Page 1 of 1
11024 - Circular Lock
Posted: Sun Apr 02, 2006 5:23 am
by ferng1021
I got WA during the contest...
I think that a lock can be deactivated if and only if
(p11-s11)%p11 + (p22-s22)%p22 = (p12-s12)%p12 + (p21-s21)%p21
am I right?
If not, what's the correct way to solve this problem?
Thank you very much!!
Posted: Sun Apr 02, 2006 10:19 am
by andresw1
You are wrong. Think again:)
Posted: Sun Apr 02, 2006 11:16 am
by rushel
I am also having a WA
my logic:
if s11%p==s12%p==s21%p==s22%p then "Yes"
else "No"
Is my approach al right.
Posted: Sun Apr 02, 2006 12:51 pm
by ferng1021
AC!!
Thank you andresw1 and rushel!!
Your
%p
is a big hint for me!!
Thank you!!
Posted: Sun Apr 02, 2006 2:19 pm
by cypressx
rushel , you are underestimating the problem , it's not that easy.
Posted: Sun Apr 02, 2006 9:08 pm
by rushel
First I misunderstood the problem and now I tried a lot but i couldnt figure out any solution. Pls someone will be kind enough to give me a hint:
Posted: Sun Apr 02, 2006 9:20 pm
by mf
Make a system of linear equations modulo p, and solve it by hand.
Then you'll find a criterion, when a solution exists.
Posted: Mon Apr 03, 2006 12:31 am
by Jan
In the contest I did a silly mistake and got WA. But now I fixed my code and get it Accepted.
Suppose S11 is unchanged, and x press(es) to S12, S22 and y press(es) to S21,S22.
Then there would be a solution if
S11 % P = (S12 + x) % P = (S21 + y) % P = (S22 + x + y) % P
Now the rest is yours.
Hope it helps.
Posted: Mon Apr 03, 2006 12:41 am
by trulo17
it can be solved in O(p) . Change everthing to %p as suggested in description.Just try to figure out how many times you should press every one of the four buttons. During contest i got ac an O(p^2) solution.
Posted: Mon Apr 03, 2006 1:23 am
by mf
It can be solved in O(1)

A solution to the puzzle exists iff s12-s11 = s22-s21 (mod gcd(p11,p12,p21,p22))
Posted: Mon Apr 03, 2006 1:52 am
by Krzysztof Duleba
How do you do gcd(p11,p12,p21,p22) in O(1)?
Posted: Mon Apr 03, 2006 2:24 am
by mf
Ah, I was assuming an O(1) gcd operation.
Bit complexity would be of course higher.
Posted: Mon Apr 03, 2006 8:38 pm
by rushel
Thank u guys. I found the System of linear equation modulo p after lot of hard work. i dont know why it takes too much time to analyze. Any way
Thanks Again.
Really It was a good problem for a programmer like me.
Posted: Thu Apr 06, 2006 8:08 pm
by rajkon
Can somebody post a list of other math problems from uva.es? Thanx

Posted: Fri Apr 07, 2006 8:28 am
by Raiyan Kamal
I don't know how to derive the system of linear equation everyone is talking about, I have solved it using GCD and subtraction only. After subtracting couple of numers, i chek if the lock is opened, i.e. 0,0,0,0 configuration or not. Can anyone explain the relation between these two methods ?