channelflow.org

====== Differences ====== This shows you the differences between two versions of the page.

--- gibson:teaching:spring-2016:math445:hw1 [2016/01/25 11:53]
gibson created
+++ gibson:teaching:spring-2016:math445:hw1 [2016/01/25 11:55] (current)
gibson
@@ Line 20: / Line 20: @@
 \end{eqnarray*}
-Do you think finite truncations of this series will converge to $\pi$ faster or slower than the Ramanujan series from [[gibson:teaching:spring-2015:math445:lab1 | lab 1]]? Take a guess before testing in Matlab!
+Do you think finite truncations of this series will converge to $\pi$ faster or slower than the Ramanujan series from [[gibson:teaching:spring-2016:math445:lab1 | lab 1]]? Take a guess before testing in Matlab!
 How many digits of accuracy do you get for two terms of this series? How many digits did you get for two terms of the Ramanujan series? What reason is there for the difference in the convergence rate of the two formulae?
@@ Line 33: / Line 33: @@
 **Problem 4:**
-Let $e_n$ be the truncation of the above series after the $n$th term, i.e. $e_1 = 1$, $e_2 = 1 + 1/1!$, etc. Devise a Matlab expression that uses the ''log10'' function to count the number of digits of accuracy in any given $e_n$.
+Let $e_n$ be the truncation of the above series after the $n$th term, i.e. $e_1 = 1$, $e_2 = 1 + 1/1!$, etc. Devise a Matlab expression that uses the ''log10'' function to count the number of digits of accuracy for a given $e_n$. Use this expression to verify your answer for problem 3.