We will hold AtCoder Regular Contest 175.
- Contest URL: https://atcoder.jp/contests/arc175
- Start Time: http://www.timeanddate.com/worldclock/fixedtime.html?iso=20240324T2100&p1=248
- Duration: 120 minutes
- Number of Tasks: 6
- Writer: Kite_kuma, nok0, zkou
- Rated range: — 2799
The point values will be 400-400-600-700-800-1000.
We are looking forward to your participation!
overlaps with EUC mirror :(
Have you ever seen in history atcoder changing their contest time for other contests? Never.Great/bad (that judgement is upto you) thing about atcoder is it never compromises its' own contest times for other contests
How can you solve these HARD problems so FAST? negative again:(
Could not code C in time :(
The code of C is actually very short.
Interesting, could you explain it? I also needed like 80+ lines :(
Finally broke the streak of rk145 and got rk149... Anyway, good contest.
I guess the authors like constructive problems veeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeery much!!!
This was a bit harder than the other recent arc for me.
Tbh ARC has very bad difficulty distribution many times because problems from C become interesting exclusively to IM+ levels, while A and B can rb solved by Specialist and below. I find all these problems very educational nonetheless but the difficulty curve is not the most optimal. I personally see no reason for ARC to be rated below 1800-1900 simply because it turns into speedforces a lot of times.
If last ARC had today's C as a D, it would have been an example of a well-balanced ARC.
Yes, maybe B and E harder as ARC173B and this contest's E?
Anything other than what we got yesterday would be great. If only problems from C were present yesterday you'd see a well distributed leaderboard, this way 80% of competition shared A and B and time was a tiebreaker.
if you have people who can only solve 0/1/2 problems, naturally its always going to be speedforces for them regardless of the balancing.
What you are suggesting is making arc easier so that its speedforces on 3 problems for you instead of 2. That isnt enjoyable for the actual audience between 2000 and 2800 rating imo. Last contest had perfect A, C and D level for me (tho CD swap personally). B was a bit too easy but fine.
The only flaw in the last contest difficulty wise imo is the DE gap.
No, I am not suggesting it be easier for me. I am saying remove A and B completely, as it's really not interesting seeing two 1000 rated problems.
In an ideal scenario I'd like to see a 1000-1400-1800-2200-2600-3100 distribution. This leaves everyone satisfied. Ofc it is not possible for an author to predict this.
How was difficulty good when it appears if you solved A you were pretty much guaranteed B, and if you solved C you were guaranteed D, yet both B-C and D-E jumps were hugeee???
ARC 174 had miles better difficulty distribution for A-C, but D messed it up as well. Either C or D did not belong there.
And how is it speedforces for me exactly if C is 1700-1800??? It's a literal filler problem between 1000 and 2100.
All ARC have interesting problems for people in 2100-3000 CF yet there's not a single problem that is not completely trivial but somewhat challenging for the lower rated participants. It's either completely trivial uninteresting shit problems like B and A yesterday, or untouchable like C yesterday.
What's the point of having it be rated for people at 1500-1700 rating when you know you'll gain 60 points if you are fast and lose 60 if you're slow? You cannot differentiate between a AtCpder green and an AtCoder blue rating, yet in ABC one beats the other by 3000-7000 places.
Is there anyone who would like to share your ideas about problem C? Thank you so much.
I feel there's a mistake in the editorial for C. In proposition 1.3, the part where the value $$$f_k(x)$$$ is broken down into 3 cases, $$$x \lt l_k$$$, $$$x \gt r_k$$$ and $$$x \in [l_k,r_k]$$$, shouldn't it be $$$l_{k+1}$$$ and $$$r_{k+1}$$$ instead of $$$l_k$$$ and $$$r_k$$$?