Using double precision, find x for which exp(x) doesn't overflow.

5 views (last 30 days)
Using double precision, find x for which exp(x) doesn't overflow.
  4 Comments
Shiksha Pandey
Shiksha Pandey on 24 Mar 2021
Edited: Rik on 24 Mar 2021
This is what I have. I'm not sure if this is the best way to go though.
expn=0;
x=1:1000;
while expn < realmax
expn=(exp(x));
maxexpn = double(expn);
xx = double(log(maxexpn));
mexp = max(maxexpn(~isinf(maxexpn)));
mx = max(xx(~isinf(xx)))
abserr = abs(mexp-realmax)
relerr = (abs(mexp-realmax))/realmax
end
Is there a shorter way to go about it?
Rik
Rik on 24 Mar 2021
I don't really understand what you're attempting to do with the while loop, but yes, there is a faster way. It requires you to think a bit more.
log2(realmax)
ans = 1024
OK, so we are looking for x where e^x is at most 2^1024.
With a little bit of algebra that should not be too hard to solve. Can you see how?

Sign in to comment.

Answers (1)

KALYAN ACHARJYA
KALYAN ACHARJYA on 24 Mar 2021
Edited: KALYAN ACHARJYA on 24 Mar 2021
All data creates in MATLAB are douple precision (Standard /By Default)
More:
e^x = 1 + (x ^ 1 / 1!) + (x ^ 2 / 2!) ......

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!