In this post in the last method the factorials were factorized. But I don't quite understand how that works.
Lets say we have
(−24)−1+(6)−1+(−2)−1
modulo a prime p, for instance 7. Then (−24)−1=2, (6)−1=6 and (−2)−1=3 (correct me if I'm wrong).
The sum is congruent to 11≡4 modulo 7 which is correct.
However, the factorized method multiplies (−24)−1 by 8 modulo 7. That is (−24)−1 (because 8≡1(mod7)) which equals 2.. that is wrong.
Am I doing something wrong here? Is 7 an exception because 8 is congruent to 1?
Answer
I think the problem is a mistake that was pointed out in the comments. Note that
−24(−24)−1≡1(modp)
6[−4(−24)−1]≡1(modp).
So we have 6−1≡−4(−24)−1. Similarly, we have (−2)−1≡12(−24)−1. Therefore, we have
(−24)−1+6−1+(−2)−1≡(−24)−1(1−4+12)≡9(24)−1
No comments:
Post a Comment