2^(2^n) is congruent to 1 mod 3?
Maths and statistics discussion, revision, exam and homework help.
-
Re: 2^(2^n) is congruent to 1 mod 3?You cold just invoke that if(Original post by TenOfThem)
because

All the bits apart from the 1 have 3 to some power so are 0 [mod3]
then
(not that it makes any difference)
-
Re: 2^(2^n) is congruent to 1 mod 3?I don't follow... are you saying 2^(2^n)=4^n? This is clearly not true, 4^n=2^(2n) not 2^(2^n).(Original post by TenOfThem)
because

All the bits apart from the 1 have 3 to some power so are 0 [mod3] -
Re: 2^(2^n) is congruent to 1 mod 3?That's(Original post by TenOfThem)
because

All the bits apart from the 1 have 3 to some power so are 0 [mod3]
and not
. (Edit: Also that's not the binomial theorem!) (Edit 2: That said, it'd be right if you used the binomial theorem correctly, and it would prove the OP's result since
.)
@OP: You can prove it by induction, using the fact that
Last edited by nuodai; 31-03-2012 at 15:00. -
Re: 2^(2^n) is congruent to 1 mod 3?Are you saying(Original post by nuodai)
That's
and not
. (Edit: Also that's not the binomial theorem!) (Edit 2: That said, it'd be right if you used the binomial theorem correctly, and it would prove the OP's result since
.)
@OP: You can prove it by induction, using the fact that
?
edit: if so why?Last edited by deejayy; 31-03-2012 at 15:05. -
Re: 2^(2^n) is congruent to 1 mod 3?Yes.
For instance(Original post by deejayy)
edit: if so why?
.
Last edited by nuodai; 31-03-2012 at 15:07. -
Re: 2^(2^n) is congruent to 1 mod 3?I've just shown you why it's true. It's because(Original post by deejayy)
Why though? I think I dont understand indices entirely.
For instance
Why is this true?
, but it is not true that
; the latter means
.
Edit: You'll need to fix your understanding of indices before you do STEP in June!
Edit 2: To elaborate, you should try and lose the tendency to think "they look like they should be equal, so I need a reason why they're not" -- you should think "they look like they should be equal, so I'm going to prove that they are". Then when you try and fail to prove that they're equal, and instead find a counterexample, it's not such a shock. Think about what these things mean.
means that you multiply
by itself
times, whereas
means that you multiply
by itself
times. Clearly
, so why should
?
Last edited by nuodai; 31-03-2012 at 15:17. -
Re: 2^(2^n) is congruent to 1 mod 3?The method involving 4 does still work, namely because(Original post by nuodai)
I've just shown you why it's true. It's because
, but it is not true that
; the latter means
.
Edit: You'll need to fix your understanding of indices before you do STEP in June!
Edit 2: To elaborate, you should try and lose the tendency to think "they look like they should be equal, so I need a reason why they're not" -- you should think "they look like they should be equal, so I'm going to prove that they are". Then when you try and fail to prove that they're equal, and instead find a counterexample, it's not such a shock. Think about what these things mean.
means that you multiply
by itself
times, whereas
means that you multiply
by itself
times. Clearly
, so why should
?
just not for the reason suggested!
-
Re: 2^(2^n) is congruent to 1 mod 3?Indeed (I said as much in my first reply!).(Original post by TheMagicMan)
The method involving 4 does still work, namely because
just not for the reason suggested!
-
Re: 2^(2^n) is congruent to 1 mod 3?It's much more elementary than induction.(Original post by ben-smith)
tbf, Mod arithmetic had been mentioned earlier but I guess people wanted a more elementary solution. -
Re: 2^(2^n) is congruent to 1 mod 3?I can't say I agree though, personally, I would just binomially expand (3-1)^(2^n) and show that the only term without a 3 in is 1. This is essentially mod arithmetic but without the assumptions.(Original post by Dog4444)
It's much more elementary than induction. -
Re: 2^(2^n) is congruent to 1 mod 3?I would probably describe it as more elegant in comparison with the induction, as it isn't more elementary.(Original post by Dog4444)
It's much more elementary than induction.
(The result you used is indeed proved by induction.)
For example, I could prove Fermat's Little theorem in the following manner.
However, this does not make it near more elementary than the proof by induction and the binomial theorem given in the above link.
Spoiler:Show
Anyway, +rep for spotting it.
I agree with this, and for everybody planning to do STEP, I think binomial and induction must be our best friends.(Original post by ben-smith)
I can't say I agree though, personally, I would just binomially expand (3-1)^(2^n) and show that the only term without a 3 in is 1. This is essentially mod arithmetic but without the assumptions.