HELP WITH NCPC

Post here if you don't find any other place for your post. But please, stay on-topic: algorithms, programming or something related to this web site and its services.

Moderator: Board moderators

Dreamer#1
Learning poster
Posts: 77
Joined: Tue Oct 07, 2003 10:07 pm

Post by Dreamer#1 »

I personally think the problemset of NCPC was pretty unfair because the 1st 6 problems were far more easier than the next 4 problems. You don't need to be a much good programmer to solve the 1st 6 problems and you need to be a very good programmer to solve the final 4 problems. First of all the problems shouldn't have been sorted according to difficulty because one of the main objective of a programming contest is to identify the easier problems because no matter whether you solve a relatively easier or difficult problem in a contest you get the same credit. So a ideal contest is also a test of identifying the easier problems. Secondly if you consider the difficulty level of the problems you will see the 1st 6 problems were relatively easier, someone who knows little bit of coding can solve them easily (provided he has a little bit of brain, knows the most simple 3-line apsp code and knows a good bit of maths, well may be the problemsetters forgot it wasn't supposed to be a math olympiad or something like that rather a programming contest where a test of coding skills are equally important). Thirdly, between the final 4 problems' difficulty level and the difficulty of the 1st 6 problems there was quite a bit of gap, which once again reflected the lack of sense in designing the problemset. Altogether despite of the 4 nice problems the problemset fails to stand out to be a good one. This is really sad. :(

Its not that I've something with the problemsetters, because I highly appreciate the problemset that the same band of people prepared for the icpc regional dhaka site. I know they are extremely talented but this time I guess too much of experimenting messed up what could easily have been a nice problemset.

Anyway all these is just my personal opinion and there are more people I know who thinks the same way and ofcourse there are people who don't.

shanto86: I guess solving the 1st 6 problems from NCPC has made you a little too confident. Your above code clearly shows you've no idea about complexity calculation or anything. Well but don't be upset cause you wouldn't have needed to be a good programmer to rank within top 4 in NCPC cause the 6 problems you have solved would get you there, doesn't matter that you can't even realise that a exhaustive dfs without any pruning will never solve problems having state space of this size. Don't be upset through time you will learn all these but the only thing to worry for is that if you face similar problemsets in future then learning all those good things won't do you any good cause you can get top rank even without knowing them. Unfortunately if you attend the waterloo contests or contests at russian sites I bet you can't solve more than 1 or 2(!) problems and that won't be good enough for even a place in top 30. Thanks god you will have more genius people(!) preparing problemsets for NCPC, so without learning much about programming you can still get top rank. :lol:

Just for the record, at NCPC only 1 team solved 1 more problem apart from the 1st 6 problems. And the last problem present at the online contest wasn't present over there, so now you know what I meant by problems sorted according to difficulty level.
shahriar_manzoor
System administrator & Problemsetter
Posts: 399
Joined: Sat Jan 12, 2002 2:00 am

hmm

Post by shahriar_manzoor »

I don't know if you are a Bangladeshi or not as your identiy does not say anything. First let me tell you that I am not upset or angry with your comments. But as far as teams in Bangladesh is concerned the problem set is ok. One year back we had three very good team in Bangladesh but now there is only one. If the NCPC problem set was like Dhaka regional then the ranklist would have been a disaster because you see in Dhaka regional the second best team solved 2 problems. So the problemset was kept that way intentionally. We are having a gap as far as world class teams are concerned. Some good teams are coming up and we just needed to inspire them. So in international sense the problemset was not fair but as national sense (of Bangladesh) we did not have many other options (May be we could make it a little harder). And there was no excuse if a good team failed to solve six problems because they were sorted :)
Observer
Guru
Posts: 570
Joined: Sat May 10, 2003 4:20 am
Location: Hong Kong

Post by Observer »

I think that the problemset is very well-set indeed! Except maybe problem A - Prime Frequency, which I believe is quite dull...... (Please don't beat me up for this comment!!!)

Thx kzaman010! I finally get accepted with "Roses"!! Yeah~~ :D
There aren't so many states we have to care about. A simple graph search gets me an AC.
7th Contest of Newbies
Date: December 31st, 2011 (Saturday)
Time: 12:00 - 16:00 (UTC)
URL: http://uva.onlinejudge.org
shahriar_manzoor
System administrator & Problemsetter
Posts: 399
Joined: Sat Jan 12, 2002 2:00 am

hmm

Post by shahriar_manzoor »

No doubt problem A was dull :D . But u know there it is good to have a problem that can be solved by novice teams. And always me and kmhasan (more frequently me) has to come up with an idea as problemsetters are generally concerned with their images (What people will think if he sets a dull problem :) ).
Dreamer#1
Learning poster
Posts: 77
Joined: Tue Oct 07, 2003 10:07 pm

Post by Dreamer#1 »

I and J are nice problems. It felt really good to get them Accepted. :)

I don't have any theoretical knowledge about the tower of hanoi problem (since I'm not a CS student) so may be I'll consult a book before I try G. :o
shanto86
Experienced poster
Posts: 160
Joined: Wed Jul 30, 2003 8:10 pm

Post by shanto86 »

Well, you were talking with prunning while searching in graph right?

so can you please tell me what is prunning? and how it is done? I would be glad if you tell me this!
Self judging is the best judging!
Cho
A great helper
Posts: 274
Joined: Wed Oct 20, 2004 11:51 pm
Location: Hong Kong

Post by Cho »

I've no much idea what this thread is discussing at the beginning since I don't know what NCPC and HSC are. (NCPC should be a regional contest, right?) But I know what you guys are talking now since you mentioned IOI, IMO which I know what they are.

So, shanto86, you are high school student? Then a lot of problems here are just to hard for you. Most of us here have taken formal courses about algorithms from university. Courses, not just one course. A lot of problems here just cannot be solved without those knowledge. Knowledge is built by many talent people in many years. If you are genius, you may be able to figure out a few algorithms without taking any course or reading any book. But this is not enough to solve all the problems. We really do know some solid knowledge that you don't know yet.

Fermats Last Theorem looks very simple, isn't it. And a student yells "Come on! professor, it's so simple. It's proof wouldn't be hard. Please tell me. I do want to learn". This is not right.

Get back to the rose problem. How prunning should be done may be the core hardness of this problem. Asking for how that is done may be too much. If I tell you what I did directly, you will lose all the fun. In ordinary shortest path problem, there is a cost associated to a path. But in this problem, I associate four costs to a path. The number of possible states would be 10x10x10x10x21x21. I skip my meaning of my "state" here. Actually, the largest block of memory in my code is int cost[10][10][10][10][21].
If it's too hard, just let it go.
unknown_error
New poster
Posts: 5
Joined: Fri Dec 10, 2004 9:42 am

Post by unknown_error »

to Mr. Shahriar Manzoor: I am not agree with you in some cases about ncpc problemset. i know at present there is no very good team in bangladesh ( buet explorer's 2 member retires very soon. ). icpc dhaka regional perfromance was not good ( it was the first contest this year here, 2004 ). but u may notice that after the regional contest few teams really tried hard in solving problem in valladolid site. some of them solve some really hard problems.

There were 5/6 problems in ncpc where u don't need to know any computer knowledge except C laguage + high school geometry. acclording to u all the teams are not good, they took almost 5hrs to get the 5/6 problems and get good positoins.

Do u think that is this way good teams will develop? if there ae 6/7 that type of problems available, no one will read or try the 3/4 problems since they will be in some sorted order. :lol:

if u follows the icpc weak regions problemset .. they set 1/2 simple problems + 3/4 problems on basic algo/number theory + 2/3 hard problmes. if the problemset doesn't reflect the need of reading books or any basic knowledge of computer science, no body will be interested to read books. Do u think the past good teams from BD influenced by solving this type of problemset?

This is only my opinion. Our team also ranked within top 10.
sohel
Guru
Posts: 856
Joined: Thu Jan 30, 2003 5:50 am
Location: New York

ditto...

Post by sohel »

Some of the points that unknown_error made was valid and I seem to agree with those. I wonder why he kept his identity secret.
shahriar_manzoor
System administrator & Problemsetter
Posts: 399
Joined: Sat Jan 12, 2002 2:00 am

hmm!

Post by shahriar_manzoor »

All people getting death penalties will not be satisfied with the laws and it is impossible for me to satisfy all the contestants.

I don't think lack of practice can be a good excuse for solving 2 problems in a contest. I personally think if the NCPC problemset was like Dhaka Regional then the result would have been similar.

Why no DP? the answer is that no problemsetter could provide us with a good DP. Because I don't think any one would consider LCS, Matrix Mult a DP now a days.

The Hanoi problem and the BFS problem was rated medium by more than one person but the contestants had other ideas, how can we counter that?

Everyone praise dhaka regional but after the dhaka regional mass critisism was present locally. Did you then tell the critisizers that the problemset was ok but we could not solve and it was our own fault?

I don't think there was any team in NCPC which was biting their nails after solving 6 problems. The teams who solved six problems solved it in last moment and if a team has not solved six can they have any point to argue?

The champion team could not solve eight for their own silly mistake, it was not the problem setters fault. But I do welcome and note your critisism but I would be glad if you could do it with your identity.

We do have a general tendency to look for good example in abroad, we always tend to consider what happens in Foreign land is good. Please don't set examples of weak regionals with these online contests as many of the weak regionas still just struggles to provide eight new and error free problems.


-Shahriar Manzoor
kmhasan
Problemsetter
Posts: 107
Joined: Fri Oct 26, 2001 2:00 am
Location: Canada
Contact:

Post by kmhasan »

I feel a bit itchy
Larry
Guru
Posts: 647
Joined: Wed Jun 26, 2002 10:12 pm
Location: Hong Kong and New York City
Contact:

Post by Larry »

I agree with KM. While it's always fun to tackle harder problems, we shouldn't make contests consisting of only harder problems, we need some easier problems too.

Even the ones that are "well-known", but maybe not to the newcomers. When I first started, I think I didn't solve a problem in 5 or 6 sets.. many friends already quitted by then..
Dreamer#1
Learning poster
Posts: 77
Joined: Tue Oct 07, 2003 10:07 pm

Post by Dreamer#1 »

I'm not writing to criticize or to blame anyone but if the contest didn't have those easy problems the outcome would have been different for sure. not everyone is lucky to be a part of a good team and with so many problems to solve its almost impossible to uplift a team by solving 5+ problems alone. but again i've none to blame for that but my fate.

but ofcourse if you consider the last 5 problems i guess after getting 9 out of 10 of them AC (can't solve G cause i lack the knowledge needed to solve it) i can safely say i could solve 4 out of the last 5 of them in real time too if only i was physically ok or if my teammates had given me a hand in solving the easy problems. not all of the top 10 teams have that kind of ability. but in any case its only myself and my luck that i can blame.

anyways its not all over yet. i still have over a year left, so, long way to go. may be next time, provided that i can form a better team... or else i've to be satisfied with something that i can do individually, topcoder.

and to all the problemsetters, solving your problems has always been a pleasure... hats off to u guys...

and to everyone... guys, no hard feelings, alrite? there is always a next time so forget the past and get charged up for the future or you've to come here and write the same thing time after time...

and to the teams that did well at ncpc... congrats! :)
kzaman010
Problemsetter
Posts: 18
Joined: Wed Jan 16, 2002 2:00 am
Location: Bangladesh
Contact:

Post by kzaman010 »

but if the contest didn't have those easy problems the outcome would have been different for sure
Are u sure? Compare the ranklist with Regional Dhaka site. The top 3 Bangladeshi teams are unchanged. NDC team did not participate in the regional and I am sure they would at least solve 2 probs in the regional if they did participate.
with so many problems to solve its almost impossible to uplift a team by solving 5+ problems alone.
May I know, how many did you solve alone? If u cannot solve easy ones (It took 5*3 hrs to solve 5-6 easy ones !!! for those who are complaining) how can I expect u can solve difficult ones.
i can safely say i could solve 4 out of the last 5 of them in real time too if only i was physically ok or if my teammates had given me a hand in solving the easy problems
If u can do so, u will definitely be the ALL TIME GREAT BANGLADESHI Contestant. I am eagerly waiting to see that.
Md. Kamruzzaman
Member, Elite problemsetter's panel
Faculty, American University Bangladesh
Graduate Student, BUET
shamim
A great helper
Posts: 498
Joined: Mon Dec 30, 2002 10:10 am
Location: Bozeman, Montana, USA

Let's Play detective

Post by shamim »

Hi, this post is regarding people involved in NCPC2004, Bangladesh, so if you are not involved you may skip it.

Dreamer, by mentioning that you were not physically sound on the day, you have geven away your true identity. And the words regarding your partners confirm that.
Last edited by shamim on Tue Jan 25, 2005 6:42 am, edited 1 time in total.
Post Reply

Return to “Other words”