Submit | All submissions | Best solutions | Back to list |
PRIMES - Printing primes |
Wersja polska | English version |
Print all prime numbers from 0 to 1000000.
No input
Output
All primes from range [0;1000000].
No example
Special thanks to numerix for the concept of this task.
Added by: | Piotr Kąkol |
Date: | 2010-04-21 |
Time limit: | 35s |
Source limit: | 50000B |
Memory limit: | 1536MB |
Cluster: | Cube (Intel G860) |
Languages: | All except: NODEJS OBJC SCM qobi VB.NET |
hide comments
|
|||||
2021-07-28 02:48:53
.22 seconds in JAVA using sieve of eratosthenes and BitSet |
|||||
2014-02-14 23:03:18 Dominique VAILLANT
@Mitch 36 in Ruby?! Superb! I'm searching... |
|||||
2010-05-02 10:08:26 Piotr KÄ…kol
@Zoltán Zámbori - Too much a bit. ;-) @debanjan - Thanks for this useful information. :-) |
|||||
2010-04-27 19:33:47 :(){ :|: & };:
Piotr I think the judge doesn't work in the way you are thinking :) Take for example this task TDPRIMES and TDKPRIME,my AC solutions gives TLE if I try re-submitting in congestion that is when the judge is checking some other solutions. The submission id:3570850 and id:3570849,I resubmitted both of my AC solution simultaneously which now gives TLE. ;-) You may experiment this with any other two problems from your solution set :-) Last edit: 2010-04-27 20:04:53 |
|||||
2010-04-27 19:16:22 Zoltán Zámbori
I don't know, but i'm able to send some numbers. Printing primes with the regexp: 2.. 2500 needs 1sec 2.. 5000 needs 4.3sec 2.. 7500 needs 11sec 2..10000 needs 22sec 2..12500 needs 38sec 2..15000 needs 60sec 2..17500 needs 92sec 2..20000 needs 130sec Last edit: 2010-04-28 06:32:27 |
|||||
2010-04-27 16:12:34 Piotr KÄ…kol
@Debanjan - You won. ;-) Good for You! But if someone sent a slow submission SPOJ would check other submissions irrespective of previous slower one. You can check it by submitting to this program a TLE program with for example ( while(1); ) and a second later a correct sumbission -> the second will be checked first. ;-) @Zoltán Zámbori - How slow? ;-> (approximately of course) Last edit: 2010-04-27 16:13:46 |
|||||
2010-04-27 06:50:13 Zoltán Zámbori
I think the shortest possible solution in Perl is build on this regexp: (1x$i)!~/^(11+)\1+$/ It is extremely short and of course extremely slow. It is not a problem for me if the time limit close out this code. |
|||||
2010-04-27 06:05:06 :(){ :|: & };:
You may check the scores now :-) Also it is not necessary that O(n^2) algorithm need that much of run-time. If you have given me about 100 secs I could write much shorter code,but that will just waste of SPOJ resources,since any infinite loop or some other bug will unnecessarily use the judge and make other contestants wait ;-) Last edit: 2010-04-27 06:30:00 |
|||||
2010-04-26 22:34:06 Piotr KÄ…kol
challenger's is using (as You can see from his time) O(n2) algorithm. We'll see who will have shorter code. ;-) Elegant? In my opinion it doesn't matter in this kind of tasks. ;-) |
|||||
2010-04-26 22:28:23 :(){ :|: & };:
Don't know how others are solving this but for me Sieve is giving more shorten + elegant solution for this task. |