I am trying to find a way to count the number of ways n different-sided dice can add up to a given number.
For example, 2 dice, 4- and 6-sided, can add up to 8 in 3 different ways: ($(2,6),(3,5),(4,4)$).
I've found a similar question but I cannot find a way to generalize it to my problem.
I can't provide a nice formula as in your referenced question, but I can show you at least half of the way.
Let's start with your
We encode the dies with polynomials and use the exponents to label the pips
If we denote with $[x^k]$ the coefficient $a_k$ of $x^k$ of a polynomial $\sum_{j=0}^{n}a_j x^j$, we observe
\begin{align*} [x^8]&(x^1+x^2+x^3+x^4)(x^1+x^2+x^3+x^4+x^5+x^6)\\ &=[x^8]x\frac{1-x^4}{1-x}x\frac{1-x^6}{1-x}\\ &=[x^6](1-x^4)(1-x^6)\frac{1}{(1-x)^2}\\ &=[x^6](1-x^4-x^6+x^{10})\sum_{j\geq 0}\binom{-2}{j}(-x)^j\\ &=[x^6](1-x^4-x^6+x^{10})\sum_{j\geq 0}\binom{j+1}{j}x^j\tag{1}\\ &=([x^6]-[x^2]-[x^0])\sum_{j\geq 0}(j+1)x^j\\ &=7-3-1\\ &=3 \end{align*}
So, the result is $3$ in accordance with your example. Now the same generalized:
We can now calculate similarly to the special case above:
\begin{align*} [x^K]&(x^1+\cdots+x^{k_1})^{j_1}\cdot...\cdot(x^1+\cdots+x^{k_l})^{j_l}\\ &=[x^K]x^{j_1}\left(\frac{1-x^{k_1}}{1-x}\right)^{j_1}\cdot...\cdot x^{j_l}\left(\frac{1-x^{k_l}}{1-x}\right)^{j_l}\\ &=[x^{K-j_1-...-j_l}](1-x^{k_1})^{j_1}\cdot...\cdot(1-x^{k_l})^{j_l}\frac{1}{(1-x)^N}\\ &=[x^{K-j_1-...-j_l}]\prod_{m=1}^{l}\left(1-x^{k_m}\right)^{j_m}\sum_{n\geq 0}\binom{-N}{n}(-x)^n\\ &=[x^{K-j_1-...-j_l}]\prod_{m=1}^{l}\left(1-x^{k_m}\right)^{j_m}\sum_{n\geq 0}\binom{N+n-1}{n}x^n\tag{2}\\ \end{align*}
Observe, that formula $(2)$ corresponds to formula $(1)$ in the special case above. So, for specific values we can continue at least in principle as we did from $(1)$.