Programming C, C++, Java, PHP, Ruby, Turing, VB
Computer Science Canada 
Programming C, C++, Java, PHP, Ruby, Turing, VB  

Username:   Password: 
 RegisterRegister   
 Ccc2013
Index -> Contests
Goto page Previous  1, 2, 3, 4, 5, 6, 7, 8  Next
View previous topic Printable versionDownload TopicSubscribe to this topicPrivate MessagesRefresh page View next topic
Author Message
linuxp




PostPosted: Sun Mar 03, 2013 1:52 am   Post subject: Re: RE:Ccc2013

Raknarg @ Sat Mar 02, 2013 11:34 pm wrote:
Im pretty sure the questions had a 1 minute limit


The online grader changed time limit to 1 second for s1-4 and 6s for s5.
Evil or Very Mad
Sponsor
Sponsor
Sponsor
sponsor
nullptr




PostPosted: Sun Mar 03, 2013 5:47 pm   Post subject: Re: Ccc2013

My solution for S5 takes about 45 seconds on my own computer (which is about 2 GHz). Will it get full marks?
Halls McSmurfin




PostPosted: Sun Mar 03, 2013 6:11 pm   Post subject: Re: Ccc2013

nullptr @ Sun Mar 03, 2013 5:47 pm wrote:
My solution for S5 takes about 45 seconds on my own computer (which is about 2 GHz). Will it get full marks?


No
ishiney




PostPosted: Sun Mar 03, 2013 6:41 pm   Post subject: Re: Ccc2013

nullptr @ Sun Mar 03, 2013 5:47 pm wrote:
My solution for S5 takes about 45 seconds on my own computer (which is about 2 GHz). Will it get full marks?


Why not enter it in yourself? http://cemclinux1.math.uwaterloo.ca/index.php
The Post-contest page is for trying out your code again, I think.
nullptr




PostPosted: Sun Mar 03, 2013 6:59 pm   Post subject: Re: Ccc2013

ishiney @ Sun Mar 03, 2013 6:41 pm wrote:
nullptr @ Sun Mar 03, 2013 5:47 pm wrote:
My solution for S5 takes about 45 seconds on my own computer (which is about 2 GHz). Will it get full marks?


Why not enter it in yourself? http://cemclinux1.math.uwaterloo.ca/index.php
The Post-contest page is for trying out your code again, I think.


Our school didn't get a grader for some reason, so I don't have any login info.
t68




PostPosted: Sun Mar 03, 2013 7:17 pm   Post subject: Re: Ccc2013

When will CEMC release the stage 2 cut-off? Looks like a lot people solved S5. Could it be 65 for stage 2?
Unnamed




PostPosted: Sun Mar 03, 2013 7:38 pm   Post subject: Re: Ccc2013

How is CEMC going to get over this Computer Judge vs. Human Judge issue?

I know they answered with the whole, 6 s = 1 minute thing, but how about this?

The "slow" S5 solution on http://mmhs.ca/ccc/index.htm gets a 12/15 from a human judge, but a 9/15 from the computer judge.

This doesn't affect me, but someone with a 65 with a 9 for S5 may be prevented from moving on to stage 2 if the cut-off was 66 - 68.
t68




PostPosted: Sun Mar 03, 2013 7:44 pm   Post subject: Re: Ccc2013

On the other side, using online judge has the advantage to prevent careless mistakes. I guess it is hard to say which one was a better option for this year's CCC.
Sponsor
Sponsor
Sponsor
sponsor
coolgod




PostPosted: Sun Mar 03, 2013 7:44 pm   Post subject: Re: Ccc2013

From what I heard they will take most of the high scores and remark it once more on their judge. Ultimately the cutoff scores and final ccc scores should be based on the judge because this would be the fairest.
t68




PostPosted: Sun Mar 03, 2013 7:56 pm   Post subject: Re: Ccc2013

or they could give one cut-off for students without on-line judge since they don't have chance to correct their codes and re-submit, and give a different cut-off for people using on-line judge as they had shorter time limit

they even have the option to remove some test cases in S5 to balance out the differences, I guess. Rolling Eyes
coolgod




PostPosted: Sun Mar 03, 2013 8:46 pm   Post subject: Re: Ccc2013

what about people who used the
t68 wrote:

or they could give one cut-off for students without on-line judge since they don't have chance to correct their codes and re-submit, and give a different cut-off for people using on-line judge as they had shorter time limit

they even have the option to remove some test cases in S5 to balance out the differences, I guess. roll
what about people given the online judge without feedback? What about people who's code worked on judge but after remark it dropped? What about people who submitted both?
A.J




PostPosted: Mon Mar 04, 2013 1:41 am   Post subject: RE:Ccc2013

Student's with a mark of (typically) > 40 have their submissions remarked by the committee, and then the top 20 submissions are decided based on the scores (there might be 21 - 23 participants invited to move on to Stage 2 if there is a tie for 20th). I suspect that the cutoff will be higher than usual this year (~60 - 65).
t68




PostPosted: Mon Mar 04, 2013 10:21 am   Post subject: Re: Ccc2013

coolgod wrote:

From what I heard they will take most of the high scores and remark it once more on their judge. Ultimately the cutoff scores and final ccc scores should be based on the judge because this would be the fairest.


People (without online judge) did not even know this 6s limit. One would use a completely different algorithm for 6s vs. 1min. It is not reasonable to change the time limit rule for them at all (post contest). It is clearly about CEMC's reputation and ethics.

For people with online judge, I am not sure that they knew this 6s time limit up front. If yes, two groups of students were writing this contest with different instructions.

For a national contest like this, this is not a desired outcome. Maybe they could invite more than 20 students this year to make it fair. Very Happy
Halls McSmurfin




PostPosted: Mon Mar 04, 2013 4:48 pm   Post subject: RE:Ccc2013

Where does the 6s runtime limit come from? S5 has a 5s runtime limit.
t68




PostPosted: Mon Mar 04, 2013 5:27 pm   Post subject: Re: RE:Ccc2013

Halls McSmurfin @ Mon Mar 04, 2013 4:48 pm wrote:
Where does the 6s runtime limit come from? S5 has a 5s runtime limit.


well, close enough compared with 1min. Smile
Display posts from previous:   
   Index -> Contests
View previous topic Tell A FriendPrintable versionDownload TopicSubscribe to this topicPrivate MessagesRefresh page View next topic

Page 7 of 8  [ 118 Posts ]
Goto page Previous  1, 2, 3, 4, 5, 6, 7, 8  Next
Jump to:   


Style:  
Search: