$-3$ is a quadratic residue iff $p equiv 1 pmod 3$ [closed]

The name of the pictureThe name of the pictureThe name of the pictureClash Royale CLAN TAG#URR8PPP











up vote
1
down vote

favorite












So this is the question:
Let $p$ be an odd prime, prove that $-3$ is a quadratic residue modulo $p$ iff $p equiv 1 pmod 3$.



My idea was:
$$left(frac-3pright) = left(frac-1pright)cdot left(frac3pright)$$



also:
$$left(frac-3pright)= (-3)^a$$
when $a$ stands for $fracp-12$.



No idea how to continue... I would appreciate any help. Thanks!







share|cite|improve this question














closed as off-topic by anomaly, Xander Henderson, Taroccoesbrocco, Brahadeesh, José Carlos Santos Aug 22 at 14:39


This question appears to be off-topic. The users who voted to close gave this specific reason:


  • "This question is missing context or other details: Please improve the question by providing additional context, which ideally includes your thoughts on the problem and any attempts you have made to solve it. This information helps others identify where you have difficulties and helps them write answers appropriate to your experience level." – anomaly, Xander Henderson, Taroccoesbrocco, Brahadeesh, José Carlos Santos
If this question can be reworded to fit the rules in the help center, please edit the question.












  • That isn't an attempt to solve the problem; you just wrote out one of the definitions of $(cdot/p)$. Use quadratic reciprocity.
    – anomaly
    Aug 21 at 16:05














up vote
1
down vote

favorite












So this is the question:
Let $p$ be an odd prime, prove that $-3$ is a quadratic residue modulo $p$ iff $p equiv 1 pmod 3$.



My idea was:
$$left(frac-3pright) = left(frac-1pright)cdot left(frac3pright)$$



also:
$$left(frac-3pright)= (-3)^a$$
when $a$ stands for $fracp-12$.



No idea how to continue... I would appreciate any help. Thanks!







share|cite|improve this question














closed as off-topic by anomaly, Xander Henderson, Taroccoesbrocco, Brahadeesh, José Carlos Santos Aug 22 at 14:39


This question appears to be off-topic. The users who voted to close gave this specific reason:


  • "This question is missing context or other details: Please improve the question by providing additional context, which ideally includes your thoughts on the problem and any attempts you have made to solve it. This information helps others identify where you have difficulties and helps them write answers appropriate to your experience level." – anomaly, Xander Henderson, Taroccoesbrocco, Brahadeesh, José Carlos Santos
If this question can be reworded to fit the rules in the help center, please edit the question.












  • That isn't an attempt to solve the problem; you just wrote out one of the definitions of $(cdot/p)$. Use quadratic reciprocity.
    – anomaly
    Aug 21 at 16:05












up vote
1
down vote

favorite









up vote
1
down vote

favorite











So this is the question:
Let $p$ be an odd prime, prove that $-3$ is a quadratic residue modulo $p$ iff $p equiv 1 pmod 3$.



My idea was:
$$left(frac-3pright) = left(frac-1pright)cdot left(frac3pright)$$



also:
$$left(frac-3pright)= (-3)^a$$
when $a$ stands for $fracp-12$.



No idea how to continue... I would appreciate any help. Thanks!







share|cite|improve this question














So this is the question:
Let $p$ be an odd prime, prove that $-3$ is a quadratic residue modulo $p$ iff $p equiv 1 pmod 3$.



My idea was:
$$left(frac-3pright) = left(frac-1pright)cdot left(frac3pright)$$



also:
$$left(frac-3pright)= (-3)^a$$
when $a$ stands for $fracp-12$.



No idea how to continue... I would appreciate any help. Thanks!









share|cite|improve this question













share|cite|improve this question




share|cite|improve this question








edited Aug 21 at 16:03









cansomeonehelpmeout

5,1783830




5,1783830










asked Aug 21 at 13:36









LonelyStudent

122




122




closed as off-topic by anomaly, Xander Henderson, Taroccoesbrocco, Brahadeesh, José Carlos Santos Aug 22 at 14:39


This question appears to be off-topic. The users who voted to close gave this specific reason:


  • "This question is missing context or other details: Please improve the question by providing additional context, which ideally includes your thoughts on the problem and any attempts you have made to solve it. This information helps others identify where you have difficulties and helps them write answers appropriate to your experience level." – anomaly, Xander Henderson, Taroccoesbrocco, Brahadeesh, José Carlos Santos
If this question can be reworded to fit the rules in the help center, please edit the question.




closed as off-topic by anomaly, Xander Henderson, Taroccoesbrocco, Brahadeesh, José Carlos Santos Aug 22 at 14:39


This question appears to be off-topic. The users who voted to close gave this specific reason:


  • "This question is missing context or other details: Please improve the question by providing additional context, which ideally includes your thoughts on the problem and any attempts you have made to solve it. This information helps others identify where you have difficulties and helps them write answers appropriate to your experience level." – anomaly, Xander Henderson, Taroccoesbrocco, Brahadeesh, José Carlos Santos
If this question can be reworded to fit the rules in the help center, please edit the question.











  • That isn't an attempt to solve the problem; you just wrote out one of the definitions of $(cdot/p)$. Use quadratic reciprocity.
    – anomaly
    Aug 21 at 16:05
















  • That isn't an attempt to solve the problem; you just wrote out one of the definitions of $(cdot/p)$. Use quadratic reciprocity.
    – anomaly
    Aug 21 at 16:05















That isn't an attempt to solve the problem; you just wrote out one of the definitions of $(cdot/p)$. Use quadratic reciprocity.
– anomaly
Aug 21 at 16:05




That isn't an attempt to solve the problem; you just wrote out one of the definitions of $(cdot/p)$. Use quadratic reciprocity.
– anomaly
Aug 21 at 16:05










2 Answers
2






active

oldest

votes

















up vote
6
down vote













Your idea is right! But note that




$$left (frac-3pright)=1iff pequiv_61$$




is a stronger and better condition than $pequiv_3 1$, why? Look at all the numbers $nequiv_3 1$ (that is, $n=1+3k$): $$1,colorred4,7,colorred10,13,colorred16,19,colorred22,dots$$ Cool, now what? Notice that you alternate between even and odd numbers, so you can just discard the even numbers. This happens because adding an odd number of $3$'s to $1$ gives an even number. But no even number on the form $1+3k$ can be prime! Therefore you need to add an even number of $3$'s, that is, adding $6$. Therefore you should look at $pequiv_6 1$ instead: $$1,7,13,19,dots$$ Now, back to business!



You can write $left (frac-3pright)=left (frac-1pright)left (frac3pright)$, now we know that $left (frac-1pright)=(-1)^fracp-12$, so we only have to worry about what $left (frac3pright)$ is.



For this, you can use a theorem known as quadratic reciprocity:




For $p,q$ distinct odd primes$$left (fracpqright)left (fracqpright)=(-1)^fracq-12fracp-12$$




Then you have to check two cases:



  • $left (frac-1pright)=1$ and $left (frac3pright)=1$

  • $left (frac-1pright)=-1$ and $left (frac3pright)=-1$

Can you take it from here?




(Edit)



I'll help you with the first case.



If we want $left (frac-1pright )=1$, then $(-1)^fracp-12=1$, which means that $fracp-12$ must be even. This is the same as saying: $$fracp-12=2kiff p=1+4kiff pequiv_4 1$$ From quadratic reciprocity we know that: $$left( frac3pright) left( fracp3right )=(-1)^fracp-12=1$$ Thus, $$left( frac3pright)=left( fracp3right)$$ (we can multiply and divide by $left( frac3pright)$, since $left( frac3pright)=pm 1$)



Now we need to find out what $left( fracp3right)$ is. Since we want $left( frac3pright)=1$, we need to have $left( fracp3right)=1$, that is, we need to find the primes $p$ such that $p$ is a quadratic residue modulo $3$. Let us look at the residues: $$beginalign1^2=1\(-1)^2=1endalign$$ So the only quadratic residue modulo $3$ is $1$, therefore we want $p=1+3niff pequiv_3 1$



We now have two conditions on $p$:



  • $pequiv_3 1$

  • $pequiv_4 1$

This means that $pequiv_121$. Now you only have to check the second case! Combining both cases you should get: $$pequiv_6 1$$ which is stronger than $pequiv_3 1$.






share|cite|improve this answer


















  • 1




    First of all thanks for the help, but if p is odd, cant i just ignore the second case? because (p-1)/2 would be even, so $(frac-3p)$= $(frac3p)$=1 or am i wrong? and after i finish proving it, how can i conclude that p=1 (mod 3)? im sorry if my question sound stupid, but i just didnt understand that subject at all..
    – LonelyStudent
    Aug 21 at 13:58











  • @LonelyStudent Not a stupid question, number theory is hard, and we have all struggled with it some time! $p$ is a prime, so $p$ is always odd unless $p=2$. But $fracp-12$ is not always even, for instance, $frac19-12=9$. I will expand my answer and help you some more.
    – cansomeonehelpmeout
    Aug 21 at 14:02


















up vote
0
down vote













The idea to use $left(dfrac-1pright)$ and $left(dfrac3pright)$ to calculate $left(dfrac-3pright)$ is a natural one...



But, IMHO it works BETTER if you go the opposite direction, and calculate $left(dfrac3pright)$ using the other two!!



This is because $-3$ happens to be the discriminant of the polynomial
$$
Phi_3(x)=x^2+x+1.
$$
And, coincidentally, the presence of zeros of $Phi_3$ modulo a prime $p$ can be decided by other means. Namely, we have the factorization $Phi_3(x)(x-1)=x^3-1$.
Implying that modulo $p>3$ the integer $a$ is a zero of $Phi_3$ if and only if $anotequiv1pmod p$ and $a^3equiv1pmod p$. In other words, $a$ is a zero if and only if it has order three in the group $BbbZ_p^*$. This group is cyclic of order $p-1$ (recall the existence of primitive roots modulo $p$), so, by Lagrange, such numbers $a$ exist if and only if $3mid(p-1)$.



OTOH, by quadratic formula, $Phi_3(x)$ has zeros modulo $p$ if and only if the discriminant $-3$ is a quadratic residue.




It follows that $left(dfrac-3pright)=1$ if and only if $pequiv1pmod3$.







share|cite|improve this answer






















  • Marking this CW because I am positive that this is a dupe. I need to commute next, so can search for one only later.
    – Jyrki Lahtonen
    Aug 22 at 6:34

















2 Answers
2






active

oldest

votes








2 Answers
2






active

oldest

votes









active

oldest

votes






active

oldest

votes








up vote
6
down vote













Your idea is right! But note that




$$left (frac-3pright)=1iff pequiv_61$$




is a stronger and better condition than $pequiv_3 1$, why? Look at all the numbers $nequiv_3 1$ (that is, $n=1+3k$): $$1,colorred4,7,colorred10,13,colorred16,19,colorred22,dots$$ Cool, now what? Notice that you alternate between even and odd numbers, so you can just discard the even numbers. This happens because adding an odd number of $3$'s to $1$ gives an even number. But no even number on the form $1+3k$ can be prime! Therefore you need to add an even number of $3$'s, that is, adding $6$. Therefore you should look at $pequiv_6 1$ instead: $$1,7,13,19,dots$$ Now, back to business!



You can write $left (frac-3pright)=left (frac-1pright)left (frac3pright)$, now we know that $left (frac-1pright)=(-1)^fracp-12$, so we only have to worry about what $left (frac3pright)$ is.



For this, you can use a theorem known as quadratic reciprocity:




For $p,q$ distinct odd primes$$left (fracpqright)left (fracqpright)=(-1)^fracq-12fracp-12$$




Then you have to check two cases:



  • $left (frac-1pright)=1$ and $left (frac3pright)=1$

  • $left (frac-1pright)=-1$ and $left (frac3pright)=-1$

Can you take it from here?




(Edit)



I'll help you with the first case.



If we want $left (frac-1pright )=1$, then $(-1)^fracp-12=1$, which means that $fracp-12$ must be even. This is the same as saying: $$fracp-12=2kiff p=1+4kiff pequiv_4 1$$ From quadratic reciprocity we know that: $$left( frac3pright) left( fracp3right )=(-1)^fracp-12=1$$ Thus, $$left( frac3pright)=left( fracp3right)$$ (we can multiply and divide by $left( frac3pright)$, since $left( frac3pright)=pm 1$)



Now we need to find out what $left( fracp3right)$ is. Since we want $left( frac3pright)=1$, we need to have $left( fracp3right)=1$, that is, we need to find the primes $p$ such that $p$ is a quadratic residue modulo $3$. Let us look at the residues: $$beginalign1^2=1\(-1)^2=1endalign$$ So the only quadratic residue modulo $3$ is $1$, therefore we want $p=1+3niff pequiv_3 1$



We now have two conditions on $p$:



  • $pequiv_3 1$

  • $pequiv_4 1$

This means that $pequiv_121$. Now you only have to check the second case! Combining both cases you should get: $$pequiv_6 1$$ which is stronger than $pequiv_3 1$.






share|cite|improve this answer


















  • 1




    First of all thanks for the help, but if p is odd, cant i just ignore the second case? because (p-1)/2 would be even, so $(frac-3p)$= $(frac3p)$=1 or am i wrong? and after i finish proving it, how can i conclude that p=1 (mod 3)? im sorry if my question sound stupid, but i just didnt understand that subject at all..
    – LonelyStudent
    Aug 21 at 13:58











  • @LonelyStudent Not a stupid question, number theory is hard, and we have all struggled with it some time! $p$ is a prime, so $p$ is always odd unless $p=2$. But $fracp-12$ is not always even, for instance, $frac19-12=9$. I will expand my answer and help you some more.
    – cansomeonehelpmeout
    Aug 21 at 14:02















up vote
6
down vote













Your idea is right! But note that




$$left (frac-3pright)=1iff pequiv_61$$




is a stronger and better condition than $pequiv_3 1$, why? Look at all the numbers $nequiv_3 1$ (that is, $n=1+3k$): $$1,colorred4,7,colorred10,13,colorred16,19,colorred22,dots$$ Cool, now what? Notice that you alternate between even and odd numbers, so you can just discard the even numbers. This happens because adding an odd number of $3$'s to $1$ gives an even number. But no even number on the form $1+3k$ can be prime! Therefore you need to add an even number of $3$'s, that is, adding $6$. Therefore you should look at $pequiv_6 1$ instead: $$1,7,13,19,dots$$ Now, back to business!



You can write $left (frac-3pright)=left (frac-1pright)left (frac3pright)$, now we know that $left (frac-1pright)=(-1)^fracp-12$, so we only have to worry about what $left (frac3pright)$ is.



For this, you can use a theorem known as quadratic reciprocity:




For $p,q$ distinct odd primes$$left (fracpqright)left (fracqpright)=(-1)^fracq-12fracp-12$$




Then you have to check two cases:



  • $left (frac-1pright)=1$ and $left (frac3pright)=1$

  • $left (frac-1pright)=-1$ and $left (frac3pright)=-1$

Can you take it from here?




(Edit)



I'll help you with the first case.



If we want $left (frac-1pright )=1$, then $(-1)^fracp-12=1$, which means that $fracp-12$ must be even. This is the same as saying: $$fracp-12=2kiff p=1+4kiff pequiv_4 1$$ From quadratic reciprocity we know that: $$left( frac3pright) left( fracp3right )=(-1)^fracp-12=1$$ Thus, $$left( frac3pright)=left( fracp3right)$$ (we can multiply and divide by $left( frac3pright)$, since $left( frac3pright)=pm 1$)



Now we need to find out what $left( fracp3right)$ is. Since we want $left( frac3pright)=1$, we need to have $left( fracp3right)=1$, that is, we need to find the primes $p$ such that $p$ is a quadratic residue modulo $3$. Let us look at the residues: $$beginalign1^2=1\(-1)^2=1endalign$$ So the only quadratic residue modulo $3$ is $1$, therefore we want $p=1+3niff pequiv_3 1$



We now have two conditions on $p$:



  • $pequiv_3 1$

  • $pequiv_4 1$

This means that $pequiv_121$. Now you only have to check the second case! Combining both cases you should get: $$pequiv_6 1$$ which is stronger than $pequiv_3 1$.






share|cite|improve this answer


















  • 1




    First of all thanks for the help, but if p is odd, cant i just ignore the second case? because (p-1)/2 would be even, so $(frac-3p)$= $(frac3p)$=1 or am i wrong? and after i finish proving it, how can i conclude that p=1 (mod 3)? im sorry if my question sound stupid, but i just didnt understand that subject at all..
    – LonelyStudent
    Aug 21 at 13:58











  • @LonelyStudent Not a stupid question, number theory is hard, and we have all struggled with it some time! $p$ is a prime, so $p$ is always odd unless $p=2$. But $fracp-12$ is not always even, for instance, $frac19-12=9$. I will expand my answer and help you some more.
    – cansomeonehelpmeout
    Aug 21 at 14:02













up vote
6
down vote










up vote
6
down vote









Your idea is right! But note that




$$left (frac-3pright)=1iff pequiv_61$$




is a stronger and better condition than $pequiv_3 1$, why? Look at all the numbers $nequiv_3 1$ (that is, $n=1+3k$): $$1,colorred4,7,colorred10,13,colorred16,19,colorred22,dots$$ Cool, now what? Notice that you alternate between even and odd numbers, so you can just discard the even numbers. This happens because adding an odd number of $3$'s to $1$ gives an even number. But no even number on the form $1+3k$ can be prime! Therefore you need to add an even number of $3$'s, that is, adding $6$. Therefore you should look at $pequiv_6 1$ instead: $$1,7,13,19,dots$$ Now, back to business!



You can write $left (frac-3pright)=left (frac-1pright)left (frac3pright)$, now we know that $left (frac-1pright)=(-1)^fracp-12$, so we only have to worry about what $left (frac3pright)$ is.



For this, you can use a theorem known as quadratic reciprocity:




For $p,q$ distinct odd primes$$left (fracpqright)left (fracqpright)=(-1)^fracq-12fracp-12$$




Then you have to check two cases:



  • $left (frac-1pright)=1$ and $left (frac3pright)=1$

  • $left (frac-1pright)=-1$ and $left (frac3pright)=-1$

Can you take it from here?




(Edit)



I'll help you with the first case.



If we want $left (frac-1pright )=1$, then $(-1)^fracp-12=1$, which means that $fracp-12$ must be even. This is the same as saying: $$fracp-12=2kiff p=1+4kiff pequiv_4 1$$ From quadratic reciprocity we know that: $$left( frac3pright) left( fracp3right )=(-1)^fracp-12=1$$ Thus, $$left( frac3pright)=left( fracp3right)$$ (we can multiply and divide by $left( frac3pright)$, since $left( frac3pright)=pm 1$)



Now we need to find out what $left( fracp3right)$ is. Since we want $left( frac3pright)=1$, we need to have $left( fracp3right)=1$, that is, we need to find the primes $p$ such that $p$ is a quadratic residue modulo $3$. Let us look at the residues: $$beginalign1^2=1\(-1)^2=1endalign$$ So the only quadratic residue modulo $3$ is $1$, therefore we want $p=1+3niff pequiv_3 1$



We now have two conditions on $p$:



  • $pequiv_3 1$

  • $pequiv_4 1$

This means that $pequiv_121$. Now you only have to check the second case! Combining both cases you should get: $$pequiv_6 1$$ which is stronger than $pequiv_3 1$.






share|cite|improve this answer














Your idea is right! But note that




$$left (frac-3pright)=1iff pequiv_61$$




is a stronger and better condition than $pequiv_3 1$, why? Look at all the numbers $nequiv_3 1$ (that is, $n=1+3k$): $$1,colorred4,7,colorred10,13,colorred16,19,colorred22,dots$$ Cool, now what? Notice that you alternate between even and odd numbers, so you can just discard the even numbers. This happens because adding an odd number of $3$'s to $1$ gives an even number. But no even number on the form $1+3k$ can be prime! Therefore you need to add an even number of $3$'s, that is, adding $6$. Therefore you should look at $pequiv_6 1$ instead: $$1,7,13,19,dots$$ Now, back to business!



You can write $left (frac-3pright)=left (frac-1pright)left (frac3pright)$, now we know that $left (frac-1pright)=(-1)^fracp-12$, so we only have to worry about what $left (frac3pright)$ is.



For this, you can use a theorem known as quadratic reciprocity:




For $p,q$ distinct odd primes$$left (fracpqright)left (fracqpright)=(-1)^fracq-12fracp-12$$




Then you have to check two cases:



  • $left (frac-1pright)=1$ and $left (frac3pright)=1$

  • $left (frac-1pright)=-1$ and $left (frac3pright)=-1$

Can you take it from here?




(Edit)



I'll help you with the first case.



If we want $left (frac-1pright )=1$, then $(-1)^fracp-12=1$, which means that $fracp-12$ must be even. This is the same as saying: $$fracp-12=2kiff p=1+4kiff pequiv_4 1$$ From quadratic reciprocity we know that: $$left( frac3pright) left( fracp3right )=(-1)^fracp-12=1$$ Thus, $$left( frac3pright)=left( fracp3right)$$ (we can multiply and divide by $left( frac3pright)$, since $left( frac3pright)=pm 1$)



Now we need to find out what $left( fracp3right)$ is. Since we want $left( frac3pright)=1$, we need to have $left( fracp3right)=1$, that is, we need to find the primes $p$ such that $p$ is a quadratic residue modulo $3$. Let us look at the residues: $$beginalign1^2=1\(-1)^2=1endalign$$ So the only quadratic residue modulo $3$ is $1$, therefore we want $p=1+3niff pequiv_3 1$



We now have two conditions on $p$:



  • $pequiv_3 1$

  • $pequiv_4 1$

This means that $pequiv_121$. Now you only have to check the second case! Combining both cases you should get: $$pequiv_6 1$$ which is stronger than $pequiv_3 1$.







share|cite|improve this answer














share|cite|improve this answer



share|cite|improve this answer








edited Aug 21 at 14:47

























answered Aug 21 at 13:46









cansomeonehelpmeout

5,1783830




5,1783830







  • 1




    First of all thanks for the help, but if p is odd, cant i just ignore the second case? because (p-1)/2 would be even, so $(frac-3p)$= $(frac3p)$=1 or am i wrong? and after i finish proving it, how can i conclude that p=1 (mod 3)? im sorry if my question sound stupid, but i just didnt understand that subject at all..
    – LonelyStudent
    Aug 21 at 13:58











  • @LonelyStudent Not a stupid question, number theory is hard, and we have all struggled with it some time! $p$ is a prime, so $p$ is always odd unless $p=2$. But $fracp-12$ is not always even, for instance, $frac19-12=9$. I will expand my answer and help you some more.
    – cansomeonehelpmeout
    Aug 21 at 14:02













  • 1




    First of all thanks for the help, but if p is odd, cant i just ignore the second case? because (p-1)/2 would be even, so $(frac-3p)$= $(frac3p)$=1 or am i wrong? and after i finish proving it, how can i conclude that p=1 (mod 3)? im sorry if my question sound stupid, but i just didnt understand that subject at all..
    – LonelyStudent
    Aug 21 at 13:58











  • @LonelyStudent Not a stupid question, number theory is hard, and we have all struggled with it some time! $p$ is a prime, so $p$ is always odd unless $p=2$. But $fracp-12$ is not always even, for instance, $frac19-12=9$. I will expand my answer and help you some more.
    – cansomeonehelpmeout
    Aug 21 at 14:02








1




1




First of all thanks for the help, but if p is odd, cant i just ignore the second case? because (p-1)/2 would be even, so $(frac-3p)$= $(frac3p)$=1 or am i wrong? and after i finish proving it, how can i conclude that p=1 (mod 3)? im sorry if my question sound stupid, but i just didnt understand that subject at all..
– LonelyStudent
Aug 21 at 13:58





First of all thanks for the help, but if p is odd, cant i just ignore the second case? because (p-1)/2 would be even, so $(frac-3p)$= $(frac3p)$=1 or am i wrong? and after i finish proving it, how can i conclude that p=1 (mod 3)? im sorry if my question sound stupid, but i just didnt understand that subject at all..
– LonelyStudent
Aug 21 at 13:58













@LonelyStudent Not a stupid question, number theory is hard, and we have all struggled with it some time! $p$ is a prime, so $p$ is always odd unless $p=2$. But $fracp-12$ is not always even, for instance, $frac19-12=9$. I will expand my answer and help you some more.
– cansomeonehelpmeout
Aug 21 at 14:02





@LonelyStudent Not a stupid question, number theory is hard, and we have all struggled with it some time! $p$ is a prime, so $p$ is always odd unless $p=2$. But $fracp-12$ is not always even, for instance, $frac19-12=9$. I will expand my answer and help you some more.
– cansomeonehelpmeout
Aug 21 at 14:02











up vote
0
down vote













The idea to use $left(dfrac-1pright)$ and $left(dfrac3pright)$ to calculate $left(dfrac-3pright)$ is a natural one...



But, IMHO it works BETTER if you go the opposite direction, and calculate $left(dfrac3pright)$ using the other two!!



This is because $-3$ happens to be the discriminant of the polynomial
$$
Phi_3(x)=x^2+x+1.
$$
And, coincidentally, the presence of zeros of $Phi_3$ modulo a prime $p$ can be decided by other means. Namely, we have the factorization $Phi_3(x)(x-1)=x^3-1$.
Implying that modulo $p>3$ the integer $a$ is a zero of $Phi_3$ if and only if $anotequiv1pmod p$ and $a^3equiv1pmod p$. In other words, $a$ is a zero if and only if it has order three in the group $BbbZ_p^*$. This group is cyclic of order $p-1$ (recall the existence of primitive roots modulo $p$), so, by Lagrange, such numbers $a$ exist if and only if $3mid(p-1)$.



OTOH, by quadratic formula, $Phi_3(x)$ has zeros modulo $p$ if and only if the discriminant $-3$ is a quadratic residue.




It follows that $left(dfrac-3pright)=1$ if and only if $pequiv1pmod3$.







share|cite|improve this answer






















  • Marking this CW because I am positive that this is a dupe. I need to commute next, so can search for one only later.
    – Jyrki Lahtonen
    Aug 22 at 6:34














up vote
0
down vote













The idea to use $left(dfrac-1pright)$ and $left(dfrac3pright)$ to calculate $left(dfrac-3pright)$ is a natural one...



But, IMHO it works BETTER if you go the opposite direction, and calculate $left(dfrac3pright)$ using the other two!!



This is because $-3$ happens to be the discriminant of the polynomial
$$
Phi_3(x)=x^2+x+1.
$$
And, coincidentally, the presence of zeros of $Phi_3$ modulo a prime $p$ can be decided by other means. Namely, we have the factorization $Phi_3(x)(x-1)=x^3-1$.
Implying that modulo $p>3$ the integer $a$ is a zero of $Phi_3$ if and only if $anotequiv1pmod p$ and $a^3equiv1pmod p$. In other words, $a$ is a zero if and only if it has order three in the group $BbbZ_p^*$. This group is cyclic of order $p-1$ (recall the existence of primitive roots modulo $p$), so, by Lagrange, such numbers $a$ exist if and only if $3mid(p-1)$.



OTOH, by quadratic formula, $Phi_3(x)$ has zeros modulo $p$ if and only if the discriminant $-3$ is a quadratic residue.




It follows that $left(dfrac-3pright)=1$ if and only if $pequiv1pmod3$.







share|cite|improve this answer






















  • Marking this CW because I am positive that this is a dupe. I need to commute next, so can search for one only later.
    – Jyrki Lahtonen
    Aug 22 at 6:34












up vote
0
down vote










up vote
0
down vote









The idea to use $left(dfrac-1pright)$ and $left(dfrac3pright)$ to calculate $left(dfrac-3pright)$ is a natural one...



But, IMHO it works BETTER if you go the opposite direction, and calculate $left(dfrac3pright)$ using the other two!!



This is because $-3$ happens to be the discriminant of the polynomial
$$
Phi_3(x)=x^2+x+1.
$$
And, coincidentally, the presence of zeros of $Phi_3$ modulo a prime $p$ can be decided by other means. Namely, we have the factorization $Phi_3(x)(x-1)=x^3-1$.
Implying that modulo $p>3$ the integer $a$ is a zero of $Phi_3$ if and only if $anotequiv1pmod p$ and $a^3equiv1pmod p$. In other words, $a$ is a zero if and only if it has order three in the group $BbbZ_p^*$. This group is cyclic of order $p-1$ (recall the existence of primitive roots modulo $p$), so, by Lagrange, such numbers $a$ exist if and only if $3mid(p-1)$.



OTOH, by quadratic formula, $Phi_3(x)$ has zeros modulo $p$ if and only if the discriminant $-3$ is a quadratic residue.




It follows that $left(dfrac-3pright)=1$ if and only if $pequiv1pmod3$.







share|cite|improve this answer














The idea to use $left(dfrac-1pright)$ and $left(dfrac3pright)$ to calculate $left(dfrac-3pright)$ is a natural one...



But, IMHO it works BETTER if you go the opposite direction, and calculate $left(dfrac3pright)$ using the other two!!



This is because $-3$ happens to be the discriminant of the polynomial
$$
Phi_3(x)=x^2+x+1.
$$
And, coincidentally, the presence of zeros of $Phi_3$ modulo a prime $p$ can be decided by other means. Namely, we have the factorization $Phi_3(x)(x-1)=x^3-1$.
Implying that modulo $p>3$ the integer $a$ is a zero of $Phi_3$ if and only if $anotequiv1pmod p$ and $a^3equiv1pmod p$. In other words, $a$ is a zero if and only if it has order three in the group $BbbZ_p^*$. This group is cyclic of order $p-1$ (recall the existence of primitive roots modulo $p$), so, by Lagrange, such numbers $a$ exist if and only if $3mid(p-1)$.



OTOH, by quadratic formula, $Phi_3(x)$ has zeros modulo $p$ if and only if the discriminant $-3$ is a quadratic residue.




It follows that $left(dfrac-3pright)=1$ if and only if $pequiv1pmod3$.








share|cite|improve this answer














share|cite|improve this answer



share|cite|improve this answer








answered Aug 22 at 6:33


























community wiki





Jyrki Lahtonen












  • Marking this CW because I am positive that this is a dupe. I need to commute next, so can search for one only later.
    – Jyrki Lahtonen
    Aug 22 at 6:34
















  • Marking this CW because I am positive that this is a dupe. I need to commute next, so can search for one only later.
    – Jyrki Lahtonen
    Aug 22 at 6:34















Marking this CW because I am positive that this is a dupe. I need to commute next, so can search for one only later.
– Jyrki Lahtonen
Aug 22 at 6:34




Marking this CW because I am positive that this is a dupe. I need to commute next, so can search for one only later.
– Jyrki Lahtonen
Aug 22 at 6:34


這個網誌中的熱門文章

How to combine Bézier curves to a surface?

Mutual Information Always Non-negative

Why am i infinitely getting the same tweet with the Twitter Search API?