Winter Challenge 2023 discussion

I’m opening this thread to talk about the contest

Personally, I was interested but I gotta admit that I wasn’t very impressed. The IDE was laggy and the tests timed out half of the time.

It also wasn’t as fun as usual. The tight time limit made it feel like a school exam more than anything. I don’t think I will participate in another event like this.

When it didn’t lag, the IDE was good though.

What do you guys think?


There is no point is discussing the contest.
Let’s forget about it and come back to bots/optimatizations.


For me tests always timed out, I had to copy my code into my IDE and craft/copy tests by hand… The experience was quite below my expectations, compared with bot challenges in-game IDE was pretty much unusable for me. Also it was not helping that copy/paste was turned off in game description area.

On the bright side, nice story line, I enjoyed puzzle descriptions and I think if in-game IDE was not in such bad shape 2 hours was quite enough for all of the puzzles.


It was fun to solve the puzzles, I’m sorry for organizers that the environment didn’t work as expected, I’d be happy to see more contests like this in future :crossed_fingers:


Hello all, the conquest was quite fun, just a little too short: in fact, i loosed few minutes at start: the introduction text make me Believe a had 2 hours when i launch the thing… my bad… lost a few more minute understanding the “new” interface, unable to enlarge the code part of the screen, loosing multiple time trying to test my code, no answer, (was a ; missing at the end of a line), but no indication…

I jumped to part 4 after few try on 2 and 3, wasn’t abble to check input, i come back after finished the 4th… and understood it was my bad…

So just for next time, please let us more time: 3 hours can be realy usefull, specialy when people like me only coming for challenge, have to do with a new interface…

I realy want to retry it, i want to finish, i haven’t the time to read the 7th, optimize my 6, and do the 3rd… and more than everything, read the full text of the story =D

thx for the fun :slight_smile:

P.S.: and with 1 more hour, anyway, ppl who have done everything in 2 hours will stay on the top, it will just make some difference between ppl that need more time like me… so… more intels, and a more “true” ranking :slight_smile: (imho)


Oh right, I forgot to mention, but at the end of the contest, the feedback page instantly redirected to the contest home page, so I couldn’t even give any feedback there.


First, thanks for the contest, I had lots of fun. But bots competitions are still better :slight_smile:
Puzzles’ descriptions/art were very flavorful and fun.

Agree on the laggy IDE part, tests were unusable most of the time.

Also couldn’t log in for the first 10 minutes. If I had these minutes back, I could’ve submit one more solution.
Would be nice next time if the contest time will be extended somewhat if there were technical issues on the server side.

Also I was puzzled (pun intended) about the scoring/leaderboard rules. And still am.
Do final scores depend on the total time when you submit your solution?
Do they depend on how many other participants already solved the task by the time?
Would I get any points if my solution passes only half of the tests?
Do I get penalty if I resubmit?

Knowing the rules beforehand would help deciding what task to take and when to submit.

Depending on the type of scoring, we could really have the content length at more than 2 hours for more relaxing/casual experience for those who want it, while still having fierce competition at the top of the ladder.
Speaking of which, would be very nice to see the ladder during the contest.

The other problem is I couldn’t see what version of a solution I had already submitted.
Combined with the “not knowing how the scoring works”, I didn’t know if it’s OK to just resubmit the latest/current version again.


Oh, I can actually answer some of that!
The blurb at the beginning did say that we could resubmit as many time as we want without penalty, and that each question had a maximum of 1000 points available (which I assume is compounded by what % of the validators are passed after submission)

That said, I couldn’t find any way to find that blurb back after I started the actual test so. It would have been nice to be able to access it any time to read those rules again.


I had that too, I thought it was my adblock or ghostery, but then I realized I had them both turned off on :)))

1 Like

What was that? This wasn’t the usual format and I realise that it was just 2 hours contest literally 2 hours before the challenge.
I thought it was going to be a duration of 2 hours and not just open for 2 hours from the release.
I went to have dinner and walk the dog so I can get 2 full hours without interrumption and when I’m back and ready I see that the challenge is over.

Next time, it would be good if is more clear that is a change in the format of the challenge. Or you can keep the challenge open for a few days but only for 2 hours once you click start.

I’m a bit disappointed as I always look forward for the challenges and now I just have to wait 3 months for the next one.

1 Like

I’m sorry but I can’t say that I had fun participating in this. There is always a “but…” for every pros.

First of all having a solo puzzle event instead of bot contest is a bit disappointing, though I understand the reason for that. Technical issues are understandable, but it’s impossible to know what’s going on without being participated in discord.

It’s nice to have a story part but I spent too much time to figure out what I’m supposed to do. It’s even worse if English is not your native language. So it would be better if there were some clear instructions in addition (like “count the number of paths in the graph that …” for problem 7)

New IDE looks nice and I think new style of code structure is good, but its overall slowness and all those little things are very frustrating. Non-intuitive controls - every time I want to expand a test case I run this test because the “play button” is not just that outlined button in the right, as it should be, but the ALL clickable bar of test case excluding the up-down arrow on the left. Constant blinking of the language assist bar - trying to start, failed, stopped and all over again, can we just turn it off permanently please? Well I hope most of this gets fixed along with serverside lags.

Time limit is also very strict. I didn’t expect such algorithm-heavy tasks, thus I might be overestimate my skills, so it turned out to be too difficult to me.
I hardly believe one can solve problem 7 in about 120/7=~20 minutes from scratch unless you know exactly the right algorithm for this kind of problem.


If the challenge is open for a day, some people could/would cheat by creating a second account.


You make a good point about the long story part.
For the time limit, I would say the opposite. I was expecting the problems to be hard enough to make sure that no one could dream to finish so the leaderboard would rely less on completion time but more on points for test cases completed. I personnaly finished the 5 first problems in about 1 hour, Only four test cases were not working at the end and I probably would have solved this if I hadn’t lost time with the server problems. And I’m not saying this to complain about the server problems but because I think it shows that it was not difficult or long enough.

1 Like

I gave up because of the timeout issues. It hung a lot for me and I got a number for 504 errors and some “Verification time elapsed” error messages from the site.

Also, when I started, it forced me to go through some of the tutorial, even though I am a long-time user of Coding-game and went through the tutorial for the challenge. I wasted a few minutes in the beginning. I thought it was part of the puzzle and then I realized it was still introduction.

If it wasn’t for those issues, I would have enjoyed the challenge. I do hope there is another one and they will have improved on the issues by then.


I didn’t have fun at all. I’ve had the count down page up on my screen all week. Yesterday I did some clashes and the “Code of the Rings” as a warm up. The last challenge I had a really great time. Worked on it every night and got in the silver league. So I was really looking forward to it. Got work all done, ready to sit down and start the 2 hour challenge. What? It’s over? I’m feeling really bitter. This is just stupidity. Why didn’t you mention the timing??? Just feeling disappointed and bitter.

EDIT: Just found the email you all sent when I registered. I guess you did mention that it was exactly 3-5 US eastern time. So I guess it was my fault I missed it. Still disappointed but less bitter. I’m just going to sit here and eat a whole bag of candy.

1 Like

Started at 6am for me, was a bit rough to code so early in the morning.

I could hardly run any tests on problem 2, 3 and 4 (problem 1 was ok for some reason).

Long format is much more enjoyable and compatible with family life.

1 Like

I personally like the short 2h format, and this kind of puzzle challenges: right or wrong, no second guessing good enough vs not. The problems were well balanced too, with a nice difficulty curve, from trivial to extremely challenging.

Now, the problems with the IDE were a bit annoying, and I spent a fair amount of time copying inputs by hand from an image or a “prettified” list of lists into my IDE. Why not make the inputs of the test cases easy to copy/paste? Or perhaps I missed a possibility somewhere?

All graphical problems were blind submits with only local tests. In the last 5 minutes I went back and could run CG tests on one of them and found a bug (both in my solution and my local test), quickly fixed it, but then after trying to run the test again, the whole thing stopped responding until the contest ended, so no idea if I submitted the fixed solution or not.

All in all, making and running local tests is a skill too, it’s fine for me if it’s part of the contest, but it’s not really in the spirit of codingame (or the presentation of the problems, here).


I was looking forward to the contest at lot but I found it very disappointing. Not because of the quality of the 7 puzzles and their presentation (these were good), but because of the overloaded servers and the terrible performance in the web IDE. I found it very frustrating. The code assist was almost never working, running the tests timed out most of the time. Not fun at all. With this bad performance, the 2 hours just weren’t enough.

This was the umpteenth contest already, and still everything was completely overloaded and unusable. Why? Was the number of participants so much higher than usual? Can’t CodinGame scale upfront to make this fly?

Also, I must have missed that Perl was not allowed in the contest. I’m sure it must have been written somewhere, but I didn’t see it. I was surprised not being able to use it. Why are 27 programming languages normally offered, but in a contest only half of these are? What is the reason behind this? Let the coders choose the best tool for each solution.


Is there some message from organizers what went wrong, if we will have chance to see the problems again, or see the leaderboard eventually?


As far as I can tell, the reduced list of languages is not specific to the challenge, but to the new IDE. I’ve seen people mention Rust too in the same boat as you.

Also, I don’t know if there was more people than usual at the same time, but there was definitely a lot less registered participants. Previous contests peaked at 16-18k, this one had 6.7k total.

1 Like