4

I have problems with the problem: What's the largest number?

\begin{align} &60^{\frac{1}{3}} \mbox{ or } 2 + 7^{\frac{1}{3}}& \end{align}

I tried using factorization, but I didn't get a good result and I dislike using an estimate.

Thanks.

Qiaochu Yuan
  • 419,620

3 Answers3

5

Compare the cubes of your numbers, which are $60$ and $$(2+7^{1/3})^3=8+3\cdot2^2\cdot 7^{1/3}+3\cdot 2\cdot 7^{2/3}+7=15+12\cdot 7^{1/3}+6\cdot 7^{2/3}.$$ Now $7^{1/3}<44/23$...

2

$$ (2+\root3\of7)^3<(2+\root3\of7)^3+(2-\root3\of7)^3=16+12\root3\of{49}. $$ Here $$ 16+12\root3\of{49}<60\Leftrightarrow\root3\of{49}<\frac{11}3. $$ The last inequality holds because both sides are positive and cubing gives a true inequality $$ 49\cdot27=1323<1331=11^3. $$

Jyrki Lahtonen
  • 133,153
  • Adding a small quantity to one side simplified it somewhat, but didn't affect the conclusion, so it worked. I was very fortunate in that sense (or the problem was designed with such a trick in mind). – Jyrki Lahtonen Mar 15 '13 at 21:17
  • The trick is justified / motivated because you know that $(2+\sqrt[3]{7})^3$ is not an integer, while 60 is. – Calvin Lin Mar 16 '13 at 00:24
  • @Calvin: Yeah, but I would have felt more certain about success (I'm sure you would have too), if I were dealing with an algebraic number of degree two. Adding a conjugate and all that. Here there were no small conjgates around, but simplification was still to be expected. – Jyrki Lahtonen Mar 16 '13 at 05:43
0

Cube both sides of the inequality.