converge or diverge sum_{2}^{infty} 1/(k ln k) use integral test
\[ \sum_{k=2}^{\infty} \frac{1}{(k ln k)} \] Now that looks better.
yes thanks
\[\sum_{k=2}^{\infty}{1\over k\ln k}\]now it looks even better :! ok let's do it! (looking through notes)
\[ \int_{2}^{\infty}\frac{1}{k \ln k} = ? \]
u-sub ??
idk
yes, u-sub should work
find out that, if that value is less than infinity then it should converge .... says wikipedia http://en.wikipedia.org/wiki/Convergence_test
if the integral converges so does the series if the integral diverges, so does the series that's my understanding
same thing as experimentX said I guess...
@experimentX I'm gonna call you exX from now on, I'm lazy :)
sure .. whatever
i set my u = x and i got stuck
no let u = ln x
well that ain't the right sub actually u=x is never the right sub
u=x is redundant you are just giving x a different name lol
hahaha ... that s funny
okay thats true
well give it a try out ... alert us when you get stuck
so i got \[\lim_{b \rightarrow \infty}\int\limits_{2}^{b} \ln(lnx)\]
am i on the right track???
yup, so does it diverge or converge?
wait a minute... I see something weird...
diverge
because it goes to infinity
do you see the same problem I do exX ? ln(2)<0 so ln(ln(2)) is undefined
yeah, the other part definitely diverges, but what about evaluation at 2 ?
ln(ln(2)) is just a constant i thought it shouldn't matter if i subtracted it from infinity
let's ask the wolf first http://www.wolframalpha.com/input/?i=integrate+from+2+to+infinity+1%2F%28x+lnx%29 wolf says does not converge.
I think it should be rather : ln(ln(inf)) - ln(ln(2)) ln 2 > 1 so lnln2 >1
oh I'm trippin' ^^^
yeah, so divergent
cool
ln2<1 not ln2<0 like I said befor
Oo ... looks like i didn't see that.
but still defined and constant, so it's all good :)
haha ... I forgot 2 < e
Join our real-time social learning platform and learn together with your friends!