I want to bring attention to an issue that may or may not have affected you:
In today's contest, 1D's statement was changed for me about 100 minutes into the round:
I don't have a screenshot of the original problem page, but before it was updated, the problem asked not only for a valid construction, but also to count the number of valid constructions, and the samples reflected that version.
Although the change was announced during the contest, I didn't see the notification because I wasn't actively viewing the contest page (because I had my IDE and ms paint open while solving). By the time I noticed the announcement around 2 hours in, I had already spent close to an hour trying to solve the counting part.
The announcement says this was caused by a display bug. I'm not sure how widespread this issue was, but I imagine I wasn't the only one affected. Please tell me if you encountered the same problem.
I do not know whose fault this is, but I think this, alongside that 1F seems to be a standard problem + author communicating with participants during contest, is grounds for making the round unrated. I would also like others' opinions on this matter.
Guess who caught it?
)
orz
orz
Auto comment: topic has been updated by temporary1 (previous revision, new revision, compare).
I 100% agree with the consensus that round should be unrated, It's quite shocking that the authors published an already known problem, and also the two main authors "history" with competitive programming. Surely this is an issue.
MikeMirzayanov KAN
Cheaters can't make good contest.
*chromate00 cough cough*
Auto comment: topic has been updated by temporary1 (previous revision, new revision, compare).
I faced a stupid issue for problem E1 , i was solving (as per the ps) with input format q and k and was getting wa on pretest 1 but couldn't figure it out because even samples were according to that (q and k) , but after that i got to know its k and q not q and k. There was no annoncement for it also
See my comment below.
I believe that it's CF mixing different revisions of statements. the "q and k" is an old revision, and the "k and q" is the correct version. We changed this a few hours before the contest.
We never know this issue during the contest :( Sorry for the inconvenience.
Also, the PS of D was buggy for me. I was solving a different problem (number of ways and all), and this bug in E1 cost me 17 minutes (you can see time diff between my first submission with wa on pretest1 and my ac one). Because of that, I missed D and got a -50 delta. So shouldn't this round be unrated?
It is possibly a bug of CF. I have no idea why it happened. For me, when I was changing the location of my VPN, the statements also changed. For example, in SGP it displays the correct statements. However, in JPN it displays the wrong statements. Even worse, when I disabled VPN, I simply can't open any statements.
We are so sorry for the inconvenience, and we are trying to find the reason behind it.
This does not look reasonable for me. I can confirm that the authors invented 1F on their own, and the fact that 1F is not new affected a rather small number of participants. Although the authors are busying sending messages in chat groups during the contest, I think their messages did not give any advantages on solving any problem. Their action is unsuitable, but this does not affect the fairness, I think.
Why is it not reasonable to at least unrate particular users on request like it's usually done in such situations? temporary1 claims to have wasted an hour on the wrong problem which is a huge disadvantage
Maybe unrated for Div1 only
because I did quite well so I want Div2 to be rated?Haven't seen a contest with some many bugs:)
can't believe i picked THIS contest to return to CF with, atp i might just go to AtCoder