Anonymous user
Talk:Polynomial synthetic division: Difference between revisions
→Dup?
(→Dup?) |
(→Dup?) |
||
| (4 intermediate revisions by 2 users not shown) | |||
Line 16:
::: About the ordering of coefficients, I have added a note about that. Indeed, I reused the exact same example input from the Polynomial long division page, but reversing the order. I think this ordering is a better fit to easily implement Synthetic Division. The loops could be reversed inside the function but I guess this would just make things more confusing, but that's possible if you really want to follow the same convention. --[[User:Lrq3000|Lrq3000]] ([[User talk:Lrq3000|talk]]) 06:49, 14 June 2015 (UTC)
:::: Uh....
:::: So, ok, the other task includes some psuedo-code which will solve the task requirements. But I know that I entirely ignored that pseudo-code when I was implementing that task. And, if nothing else, my implementation was not recursive (except in the sense that we can think of numbers as representing recursive processes).
:::: And, in general, I try to ignore statements about the details of how things should be done and instead try to focus on what happens to the data, when implementing tasks here (or, anywhere). Any other approach doesn't really make sense to me. The differences between languages often enough are more significant than such algorithmic details. (For example, when you get to thinking about what optimizing compilers can do, or when you get into entirely different modes of notation.)
:::: So, if that's the only difference, I'd be inclined to say maybe just bring this entire page (along with the implementation) over to the other task page, and update the task description to be more clear. --[[User:Rdm|Rdm]] ([[User talk:Rdm|talk]]) 14:30, 14 June 2015 (UTC)
::::: This is NOT an implementation detail, it's a difference in the algorithm that drastically reduce the number of calculations required to find the solution. The [https://en.wikipedia.org/wiki/Computational_complexity_theory computational complexity] is different, so that asymptotically (ie, when using big polynomials), you will never reach the same performance as Synthetic Division if you use Long Division, no matter how you implement it or the language you use. Avoiding the recursive loop of Long Division is possible, I also did it, but it's a simple loop flattening trick, it doesn't change the computational cost (you will do the same number of calculations and iterations as if you did it recursively). This difference between the two algos is similar to the difference that separates Merge Sort and Quick Sort, and they both have a separate page: the algorithms are different and thus they incur a different computational cost. Of course, the goal is the same, just like those sorting algorithm, so maybe both Polynomial Long Division and Polynomial Synthetic Division should be put under the same subcategory Polynomial Division. --[[User:Lrq3000|Lrq3000]] ([[User talk:Lrq3000|talk]]) 17:28, 3 July 2015 (UTC)
:::::: That would be correct if the other page required a specific algorithm.
:::::: But instead, it merely suggests a [bad] algorithm. --[[User:Rdm|Rdm]] ([[User talk:Rdm|talk]]) 19:51, 3 July 2015 (UTC)
:::::: I don't understand, what do you mean by "the other page" and "bad algorithm"? --[[User:Lrq3000|Lrq3000]] ([[User talk:Lrq3000|talk]]) 19:44, 8 July 2015 (UTC)
| |||