Re: Lab
6 views
Skip to first unread message
Soroosh Yazdani
Dec 6, 2006, 12:32:20 AM12/6/06
to Rekha Thangellapalli, math128a...@groups.l.google.com
I made a post about this in the newsgroup. Yes, I think it is a good idea
to consider the values of h=1/2,1/4,...,1/1024. However, do not print the
values of the functions out!!! In fact, printing the values of the functions,
unless you are planning on showing me something really specific, is not a
useful thing. For each h, take the difference between y(1) and your last
approximation, as measuring the global error. Analyse how this global
error behaves for different h's, and different methods.
to consider the values of h=1/2,1/4,...,1/1024. However, do not print the
values of the functions out!!! In fact, printing the values of the functions,
unless you are planning on showing me something really specific, is not a
useful thing. For each h, take the difference between y(1) and your last
approximation, as measuring the global error. Analyse how this global
error behaves for different h's, and different methods.
Soroosh
On Tue, Dec 05, 2006 at 04:35:11PM -0800, Rekha Thangellapalli wrote:
> Dear Soroosh,
> Sorry to email you so late regarding the lab but hopefully you'll find time
> to email me back by tonight. I heard from a friend that for part 1, you
> wanted us to observe values starting from 1/2 to 1/1024 for h. As this seems
> a little tedious, I just wanted to confirm with you before actually printing
> out my lab.
>
> Thank you so much!
> Rekha Thangellapalli
Reply all
Reply to author
Forward
0 new messages