Ways for Two Dice to Sum to 8: $x_1 +x_2 =8$












1












$begingroup$


I am trying to determine the number of ways two dice can be rolled to have a sum of 8. This question can be easily brute-forced, but I want to use a technique that can be applied to similar questions that cannot be brute-forced so easily.



My solution was as follows,
$$x_1 +x_2 = 8, x_1, x_2 geq 1$$
$$x_1+1+x_2+1 = 8 rightarrow x_1+x_2 =6$$
Now I can use the stars and bars method.
So $C(6 +(2-1), (2-1))$ should be the answer, which is equal to $7$. But when I brute-forced this question, I only found $5$ solutions. What did I do wrong?



Normally if I wanted to find, say for example, the number of integer solutions to $x_1+x_2+x_3 = 12, x_i geq 0$, I would apply the same method and do $C(12+(3-1), (3-1))$. What am I doing wrong?










share|cite|improve this question











$endgroup$








  • 1




    $begingroup$
    have you considered the domain of $x$? there are certainly not 7 solutions for $x_1+x_2=6$ given your definition of $x_1$ and $x_2$ because there is another constraint $xleq 5$ in addition to $xgeq0$
    $endgroup$
    – MoonKnight
    Nov 30 '18 at 18:24












  • $begingroup$
    Once I manipulate the question to $x_1+x_2 = 6$, then $0 leq x leq 6$, correct? I got rid of the domain restriction by this manipulation.
    $endgroup$
    – Danielle
    Nov 30 '18 at 18:25






  • 1




    $begingroup$
    no, originally $1leq x leq 6$, since dice can only give you 1 through 6. After your transformation, $0 leq x leq 5$.
    $endgroup$
    – MoonKnight
    Nov 30 '18 at 18:33










  • $begingroup$
    I'm not sure about that. Since I'm turning the question into solving for integer solutions, the upper bound should not be affected by adding 1 to each x. Or have I misunderstood this technique?
    $endgroup$
    – Danielle
    Nov 30 '18 at 18:36






  • 1




    $begingroup$
    Well, if the original problem do not have an upper bound, your transformation would not result in one neither. But if the original problem has an upper bound ($xleq6$ here), then your transformation need to transform that upper bound as well.
    $endgroup$
    – MoonKnight
    Nov 30 '18 at 18:44
















1












$begingroup$


I am trying to determine the number of ways two dice can be rolled to have a sum of 8. This question can be easily brute-forced, but I want to use a technique that can be applied to similar questions that cannot be brute-forced so easily.



My solution was as follows,
$$x_1 +x_2 = 8, x_1, x_2 geq 1$$
$$x_1+1+x_2+1 = 8 rightarrow x_1+x_2 =6$$
Now I can use the stars and bars method.
So $C(6 +(2-1), (2-1))$ should be the answer, which is equal to $7$. But when I brute-forced this question, I only found $5$ solutions. What did I do wrong?



Normally if I wanted to find, say for example, the number of integer solutions to $x_1+x_2+x_3 = 12, x_i geq 0$, I would apply the same method and do $C(12+(3-1), (3-1))$. What am I doing wrong?










share|cite|improve this question











$endgroup$








  • 1




    $begingroup$
    have you considered the domain of $x$? there are certainly not 7 solutions for $x_1+x_2=6$ given your definition of $x_1$ and $x_2$ because there is another constraint $xleq 5$ in addition to $xgeq0$
    $endgroup$
    – MoonKnight
    Nov 30 '18 at 18:24












  • $begingroup$
    Once I manipulate the question to $x_1+x_2 = 6$, then $0 leq x leq 6$, correct? I got rid of the domain restriction by this manipulation.
    $endgroup$
    – Danielle
    Nov 30 '18 at 18:25






  • 1




    $begingroup$
    no, originally $1leq x leq 6$, since dice can only give you 1 through 6. After your transformation, $0 leq x leq 5$.
    $endgroup$
    – MoonKnight
    Nov 30 '18 at 18:33










  • $begingroup$
    I'm not sure about that. Since I'm turning the question into solving for integer solutions, the upper bound should not be affected by adding 1 to each x. Or have I misunderstood this technique?
    $endgroup$
    – Danielle
    Nov 30 '18 at 18:36






  • 1




    $begingroup$
    Well, if the original problem do not have an upper bound, your transformation would not result in one neither. But if the original problem has an upper bound ($xleq6$ here), then your transformation need to transform that upper bound as well.
    $endgroup$
    – MoonKnight
    Nov 30 '18 at 18:44














1












1








1





$begingroup$


I am trying to determine the number of ways two dice can be rolled to have a sum of 8. This question can be easily brute-forced, but I want to use a technique that can be applied to similar questions that cannot be brute-forced so easily.



My solution was as follows,
$$x_1 +x_2 = 8, x_1, x_2 geq 1$$
$$x_1+1+x_2+1 = 8 rightarrow x_1+x_2 =6$$
Now I can use the stars and bars method.
So $C(6 +(2-1), (2-1))$ should be the answer, which is equal to $7$. But when I brute-forced this question, I only found $5$ solutions. What did I do wrong?



Normally if I wanted to find, say for example, the number of integer solutions to $x_1+x_2+x_3 = 12, x_i geq 0$, I would apply the same method and do $C(12+(3-1), (3-1))$. What am I doing wrong?










share|cite|improve this question











$endgroup$




I am trying to determine the number of ways two dice can be rolled to have a sum of 8. This question can be easily brute-forced, but I want to use a technique that can be applied to similar questions that cannot be brute-forced so easily.



My solution was as follows,
$$x_1 +x_2 = 8, x_1, x_2 geq 1$$
$$x_1+1+x_2+1 = 8 rightarrow x_1+x_2 =6$$
Now I can use the stars and bars method.
So $C(6 +(2-1), (2-1))$ should be the answer, which is equal to $7$. But when I brute-forced this question, I only found $5$ solutions. What did I do wrong?



Normally if I wanted to find, say for example, the number of integer solutions to $x_1+x_2+x_3 = 12, x_i geq 0$, I would apply the same method and do $C(12+(3-1), (3-1))$. What am I doing wrong?







combinatorics dice






share|cite|improve this question















share|cite|improve this question













share|cite|improve this question




share|cite|improve this question








edited Nov 30 '18 at 20:03









N. F. Taussig

43.7k93355




43.7k93355










asked Nov 30 '18 at 18:20









DanielleDanielle

2179




2179








  • 1




    $begingroup$
    have you considered the domain of $x$? there are certainly not 7 solutions for $x_1+x_2=6$ given your definition of $x_1$ and $x_2$ because there is another constraint $xleq 5$ in addition to $xgeq0$
    $endgroup$
    – MoonKnight
    Nov 30 '18 at 18:24












  • $begingroup$
    Once I manipulate the question to $x_1+x_2 = 6$, then $0 leq x leq 6$, correct? I got rid of the domain restriction by this manipulation.
    $endgroup$
    – Danielle
    Nov 30 '18 at 18:25






  • 1




    $begingroup$
    no, originally $1leq x leq 6$, since dice can only give you 1 through 6. After your transformation, $0 leq x leq 5$.
    $endgroup$
    – MoonKnight
    Nov 30 '18 at 18:33










  • $begingroup$
    I'm not sure about that. Since I'm turning the question into solving for integer solutions, the upper bound should not be affected by adding 1 to each x. Or have I misunderstood this technique?
    $endgroup$
    – Danielle
    Nov 30 '18 at 18:36






  • 1




    $begingroup$
    Well, if the original problem do not have an upper bound, your transformation would not result in one neither. But if the original problem has an upper bound ($xleq6$ here), then your transformation need to transform that upper bound as well.
    $endgroup$
    – MoonKnight
    Nov 30 '18 at 18:44














  • 1




    $begingroup$
    have you considered the domain of $x$? there are certainly not 7 solutions for $x_1+x_2=6$ given your definition of $x_1$ and $x_2$ because there is another constraint $xleq 5$ in addition to $xgeq0$
    $endgroup$
    – MoonKnight
    Nov 30 '18 at 18:24












  • $begingroup$
    Once I manipulate the question to $x_1+x_2 = 6$, then $0 leq x leq 6$, correct? I got rid of the domain restriction by this manipulation.
    $endgroup$
    – Danielle
    Nov 30 '18 at 18:25






  • 1




    $begingroup$
    no, originally $1leq x leq 6$, since dice can only give you 1 through 6. After your transformation, $0 leq x leq 5$.
    $endgroup$
    – MoonKnight
    Nov 30 '18 at 18:33










  • $begingroup$
    I'm not sure about that. Since I'm turning the question into solving for integer solutions, the upper bound should not be affected by adding 1 to each x. Or have I misunderstood this technique?
    $endgroup$
    – Danielle
    Nov 30 '18 at 18:36






  • 1




    $begingroup$
    Well, if the original problem do not have an upper bound, your transformation would not result in one neither. But if the original problem has an upper bound ($xleq6$ here), then your transformation need to transform that upper bound as well.
    $endgroup$
    – MoonKnight
    Nov 30 '18 at 18:44








1




1




$begingroup$
have you considered the domain of $x$? there are certainly not 7 solutions for $x_1+x_2=6$ given your definition of $x_1$ and $x_2$ because there is another constraint $xleq 5$ in addition to $xgeq0$
$endgroup$
– MoonKnight
Nov 30 '18 at 18:24






$begingroup$
have you considered the domain of $x$? there are certainly not 7 solutions for $x_1+x_2=6$ given your definition of $x_1$ and $x_2$ because there is another constraint $xleq 5$ in addition to $xgeq0$
$endgroup$
– MoonKnight
Nov 30 '18 at 18:24














$begingroup$
Once I manipulate the question to $x_1+x_2 = 6$, then $0 leq x leq 6$, correct? I got rid of the domain restriction by this manipulation.
$endgroup$
– Danielle
Nov 30 '18 at 18:25




$begingroup$
Once I manipulate the question to $x_1+x_2 = 6$, then $0 leq x leq 6$, correct? I got rid of the domain restriction by this manipulation.
$endgroup$
– Danielle
Nov 30 '18 at 18:25




1




1




$begingroup$
no, originally $1leq x leq 6$, since dice can only give you 1 through 6. After your transformation, $0 leq x leq 5$.
$endgroup$
– MoonKnight
Nov 30 '18 at 18:33




$begingroup$
no, originally $1leq x leq 6$, since dice can only give you 1 through 6. After your transformation, $0 leq x leq 5$.
$endgroup$
– MoonKnight
Nov 30 '18 at 18:33












$begingroup$
I'm not sure about that. Since I'm turning the question into solving for integer solutions, the upper bound should not be affected by adding 1 to each x. Or have I misunderstood this technique?
$endgroup$
– Danielle
Nov 30 '18 at 18:36




$begingroup$
I'm not sure about that. Since I'm turning the question into solving for integer solutions, the upper bound should not be affected by adding 1 to each x. Or have I misunderstood this technique?
$endgroup$
– Danielle
Nov 30 '18 at 18:36




1




1




$begingroup$
Well, if the original problem do not have an upper bound, your transformation would not result in one neither. But if the original problem has an upper bound ($xleq6$ here), then your transformation need to transform that upper bound as well.
$endgroup$
– MoonKnight
Nov 30 '18 at 18:44




$begingroup$
Well, if the original problem do not have an upper bound, your transformation would not result in one neither. But if the original problem has an upper bound ($xleq6$ here), then your transformation need to transform that upper bound as well.
$endgroup$
– MoonKnight
Nov 30 '18 at 18:44










3 Answers
3






active

oldest

votes


















1












$begingroup$

You stated that you found that there are exactly $5$ solutions by listing all the possibilities, so I will assume you are considering six-sided dice.



What you have failed to do is account for the fact that the largest number a six-sided die can display is $6$.




If two six-sided dice are rolled, in how many ways can the numbers on the top faces of the dice sum to $8$?




Suppose we roll a blue die and a green die. Let $x_1$, $x_2$ denote, respectively, the numbers on the top face of the blue die and top face of the green die. Then
$$x_1 + x_2 = 8 tag{1}$$
where $1 leq x_1, x_2 leq 6$.



You opted to convert this to a problem in the nonnegative integers, so let's let $x_1' = x_1 - 1$ and $x_2' = x_2 - 1$. Since $x_1$ and $x_2$ are positive integers, $x_1'$ and $x_2'$ are nonnegative integers. Substituting $x_1' + 1$ for $x_1$ and $x_2' + 1$ for $x_2$ in equation 1 and simplifying yields
$$x_1' + x_2' = 6 tag{2}$$
Since equation 1 is an equation in positive integers not exceeding $6$, equation 2 is an equation in nonnegative integers not exceeding $5$.



Without that restriction, a solution of equation 2 corresponds to the placement of one addition sign in a row of six ones. For instance,
$$1 1 + 1 1 1 1$$
corresponds to the solution $x_1' = 2$ and $x_2' = 4$. The number of such solutions is
$$binom{6 + 2 - 1}{2 - 1} = binom{7}{1} = 7$$
since we must choose which of the seven positions required for six ones and one addition sign will be filled with an addition sign.



From these, we must subtract those solutions in which one of the variables exceeds $5$.



There are two ways to choose which variable violates the restriction that $x_1', x_2' leq 5$. If a variable does violate that restriction, it must be equal to $6$ and the other must be equal to $0$, so there is only one possible solution of equation 2 for each variable that could violate the restriction that a variable cannot exceed $5$. Since there are two such variables, there are two solutions of equation 2 in which an upper bound is violated.



Hence, the number of possible ways for the numbers on the top faces of two six-sided dice to have a sum of $8$ is
$$binom{7}{1} - binom{2}{1} = 7 - 2 = 5$$






share|cite|improve this answer









$endgroup$





















    1












    $begingroup$

    Hint: You have six choices for the first die, however if you roll a $1$ on the first die then no matter what you roll on the second die you cannot get to $8$. Thus, we want $x_1,x_2$ such that $x_1+x_2=8$ and $2leq x_1,x_2leq 6$.






    share|cite|improve this answer









    $endgroup$













    • $begingroup$
      This makes sense. As such, I would try to remove these restrictions through manipulation. So $x_1 +2 + x_2 +1 = 8 rightarrow x_1+x_2 = 5$. But this still ends up being C(6,1). I assume I do not have to worry about the upper bound on $x_2$ since the total sum is less than 6.
      $endgroup$
      – Danielle
      Nov 30 '18 at 18:33










    • $begingroup$
      Yeah basically you want $x_1$ between $0$ and $4$ because $x_2$ is uniquely determined once $x_1$ is chosen, so you get five combinations.
      $endgroup$
      – Dave
      Nov 30 '18 at 18:42



















    1












    $begingroup$

    Your sum count is wrong because you are including the cases where some of the $x_i>6.$ For example, $7+3+2=12,$ but you don't want to count that case.




    Summary:



    The number of ways of getting the sum $S$ from rolling $k$ fair $n$-sided dice labeled $1$ to $n$ is:
    $$sum_{i=0}^{k}(-1)^ibinom{k}{i}binom{S-in-1}{k-1}$$
    You have to be careful evalating this, because $binom{m}{j}$ is not a polynomial in $m$ for fixed $j$, since it is $0$ when $m<j.$



    A trick to make it easier is that you can replace any $S$ with $(n+1)k-S$ and get the same count. So you can always use $Sleq frac{(n+1)k}{2}$, in which case you can ignore terms with $i>frac{k+1}{2}$ since then $binom{S-in-1}{k-1}=0.$




    Longer:



    This is a case for either generating functions or an inclusion-exclusion argument.



    The generating approach applied to the sum of two fair six-sided dice is to compute the coefficients of $x^8$ in:



    $$(x+x^2+x^3+cdots + x^6)^2$$



    or the coefficient of $x^6$ in:



    $$(1+x+cdots+x^5)^2$$



    Now, this might seem harder, but it isn't, because $1+x+cdots+x^5=frac{1-x^6}{1-x}$ So you are seeking the coefficient if $x^6$ in:



    $$(1-x^6)^2frac{1}{(1-x)^2}tag{1}=(1-2x^6+x^{12})frac{1}{(1-x)^2}$$



    And we have the general result:



    $$frac{1}{(1-x)^{k}}=sum_{m=0}^{infty} binom{m+k-1}{k-1}x^m tag{2}$$



    So when $k=2$ you get:



    $$frac{1}{(1-x)^2}=sum_{m=0}^{infty}(m+1)x^m$$



    Multiplying by $(1-x^6)^2=1-2x^6+x^{12}$ this means the coefficient of $x^j$ in $(1)$ is:



    $$binom{j+1}{1}-2binom{j-5}{1}+binom{j-11}{1}$$



    We can't replace $binom{n}{1}$ with $n$ because we need to $binom{n}1=0$ when $n<1.$ The term $binom{j-11}{1}$ is only ever really used to "zero out" the cases where $binom{j+1}{1}-2binom{j-5}{1}$ is negative.



    When seeking the number of ways to get a sum of $8$, we use $j=8-2=6$ this is $7-2=5.$



    When the sum we are seeking, $S,$ is between $2$ and $7$, inclusive, this formula gives us a count of:



    $$S-1$$



    because the term $binom{S-7}{1}=0$ when $Sleq 7.$



    and when $7<Sleq 12$ this number of ways is:



    $$S-1 - 2(S-7)=13-S$$



    When $S>12$ all the sum is:



    $$(S-1)-2(S-7)+(S-13)=0$$



    And when $S<2$ all the terms are zero.





    More generally, if you are rolling a set of $k$ $n$-sided dice, you get,via $(2)$, then the number of ways to get a total of $S$ is the coefficient of $x^{S-k}$ in:



    $$(1-x^n)^ksum_{m=0}^{infty}binom{m+k-1}{k-1}x^m$$



    Thus we get that the number of ways to get a sum $S$ from $k$ $n$-sided dice is: $$sum_{i=0}^{k}(-1)^ibinom{k}{i}binom{S-in-1}{k-1}$$



    You can also prove this formula using inclusion-exclusion, rather than using generating functions.



    Example: when throwing three eight-sided dice, the number of ways to get a sum of $S$ is:



    $$binom{S-1}{2}-3binom{S-9}{2}+3binom{S-17}{2}-binom{S-25}{2}$$



    Which yields:
    $$begin{cases}
    0&S<3\
    frac{(S-1)(S-2)}{2}&3leq S<11\
    27S-S^2-134&11leq S<19\
    frac{(25-S)(26-S)}{2}&19leq Sleq 24\
    0&25leq S
    end{cases}
    $$



    You also have, in general, the value for $S$ and $(n+1)k-S$ should be equal. This means that you can always solve for $S$ by using at most the first $frac{k+1}{2}$ terms of the sum, with the other terms zero.





    When dealing with differently-sided dice, you get the same kind of result, but the numerator is different.



    For example, a four- and six-sided die rolled together gives you:



    $$x^2(1-x^4)(1-x^6)sum_{m=0}^{infty}binom{m+1}{1}x^m$$



    Then the coefficient of $x^S$ in this is:



    $$binom{S-1}{1}-binom{S-5}{1}-binom{S-7}{1}+binom{S-11}{1}$$






    share|cite|improve this answer











    $endgroup$













      Your Answer





      StackExchange.ifUsing("editor", function () {
      return StackExchange.using("mathjaxEditing", function () {
      StackExchange.MarkdownEditor.creationCallbacks.add(function (editor, postfix) {
      StackExchange.mathjaxEditing.prepareWmdForMathJax(editor, postfix, [["$", "$"], ["\\(","\\)"]]);
      });
      });
      }, "mathjax-editing");

      StackExchange.ready(function() {
      var channelOptions = {
      tags: "".split(" "),
      id: "69"
      };
      initTagRenderer("".split(" "), "".split(" "), channelOptions);

      StackExchange.using("externalEditor", function() {
      // Have to fire editor after snippets, if snippets enabled
      if (StackExchange.settings.snippets.snippetsEnabled) {
      StackExchange.using("snippets", function() {
      createEditor();
      });
      }
      else {
      createEditor();
      }
      });

      function createEditor() {
      StackExchange.prepareEditor({
      heartbeatType: 'answer',
      autoActivateHeartbeat: false,
      convertImagesToLinks: true,
      noModals: true,
      showLowRepImageUploadWarning: true,
      reputationToPostImages: 10,
      bindNavPrevention: true,
      postfix: "",
      imageUploader: {
      brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
      contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
      allowUrls: true
      },
      noCode: true, onDemand: true,
      discardSelector: ".discard-answer"
      ,immediatelyShowMarkdownHelp:true
      });


      }
      });














      draft saved

      draft discarded


















      StackExchange.ready(
      function () {
      StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3020438%2fways-for-two-dice-to-sum-to-8-x-1-x-2-8%23new-answer', 'question_page');
      }
      );

      Post as a guest















      Required, but never shown

























      3 Answers
      3






      active

      oldest

      votes








      3 Answers
      3






      active

      oldest

      votes









      active

      oldest

      votes






      active

      oldest

      votes









      1












      $begingroup$

      You stated that you found that there are exactly $5$ solutions by listing all the possibilities, so I will assume you are considering six-sided dice.



      What you have failed to do is account for the fact that the largest number a six-sided die can display is $6$.




      If two six-sided dice are rolled, in how many ways can the numbers on the top faces of the dice sum to $8$?




      Suppose we roll a blue die and a green die. Let $x_1$, $x_2$ denote, respectively, the numbers on the top face of the blue die and top face of the green die. Then
      $$x_1 + x_2 = 8 tag{1}$$
      where $1 leq x_1, x_2 leq 6$.



      You opted to convert this to a problem in the nonnegative integers, so let's let $x_1' = x_1 - 1$ and $x_2' = x_2 - 1$. Since $x_1$ and $x_2$ are positive integers, $x_1'$ and $x_2'$ are nonnegative integers. Substituting $x_1' + 1$ for $x_1$ and $x_2' + 1$ for $x_2$ in equation 1 and simplifying yields
      $$x_1' + x_2' = 6 tag{2}$$
      Since equation 1 is an equation in positive integers not exceeding $6$, equation 2 is an equation in nonnegative integers not exceeding $5$.



      Without that restriction, a solution of equation 2 corresponds to the placement of one addition sign in a row of six ones. For instance,
      $$1 1 + 1 1 1 1$$
      corresponds to the solution $x_1' = 2$ and $x_2' = 4$. The number of such solutions is
      $$binom{6 + 2 - 1}{2 - 1} = binom{7}{1} = 7$$
      since we must choose which of the seven positions required for six ones and one addition sign will be filled with an addition sign.



      From these, we must subtract those solutions in which one of the variables exceeds $5$.



      There are two ways to choose which variable violates the restriction that $x_1', x_2' leq 5$. If a variable does violate that restriction, it must be equal to $6$ and the other must be equal to $0$, so there is only one possible solution of equation 2 for each variable that could violate the restriction that a variable cannot exceed $5$. Since there are two such variables, there are two solutions of equation 2 in which an upper bound is violated.



      Hence, the number of possible ways for the numbers on the top faces of two six-sided dice to have a sum of $8$ is
      $$binom{7}{1} - binom{2}{1} = 7 - 2 = 5$$






      share|cite|improve this answer









      $endgroup$


















        1












        $begingroup$

        You stated that you found that there are exactly $5$ solutions by listing all the possibilities, so I will assume you are considering six-sided dice.



        What you have failed to do is account for the fact that the largest number a six-sided die can display is $6$.




        If two six-sided dice are rolled, in how many ways can the numbers on the top faces of the dice sum to $8$?




        Suppose we roll a blue die and a green die. Let $x_1$, $x_2$ denote, respectively, the numbers on the top face of the blue die and top face of the green die. Then
        $$x_1 + x_2 = 8 tag{1}$$
        where $1 leq x_1, x_2 leq 6$.



        You opted to convert this to a problem in the nonnegative integers, so let's let $x_1' = x_1 - 1$ and $x_2' = x_2 - 1$. Since $x_1$ and $x_2$ are positive integers, $x_1'$ and $x_2'$ are nonnegative integers. Substituting $x_1' + 1$ for $x_1$ and $x_2' + 1$ for $x_2$ in equation 1 and simplifying yields
        $$x_1' + x_2' = 6 tag{2}$$
        Since equation 1 is an equation in positive integers not exceeding $6$, equation 2 is an equation in nonnegative integers not exceeding $5$.



        Without that restriction, a solution of equation 2 corresponds to the placement of one addition sign in a row of six ones. For instance,
        $$1 1 + 1 1 1 1$$
        corresponds to the solution $x_1' = 2$ and $x_2' = 4$. The number of such solutions is
        $$binom{6 + 2 - 1}{2 - 1} = binom{7}{1} = 7$$
        since we must choose which of the seven positions required for six ones and one addition sign will be filled with an addition sign.



        From these, we must subtract those solutions in which one of the variables exceeds $5$.



        There are two ways to choose which variable violates the restriction that $x_1', x_2' leq 5$. If a variable does violate that restriction, it must be equal to $6$ and the other must be equal to $0$, so there is only one possible solution of equation 2 for each variable that could violate the restriction that a variable cannot exceed $5$. Since there are two such variables, there are two solutions of equation 2 in which an upper bound is violated.



        Hence, the number of possible ways for the numbers on the top faces of two six-sided dice to have a sum of $8$ is
        $$binom{7}{1} - binom{2}{1} = 7 - 2 = 5$$






        share|cite|improve this answer









        $endgroup$
















          1












          1








          1





          $begingroup$

          You stated that you found that there are exactly $5$ solutions by listing all the possibilities, so I will assume you are considering six-sided dice.



          What you have failed to do is account for the fact that the largest number a six-sided die can display is $6$.




          If two six-sided dice are rolled, in how many ways can the numbers on the top faces of the dice sum to $8$?




          Suppose we roll a blue die and a green die. Let $x_1$, $x_2$ denote, respectively, the numbers on the top face of the blue die and top face of the green die. Then
          $$x_1 + x_2 = 8 tag{1}$$
          where $1 leq x_1, x_2 leq 6$.



          You opted to convert this to a problem in the nonnegative integers, so let's let $x_1' = x_1 - 1$ and $x_2' = x_2 - 1$. Since $x_1$ and $x_2$ are positive integers, $x_1'$ and $x_2'$ are nonnegative integers. Substituting $x_1' + 1$ for $x_1$ and $x_2' + 1$ for $x_2$ in equation 1 and simplifying yields
          $$x_1' + x_2' = 6 tag{2}$$
          Since equation 1 is an equation in positive integers not exceeding $6$, equation 2 is an equation in nonnegative integers not exceeding $5$.



          Without that restriction, a solution of equation 2 corresponds to the placement of one addition sign in a row of six ones. For instance,
          $$1 1 + 1 1 1 1$$
          corresponds to the solution $x_1' = 2$ and $x_2' = 4$. The number of such solutions is
          $$binom{6 + 2 - 1}{2 - 1} = binom{7}{1} = 7$$
          since we must choose which of the seven positions required for six ones and one addition sign will be filled with an addition sign.



          From these, we must subtract those solutions in which one of the variables exceeds $5$.



          There are two ways to choose which variable violates the restriction that $x_1', x_2' leq 5$. If a variable does violate that restriction, it must be equal to $6$ and the other must be equal to $0$, so there is only one possible solution of equation 2 for each variable that could violate the restriction that a variable cannot exceed $5$. Since there are two such variables, there are two solutions of equation 2 in which an upper bound is violated.



          Hence, the number of possible ways for the numbers on the top faces of two six-sided dice to have a sum of $8$ is
          $$binom{7}{1} - binom{2}{1} = 7 - 2 = 5$$






          share|cite|improve this answer









          $endgroup$



          You stated that you found that there are exactly $5$ solutions by listing all the possibilities, so I will assume you are considering six-sided dice.



          What you have failed to do is account for the fact that the largest number a six-sided die can display is $6$.




          If two six-sided dice are rolled, in how many ways can the numbers on the top faces of the dice sum to $8$?




          Suppose we roll a blue die and a green die. Let $x_1$, $x_2$ denote, respectively, the numbers on the top face of the blue die and top face of the green die. Then
          $$x_1 + x_2 = 8 tag{1}$$
          where $1 leq x_1, x_2 leq 6$.



          You opted to convert this to a problem in the nonnegative integers, so let's let $x_1' = x_1 - 1$ and $x_2' = x_2 - 1$. Since $x_1$ and $x_2$ are positive integers, $x_1'$ and $x_2'$ are nonnegative integers. Substituting $x_1' + 1$ for $x_1$ and $x_2' + 1$ for $x_2$ in equation 1 and simplifying yields
          $$x_1' + x_2' = 6 tag{2}$$
          Since equation 1 is an equation in positive integers not exceeding $6$, equation 2 is an equation in nonnegative integers not exceeding $5$.



          Without that restriction, a solution of equation 2 corresponds to the placement of one addition sign in a row of six ones. For instance,
          $$1 1 + 1 1 1 1$$
          corresponds to the solution $x_1' = 2$ and $x_2' = 4$. The number of such solutions is
          $$binom{6 + 2 - 1}{2 - 1} = binom{7}{1} = 7$$
          since we must choose which of the seven positions required for six ones and one addition sign will be filled with an addition sign.



          From these, we must subtract those solutions in which one of the variables exceeds $5$.



          There are two ways to choose which variable violates the restriction that $x_1', x_2' leq 5$. If a variable does violate that restriction, it must be equal to $6$ and the other must be equal to $0$, so there is only one possible solution of equation 2 for each variable that could violate the restriction that a variable cannot exceed $5$. Since there are two such variables, there are two solutions of equation 2 in which an upper bound is violated.



          Hence, the number of possible ways for the numbers on the top faces of two six-sided dice to have a sum of $8$ is
          $$binom{7}{1} - binom{2}{1} = 7 - 2 = 5$$







          share|cite|improve this answer












          share|cite|improve this answer



          share|cite|improve this answer










          answered Nov 30 '18 at 20:45









          N. F. TaussigN. F. Taussig

          43.7k93355




          43.7k93355























              1












              $begingroup$

              Hint: You have six choices for the first die, however if you roll a $1$ on the first die then no matter what you roll on the second die you cannot get to $8$. Thus, we want $x_1,x_2$ such that $x_1+x_2=8$ and $2leq x_1,x_2leq 6$.






              share|cite|improve this answer









              $endgroup$













              • $begingroup$
                This makes sense. As such, I would try to remove these restrictions through manipulation. So $x_1 +2 + x_2 +1 = 8 rightarrow x_1+x_2 = 5$. But this still ends up being C(6,1). I assume I do not have to worry about the upper bound on $x_2$ since the total sum is less than 6.
                $endgroup$
                – Danielle
                Nov 30 '18 at 18:33










              • $begingroup$
                Yeah basically you want $x_1$ between $0$ and $4$ because $x_2$ is uniquely determined once $x_1$ is chosen, so you get five combinations.
                $endgroup$
                – Dave
                Nov 30 '18 at 18:42
















              1












              $begingroup$

              Hint: You have six choices for the first die, however if you roll a $1$ on the first die then no matter what you roll on the second die you cannot get to $8$. Thus, we want $x_1,x_2$ such that $x_1+x_2=8$ and $2leq x_1,x_2leq 6$.






              share|cite|improve this answer









              $endgroup$













              • $begingroup$
                This makes sense. As such, I would try to remove these restrictions through manipulation. So $x_1 +2 + x_2 +1 = 8 rightarrow x_1+x_2 = 5$. But this still ends up being C(6,1). I assume I do not have to worry about the upper bound on $x_2$ since the total sum is less than 6.
                $endgroup$
                – Danielle
                Nov 30 '18 at 18:33










              • $begingroup$
                Yeah basically you want $x_1$ between $0$ and $4$ because $x_2$ is uniquely determined once $x_1$ is chosen, so you get five combinations.
                $endgroup$
                – Dave
                Nov 30 '18 at 18:42














              1












              1








              1





              $begingroup$

              Hint: You have six choices for the first die, however if you roll a $1$ on the first die then no matter what you roll on the second die you cannot get to $8$. Thus, we want $x_1,x_2$ such that $x_1+x_2=8$ and $2leq x_1,x_2leq 6$.






              share|cite|improve this answer









              $endgroup$



              Hint: You have six choices for the first die, however if you roll a $1$ on the first die then no matter what you roll on the second die you cannot get to $8$. Thus, we want $x_1,x_2$ such that $x_1+x_2=8$ and $2leq x_1,x_2leq 6$.







              share|cite|improve this answer












              share|cite|improve this answer



              share|cite|improve this answer










              answered Nov 30 '18 at 18:25









              DaveDave

              8,77711033




              8,77711033












              • $begingroup$
                This makes sense. As such, I would try to remove these restrictions through manipulation. So $x_1 +2 + x_2 +1 = 8 rightarrow x_1+x_2 = 5$. But this still ends up being C(6,1). I assume I do not have to worry about the upper bound on $x_2$ since the total sum is less than 6.
                $endgroup$
                – Danielle
                Nov 30 '18 at 18:33










              • $begingroup$
                Yeah basically you want $x_1$ between $0$ and $4$ because $x_2$ is uniquely determined once $x_1$ is chosen, so you get five combinations.
                $endgroup$
                – Dave
                Nov 30 '18 at 18:42


















              • $begingroup$
                This makes sense. As such, I would try to remove these restrictions through manipulation. So $x_1 +2 + x_2 +1 = 8 rightarrow x_1+x_2 = 5$. But this still ends up being C(6,1). I assume I do not have to worry about the upper bound on $x_2$ since the total sum is less than 6.
                $endgroup$
                – Danielle
                Nov 30 '18 at 18:33










              • $begingroup$
                Yeah basically you want $x_1$ between $0$ and $4$ because $x_2$ is uniquely determined once $x_1$ is chosen, so you get five combinations.
                $endgroup$
                – Dave
                Nov 30 '18 at 18:42
















              $begingroup$
              This makes sense. As such, I would try to remove these restrictions through manipulation. So $x_1 +2 + x_2 +1 = 8 rightarrow x_1+x_2 = 5$. But this still ends up being C(6,1). I assume I do not have to worry about the upper bound on $x_2$ since the total sum is less than 6.
              $endgroup$
              – Danielle
              Nov 30 '18 at 18:33




              $begingroup$
              This makes sense. As such, I would try to remove these restrictions through manipulation. So $x_1 +2 + x_2 +1 = 8 rightarrow x_1+x_2 = 5$. But this still ends up being C(6,1). I assume I do not have to worry about the upper bound on $x_2$ since the total sum is less than 6.
              $endgroup$
              – Danielle
              Nov 30 '18 at 18:33












              $begingroup$
              Yeah basically you want $x_1$ between $0$ and $4$ because $x_2$ is uniquely determined once $x_1$ is chosen, so you get five combinations.
              $endgroup$
              – Dave
              Nov 30 '18 at 18:42




              $begingroup$
              Yeah basically you want $x_1$ between $0$ and $4$ because $x_2$ is uniquely determined once $x_1$ is chosen, so you get five combinations.
              $endgroup$
              – Dave
              Nov 30 '18 at 18:42











              1












              $begingroup$

              Your sum count is wrong because you are including the cases where some of the $x_i>6.$ For example, $7+3+2=12,$ but you don't want to count that case.




              Summary:



              The number of ways of getting the sum $S$ from rolling $k$ fair $n$-sided dice labeled $1$ to $n$ is:
              $$sum_{i=0}^{k}(-1)^ibinom{k}{i}binom{S-in-1}{k-1}$$
              You have to be careful evalating this, because $binom{m}{j}$ is not a polynomial in $m$ for fixed $j$, since it is $0$ when $m<j.$



              A trick to make it easier is that you can replace any $S$ with $(n+1)k-S$ and get the same count. So you can always use $Sleq frac{(n+1)k}{2}$, in which case you can ignore terms with $i>frac{k+1}{2}$ since then $binom{S-in-1}{k-1}=0.$




              Longer:



              This is a case for either generating functions or an inclusion-exclusion argument.



              The generating approach applied to the sum of two fair six-sided dice is to compute the coefficients of $x^8$ in:



              $$(x+x^2+x^3+cdots + x^6)^2$$



              or the coefficient of $x^6$ in:



              $$(1+x+cdots+x^5)^2$$



              Now, this might seem harder, but it isn't, because $1+x+cdots+x^5=frac{1-x^6}{1-x}$ So you are seeking the coefficient if $x^6$ in:



              $$(1-x^6)^2frac{1}{(1-x)^2}tag{1}=(1-2x^6+x^{12})frac{1}{(1-x)^2}$$



              And we have the general result:



              $$frac{1}{(1-x)^{k}}=sum_{m=0}^{infty} binom{m+k-1}{k-1}x^m tag{2}$$



              So when $k=2$ you get:



              $$frac{1}{(1-x)^2}=sum_{m=0}^{infty}(m+1)x^m$$



              Multiplying by $(1-x^6)^2=1-2x^6+x^{12}$ this means the coefficient of $x^j$ in $(1)$ is:



              $$binom{j+1}{1}-2binom{j-5}{1}+binom{j-11}{1}$$



              We can't replace $binom{n}{1}$ with $n$ because we need to $binom{n}1=0$ when $n<1.$ The term $binom{j-11}{1}$ is only ever really used to "zero out" the cases where $binom{j+1}{1}-2binom{j-5}{1}$ is negative.



              When seeking the number of ways to get a sum of $8$, we use $j=8-2=6$ this is $7-2=5.$



              When the sum we are seeking, $S,$ is between $2$ and $7$, inclusive, this formula gives us a count of:



              $$S-1$$



              because the term $binom{S-7}{1}=0$ when $Sleq 7.$



              and when $7<Sleq 12$ this number of ways is:



              $$S-1 - 2(S-7)=13-S$$



              When $S>12$ all the sum is:



              $$(S-1)-2(S-7)+(S-13)=0$$



              And when $S<2$ all the terms are zero.





              More generally, if you are rolling a set of $k$ $n$-sided dice, you get,via $(2)$, then the number of ways to get a total of $S$ is the coefficient of $x^{S-k}$ in:



              $$(1-x^n)^ksum_{m=0}^{infty}binom{m+k-1}{k-1}x^m$$



              Thus we get that the number of ways to get a sum $S$ from $k$ $n$-sided dice is: $$sum_{i=0}^{k}(-1)^ibinom{k}{i}binom{S-in-1}{k-1}$$



              You can also prove this formula using inclusion-exclusion, rather than using generating functions.



              Example: when throwing three eight-sided dice, the number of ways to get a sum of $S$ is:



              $$binom{S-1}{2}-3binom{S-9}{2}+3binom{S-17}{2}-binom{S-25}{2}$$



              Which yields:
              $$begin{cases}
              0&S<3\
              frac{(S-1)(S-2)}{2}&3leq S<11\
              27S-S^2-134&11leq S<19\
              frac{(25-S)(26-S)}{2}&19leq Sleq 24\
              0&25leq S
              end{cases}
              $$



              You also have, in general, the value for $S$ and $(n+1)k-S$ should be equal. This means that you can always solve for $S$ by using at most the first $frac{k+1}{2}$ terms of the sum, with the other terms zero.





              When dealing with differently-sided dice, you get the same kind of result, but the numerator is different.



              For example, a four- and six-sided die rolled together gives you:



              $$x^2(1-x^4)(1-x^6)sum_{m=0}^{infty}binom{m+1}{1}x^m$$



              Then the coefficient of $x^S$ in this is:



              $$binom{S-1}{1}-binom{S-5}{1}-binom{S-7}{1}+binom{S-11}{1}$$






              share|cite|improve this answer











              $endgroup$


















                1












                $begingroup$

                Your sum count is wrong because you are including the cases where some of the $x_i>6.$ For example, $7+3+2=12,$ but you don't want to count that case.




                Summary:



                The number of ways of getting the sum $S$ from rolling $k$ fair $n$-sided dice labeled $1$ to $n$ is:
                $$sum_{i=0}^{k}(-1)^ibinom{k}{i}binom{S-in-1}{k-1}$$
                You have to be careful evalating this, because $binom{m}{j}$ is not a polynomial in $m$ for fixed $j$, since it is $0$ when $m<j.$



                A trick to make it easier is that you can replace any $S$ with $(n+1)k-S$ and get the same count. So you can always use $Sleq frac{(n+1)k}{2}$, in which case you can ignore terms with $i>frac{k+1}{2}$ since then $binom{S-in-1}{k-1}=0.$




                Longer:



                This is a case for either generating functions or an inclusion-exclusion argument.



                The generating approach applied to the sum of two fair six-sided dice is to compute the coefficients of $x^8$ in:



                $$(x+x^2+x^3+cdots + x^6)^2$$



                or the coefficient of $x^6$ in:



                $$(1+x+cdots+x^5)^2$$



                Now, this might seem harder, but it isn't, because $1+x+cdots+x^5=frac{1-x^6}{1-x}$ So you are seeking the coefficient if $x^6$ in:



                $$(1-x^6)^2frac{1}{(1-x)^2}tag{1}=(1-2x^6+x^{12})frac{1}{(1-x)^2}$$



                And we have the general result:



                $$frac{1}{(1-x)^{k}}=sum_{m=0}^{infty} binom{m+k-1}{k-1}x^m tag{2}$$



                So when $k=2$ you get:



                $$frac{1}{(1-x)^2}=sum_{m=0}^{infty}(m+1)x^m$$



                Multiplying by $(1-x^6)^2=1-2x^6+x^{12}$ this means the coefficient of $x^j$ in $(1)$ is:



                $$binom{j+1}{1}-2binom{j-5}{1}+binom{j-11}{1}$$



                We can't replace $binom{n}{1}$ with $n$ because we need to $binom{n}1=0$ when $n<1.$ The term $binom{j-11}{1}$ is only ever really used to "zero out" the cases where $binom{j+1}{1}-2binom{j-5}{1}$ is negative.



                When seeking the number of ways to get a sum of $8$, we use $j=8-2=6$ this is $7-2=5.$



                When the sum we are seeking, $S,$ is between $2$ and $7$, inclusive, this formula gives us a count of:



                $$S-1$$



                because the term $binom{S-7}{1}=0$ when $Sleq 7.$



                and when $7<Sleq 12$ this number of ways is:



                $$S-1 - 2(S-7)=13-S$$



                When $S>12$ all the sum is:



                $$(S-1)-2(S-7)+(S-13)=0$$



                And when $S<2$ all the terms are zero.





                More generally, if you are rolling a set of $k$ $n$-sided dice, you get,via $(2)$, then the number of ways to get a total of $S$ is the coefficient of $x^{S-k}$ in:



                $$(1-x^n)^ksum_{m=0}^{infty}binom{m+k-1}{k-1}x^m$$



                Thus we get that the number of ways to get a sum $S$ from $k$ $n$-sided dice is: $$sum_{i=0}^{k}(-1)^ibinom{k}{i}binom{S-in-1}{k-1}$$



                You can also prove this formula using inclusion-exclusion, rather than using generating functions.



                Example: when throwing three eight-sided dice, the number of ways to get a sum of $S$ is:



                $$binom{S-1}{2}-3binom{S-9}{2}+3binom{S-17}{2}-binom{S-25}{2}$$



                Which yields:
                $$begin{cases}
                0&S<3\
                frac{(S-1)(S-2)}{2}&3leq S<11\
                27S-S^2-134&11leq S<19\
                frac{(25-S)(26-S)}{2}&19leq Sleq 24\
                0&25leq S
                end{cases}
                $$



                You also have, in general, the value for $S$ and $(n+1)k-S$ should be equal. This means that you can always solve for $S$ by using at most the first $frac{k+1}{2}$ terms of the sum, with the other terms zero.





                When dealing with differently-sided dice, you get the same kind of result, but the numerator is different.



                For example, a four- and six-sided die rolled together gives you:



                $$x^2(1-x^4)(1-x^6)sum_{m=0}^{infty}binom{m+1}{1}x^m$$



                Then the coefficient of $x^S$ in this is:



                $$binom{S-1}{1}-binom{S-5}{1}-binom{S-7}{1}+binom{S-11}{1}$$






                share|cite|improve this answer











                $endgroup$
















                  1












                  1








                  1





                  $begingroup$

                  Your sum count is wrong because you are including the cases where some of the $x_i>6.$ For example, $7+3+2=12,$ but you don't want to count that case.




                  Summary:



                  The number of ways of getting the sum $S$ from rolling $k$ fair $n$-sided dice labeled $1$ to $n$ is:
                  $$sum_{i=0}^{k}(-1)^ibinom{k}{i}binom{S-in-1}{k-1}$$
                  You have to be careful evalating this, because $binom{m}{j}$ is not a polynomial in $m$ for fixed $j$, since it is $0$ when $m<j.$



                  A trick to make it easier is that you can replace any $S$ with $(n+1)k-S$ and get the same count. So you can always use $Sleq frac{(n+1)k}{2}$, in which case you can ignore terms with $i>frac{k+1}{2}$ since then $binom{S-in-1}{k-1}=0.$




                  Longer:



                  This is a case for either generating functions or an inclusion-exclusion argument.



                  The generating approach applied to the sum of two fair six-sided dice is to compute the coefficients of $x^8$ in:



                  $$(x+x^2+x^3+cdots + x^6)^2$$



                  or the coefficient of $x^6$ in:



                  $$(1+x+cdots+x^5)^2$$



                  Now, this might seem harder, but it isn't, because $1+x+cdots+x^5=frac{1-x^6}{1-x}$ So you are seeking the coefficient if $x^6$ in:



                  $$(1-x^6)^2frac{1}{(1-x)^2}tag{1}=(1-2x^6+x^{12})frac{1}{(1-x)^2}$$



                  And we have the general result:



                  $$frac{1}{(1-x)^{k}}=sum_{m=0}^{infty} binom{m+k-1}{k-1}x^m tag{2}$$



                  So when $k=2$ you get:



                  $$frac{1}{(1-x)^2}=sum_{m=0}^{infty}(m+1)x^m$$



                  Multiplying by $(1-x^6)^2=1-2x^6+x^{12}$ this means the coefficient of $x^j$ in $(1)$ is:



                  $$binom{j+1}{1}-2binom{j-5}{1}+binom{j-11}{1}$$



                  We can't replace $binom{n}{1}$ with $n$ because we need to $binom{n}1=0$ when $n<1.$ The term $binom{j-11}{1}$ is only ever really used to "zero out" the cases where $binom{j+1}{1}-2binom{j-5}{1}$ is negative.



                  When seeking the number of ways to get a sum of $8$, we use $j=8-2=6$ this is $7-2=5.$



                  When the sum we are seeking, $S,$ is between $2$ and $7$, inclusive, this formula gives us a count of:



                  $$S-1$$



                  because the term $binom{S-7}{1}=0$ when $Sleq 7.$



                  and when $7<Sleq 12$ this number of ways is:



                  $$S-1 - 2(S-7)=13-S$$



                  When $S>12$ all the sum is:



                  $$(S-1)-2(S-7)+(S-13)=0$$



                  And when $S<2$ all the terms are zero.





                  More generally, if you are rolling a set of $k$ $n$-sided dice, you get,via $(2)$, then the number of ways to get a total of $S$ is the coefficient of $x^{S-k}$ in:



                  $$(1-x^n)^ksum_{m=0}^{infty}binom{m+k-1}{k-1}x^m$$



                  Thus we get that the number of ways to get a sum $S$ from $k$ $n$-sided dice is: $$sum_{i=0}^{k}(-1)^ibinom{k}{i}binom{S-in-1}{k-1}$$



                  You can also prove this formula using inclusion-exclusion, rather than using generating functions.



                  Example: when throwing three eight-sided dice, the number of ways to get a sum of $S$ is:



                  $$binom{S-1}{2}-3binom{S-9}{2}+3binom{S-17}{2}-binom{S-25}{2}$$



                  Which yields:
                  $$begin{cases}
                  0&S<3\
                  frac{(S-1)(S-2)}{2}&3leq S<11\
                  27S-S^2-134&11leq S<19\
                  frac{(25-S)(26-S)}{2}&19leq Sleq 24\
                  0&25leq S
                  end{cases}
                  $$



                  You also have, in general, the value for $S$ and $(n+1)k-S$ should be equal. This means that you can always solve for $S$ by using at most the first $frac{k+1}{2}$ terms of the sum, with the other terms zero.





                  When dealing with differently-sided dice, you get the same kind of result, but the numerator is different.



                  For example, a four- and six-sided die rolled together gives you:



                  $$x^2(1-x^4)(1-x^6)sum_{m=0}^{infty}binom{m+1}{1}x^m$$



                  Then the coefficient of $x^S$ in this is:



                  $$binom{S-1}{1}-binom{S-5}{1}-binom{S-7}{1}+binom{S-11}{1}$$






                  share|cite|improve this answer











                  $endgroup$



                  Your sum count is wrong because you are including the cases where some of the $x_i>6.$ For example, $7+3+2=12,$ but you don't want to count that case.




                  Summary:



                  The number of ways of getting the sum $S$ from rolling $k$ fair $n$-sided dice labeled $1$ to $n$ is:
                  $$sum_{i=0}^{k}(-1)^ibinom{k}{i}binom{S-in-1}{k-1}$$
                  You have to be careful evalating this, because $binom{m}{j}$ is not a polynomial in $m$ for fixed $j$, since it is $0$ when $m<j.$



                  A trick to make it easier is that you can replace any $S$ with $(n+1)k-S$ and get the same count. So you can always use $Sleq frac{(n+1)k}{2}$, in which case you can ignore terms with $i>frac{k+1}{2}$ since then $binom{S-in-1}{k-1}=0.$




                  Longer:



                  This is a case for either generating functions or an inclusion-exclusion argument.



                  The generating approach applied to the sum of two fair six-sided dice is to compute the coefficients of $x^8$ in:



                  $$(x+x^2+x^3+cdots + x^6)^2$$



                  or the coefficient of $x^6$ in:



                  $$(1+x+cdots+x^5)^2$$



                  Now, this might seem harder, but it isn't, because $1+x+cdots+x^5=frac{1-x^6}{1-x}$ So you are seeking the coefficient if $x^6$ in:



                  $$(1-x^6)^2frac{1}{(1-x)^2}tag{1}=(1-2x^6+x^{12})frac{1}{(1-x)^2}$$



                  And we have the general result:



                  $$frac{1}{(1-x)^{k}}=sum_{m=0}^{infty} binom{m+k-1}{k-1}x^m tag{2}$$



                  So when $k=2$ you get:



                  $$frac{1}{(1-x)^2}=sum_{m=0}^{infty}(m+1)x^m$$



                  Multiplying by $(1-x^6)^2=1-2x^6+x^{12}$ this means the coefficient of $x^j$ in $(1)$ is:



                  $$binom{j+1}{1}-2binom{j-5}{1}+binom{j-11}{1}$$



                  We can't replace $binom{n}{1}$ with $n$ because we need to $binom{n}1=0$ when $n<1.$ The term $binom{j-11}{1}$ is only ever really used to "zero out" the cases where $binom{j+1}{1}-2binom{j-5}{1}$ is negative.



                  When seeking the number of ways to get a sum of $8$, we use $j=8-2=6$ this is $7-2=5.$



                  When the sum we are seeking, $S,$ is between $2$ and $7$, inclusive, this formula gives us a count of:



                  $$S-1$$



                  because the term $binom{S-7}{1}=0$ when $Sleq 7.$



                  and when $7<Sleq 12$ this number of ways is:



                  $$S-1 - 2(S-7)=13-S$$



                  When $S>12$ all the sum is:



                  $$(S-1)-2(S-7)+(S-13)=0$$



                  And when $S<2$ all the terms are zero.





                  More generally, if you are rolling a set of $k$ $n$-sided dice, you get,via $(2)$, then the number of ways to get a total of $S$ is the coefficient of $x^{S-k}$ in:



                  $$(1-x^n)^ksum_{m=0}^{infty}binom{m+k-1}{k-1}x^m$$



                  Thus we get that the number of ways to get a sum $S$ from $k$ $n$-sided dice is: $$sum_{i=0}^{k}(-1)^ibinom{k}{i}binom{S-in-1}{k-1}$$



                  You can also prove this formula using inclusion-exclusion, rather than using generating functions.



                  Example: when throwing three eight-sided dice, the number of ways to get a sum of $S$ is:



                  $$binom{S-1}{2}-3binom{S-9}{2}+3binom{S-17}{2}-binom{S-25}{2}$$



                  Which yields:
                  $$begin{cases}
                  0&S<3\
                  frac{(S-1)(S-2)}{2}&3leq S<11\
                  27S-S^2-134&11leq S<19\
                  frac{(25-S)(26-S)}{2}&19leq Sleq 24\
                  0&25leq S
                  end{cases}
                  $$



                  You also have, in general, the value for $S$ and $(n+1)k-S$ should be equal. This means that you can always solve for $S$ by using at most the first $frac{k+1}{2}$ terms of the sum, with the other terms zero.





                  When dealing with differently-sided dice, you get the same kind of result, but the numerator is different.



                  For example, a four- and six-sided die rolled together gives you:



                  $$x^2(1-x^4)(1-x^6)sum_{m=0}^{infty}binom{m+1}{1}x^m$$



                  Then the coefficient of $x^S$ in this is:



                  $$binom{S-1}{1}-binom{S-5}{1}-binom{S-7}{1}+binom{S-11}{1}$$







                  share|cite|improve this answer














                  share|cite|improve this answer



                  share|cite|improve this answer








                  edited Nov 30 '18 at 21:04

























                  answered Nov 30 '18 at 19:11









                  Thomas AndrewsThomas Andrews

                  130k11146297




                  130k11146297






























                      draft saved

                      draft discarded




















































                      Thanks for contributing an answer to Mathematics Stack Exchange!


                      • Please be sure to answer the question. Provide details and share your research!

                      But avoid



                      • Asking for help, clarification, or responding to other answers.

                      • Making statements based on opinion; back them up with references or personal experience.


                      Use MathJax to format equations. MathJax reference.


                      To learn more, see our tips on writing great answers.




                      draft saved


                      draft discarded














                      StackExchange.ready(
                      function () {
                      StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3020438%2fways-for-two-dice-to-sum-to-8-x-1-x-2-8%23new-answer', 'question_page');
                      }
                      );

                      Post as a guest















                      Required, but never shown





















































                      Required, but never shown














                      Required, but never shown












                      Required, but never shown







                      Required, but never shown

































                      Required, but never shown














                      Required, but never shown












                      Required, but never shown







                      Required, but never shown







                      Popular posts from this blog

                      Quarter-circle Tiles

                      build a pushdown automaton that recognizes the reverse language of a given pushdown automaton?

                      Mont Emei