be rewritten as 40 plus 8. This 4 right here actually represents a 40. So right here we're not really multiplying 7 times 4, we're actually multiplying 7 times 40. And 7 times 40 isn't just 28, it's 280. And 280, how can we think about that? We could say that's two 100's plus eight 10's. And that's exactly what we wrote right here. Notice: this column or I'm sorry, this diagonal right here, I already told you, it was the 10's diagonal. And we multiplied 7 times 40. We put the 8 right here in the 10's diagonal. So that means eight 10's. 7 times 40 is two 100's. We wrote a 2 in the 100's diagonal. And eight 10's. That's what this 2 8 here is. We actually wrote 280. Let's keep going. When I multiply 2 times 4. You might say, oh, 2 times 4. That's 8. What am I really doing? This is the 2 in 27. This is really a 20 and this is really a 40. So 20 times 40 is equal to just 8 with two 0's. Is equal to 800. And what did we do? We multiplied 2 times 4 and we said, oh, 2 times 4 is 8. We wrote a 0 and en 8 just like that. But notice where we wrote the 8. We wrote the 8 in the 100's diagonal. Let me use a different color. We wrote it in the 100's diagonal. So we literally wrote-- even though it looked like we multiplied 2 times 4 and saying it's 8, the way we accounted for it, we really did 20 times 40 is equal to eight 100's. Remember, this is the 100's diagonal, this whole thing right there. And we can keep going. When you multiply 7 times 8. Remember, this is really 7-- well, this is the 7 in 27, so it's just a regular 7. This is the 8 in 48, so it's just a regular 8. 7 times 8 is 56. You write a 6 in the 1's place. 56 is just five 10's and one 6. So it's five 10's in the 10's diagonal and one 6, 56. Then when you multiply 2 times 8 notice, that's not really just 2 times 8. I mean we did write it's just 16 when we did the problem over here, but we're actually multiplying 20. This is a 20 times 8. 20 times 8 is equal to 160. Or you could say it's 100, notice the 1 in the 100's diagonal-- and six 10's. That's what 160 is. So what we did by doing this lattice multiplication, is we accounted all of the digits, the right digits in the right places. We put the 6 in the 1's. We put the 6, the 5, and the 8 in the 10's place. We put the 1, the 8, and the 2 in the 100's place. And we put nothing right now in the 1000's place. Then, now that we're done with all the multiplication we can actually do our adding up. And then you just keep adding, and if there's something that goes over to the next place you just carry that number. So 6 in the 1's place, well, that's just a 6. Then you go the 10's place. 8 plus 5 plus 6 is what? 8 plus 5 is 13. Plus 6 is 19. But notice, we're in the 10's place. It's nineteen 10's or we could say it's nine 10's and 100. We carry the 1 up here, if you can see it, it's in the 100's place. Now we add up all the 100's. 100 plus 200 plus 800 plus 100. Or, what is this? 1,200. Well you write 2 in the 100's place. 1,200 is the same thing as two 100's plus 1,000. And now you only have 1,000 in your 1,000's diagonal. And so you write that 1 right there. That's exactly how we did it. The same reasoning applied to the more complex problem. We can label our places. This was the 1's place right there. And it made sense. When we multiplied the 9 times the 7 those are just literally 9's and 7's and 63. Six 10's and three 1's. This right here is the 10's diagonal. Then we got six 10's and three 1's. When we multiplied 9 times 80-- remember, 787, that's the same thing as seven 100's plus eight 10's plus seven, just regular seven 1's. So this 9 times 8 really 9 times 80. 9 times 80 is 720. Seven 100's-- this is the 100's place. Seven 100's and 20-- two 10's just right there. And you can keep going. This up here, this is the 1000's place. This is the 10,000's. I'll write it like that. This is the 100,000's place. And then this was the 1,000,000's place. So we did all of our multiplication at once, accounted for things in their proper place based on what those numbers really are. This entry right here, it looks like we just multiplied 4 times 8 and got 32, but we actually were multiplying 400-- this is a 400-- times 80. And 400 times 80 is equal to 3 2 and three 0's. Is equal to 32,000. And the way we counted for it-- notice, we put a 2 right there and what diagonal is that? That is the 1,000's diagonal. So we say it's 2,000 and three 10,000's. So three 10,000 and two 1,000's. That's 32,000. So hopefully that gives you an understanding. I mean it's fun to maybe do some lattice multiplication and get practice, but sometimes it looks like this bizarre magical thing. But hopefully from this video you understand that all it is is just a different way of keeping track of where the 1's, 10's, and 100's place are. With the advantage that it's kind of nice and compartmentalized, it doesn't take up a lot of space. And, it allows you to do all your multiplication at once and then, switch your brain into addition and carrying mode.