can someone help me :( I solved almost 950 problems on codeforces of difficulty level till 2000 but still I am not able to perform well at contest.. where I am going wrong
→ Обратите внимание
До соревнования
CodeTON Round 9 (Div. 1 + Div. 2, Rated, Prizes!)
25:01:09
Зарегистрироваться »
CodeTON Round 9 (Div. 1 + Div. 2, Rated, Prizes!)
25:01:09
Зарегистрироваться »
*есть доп. регистрация
→ Лидеры (рейтинг)
| № | Пользователь | Рейтинг |
|---|---|---|
| 1 | tourist | 4009 |
| 2 | jiangly | 3823 |
| 3 | Benq | 3738 |
| 4 | Radewoosh | 3633 |
| 5 | jqdai0815 | 3620 |
| 6 | orzdevinwang | 3529 |
| 7 | ecnerwala | 3446 |
| 8 | Um_nik | 3396 |
| 9 | ksun48 | 3390 |
| 10 | gamegame | 3386 |
| Страны | Города | Организации | Всё → |
→ Лидеры (вклад)
| № | Пользователь | Вклад |
|---|---|---|
| 1 | cry | 167 |
| 2 | Um_nik | 163 |
| 3 | maomao90 | 162 |
| 3 | atcoder_official | 162 |
| 5 | adamant | 159 |
| 6 | -is-this-fft- | 158 |
| 7 | awoo | 157 |
| 8 | TheScrasse | 154 |
| 9 | Dominater069 | 153 |
| 9 | nor | 153 |
→ Найти пользователя
→ Прямой эфир
Codeforces (c) Copyright 2010-2024 Михаил Мирзаянов
Соревнования по программированию 2.0
Время на сервере: 22.11.2024 16:33:52 (j2).
Десктопная версия, переключиться на мобильную.
При поддержке
Auto comment: topic has been updated by btme (previous revision, new revision, compare).
Do you know where you lost the match?... See your calendar between Jan and Feb...
2 months break --> Restart again
You can learn few new topics maybe:-)
What the hell is there to dislike isn't it a fact!
my college xam were going on i didn't had an option
I sometimes skip my exams over a codeforces round. Not an advice.
The Truth is 'nobody knows'.Life is just unfair. You just gotta accept it.
I too feel the same..
Solve more hard problems. Don't limit yourself to any difficulty range. Solve problems. I mean problems. And problems aren't easy to solve. If you find a problem easy for you left it. It's also a good practice to see editorials of good problems and you can know how good coders think. And maybe you are solving so many problems with your comfortable tags. Don't see tags, solve randomly, don't see failed test cases etc.
Solve problems that present a problem to you, and take actual thinking to solve. Don't just answer questions like a math test, because how will that help you learn and improve?
Can feel your frustration, my name is too ironic for my stagnant profile.
u reached expert by solving 300 problems ORZ
Nah I just remained an expert, because I don't practice :( I deserve to be stagnant anyway, good luck!
121 here... nah jk I actually solved problems before joining codeforces.
"the master has failed more times than the beginner has tried"
Or maybe not...
check you stats in codeforces visualizer
One thing I can say is ( even thought I did not have great rating profile ) try to solve problems in less attempts . You can see if you submit your code there is just 28% chance that your code get accepted
For motivation : The eligibility to see best in life is to see worst in life
ALL THE BEST !!
Indeed... comparatively to me, there are many times that btme just starts spamming solutions, even in contests, where this can be heavily penalized. Try to think why a solution would work (or wouldn't work) memory-wise, time-wise. Try to think about corner cases, don't just accept that it feels intuitive enough so it must work. Of course, I'm not saying write full mathematical proofs, but at least see why it would be a general solution to every test case. If you change something, usually try to justify to yourself why the previous code would have been bugged in the first place (does this change actually change the output of a particular test you'd have in mind? Would the particular case you'd have in mind even be possible or are you making the change for hardly any reason?)
problem is that i can't give so much time on problem like B.. if I start proving it will take a lot of time :(
That time is likely less than the penalty you'll get for more WA's as well as the extra stress you will put on yourself to solve it faster. Just relax. You came here to codeforces to have fun. So just have fun. If you solve another problem, good! But otherwise, you know that you have something you can learn from and do better next contest.
(Kinda cliche but it's true)
mood. (but i feel u)
tbh, I am in no place to give tips but you should start giving virtual contests(if you haven't tried that) , I started in Jan this year and till April I couldn't even reach pupil stably. I knew I could do 2-3 problems in Div2s but in contest I just couldn't do anything. I started giving VCs from Apr 4, reached pupil on Apr 21 :). I kept giving VCs everyday after that , thats why my graph is dark green there lol. It helps tbh. Also meditate maybe to stay calm, because I used to be really scared in real contests but with each VC I started to get more calm. I already knew when I was specialist level. Became specialist in next 10 days. rn I know I can upgrade again:) but I have some other engagements also, might take a month.
Can u share grp links where did u give VC
don't have a group, I just started from #570(random really) and made sure that I dont solve problems above that contest id in random practice.
recently i started doing VCs..i can feel some improvement in my speed...what rated problems did u solve for that transition from pupil to specialist? or you just did random VCs?
they were random only, in the starting I had too much free time, so I did ladders also, but wont recommend
why wouldn't you recommend ladders(a20j right?),Is there something wrong with them?
Actually,I have just started with ladders & wanted to know if I'm doing something wrong.
ladders are really good to start, dw, I meant if you have less time then prioritize VC , cuz ladders are random problems and to be good at contests you need other things like speed and accuracy also, so VCs are better
Could you please tell us more about how do you solve these problems? Are you picking specific topics, which are based on problem tags when training? If yes, then this gives you some extra information, which is not available during contests. Are tags enabled for unsolved problems on the codeforces platform in your account settings? If yes, then it's a major spoiler too. Do you have a habit of reading editorials for difficult problems right before implementing and submitting solutions for these problems? Well, you get the idea. The "number of solved problems" and their "difficulty" statistics may be interpreted in different ways, depending on your personal training style. I think that these stats are just not directly comparable for different accounts.
I'm still trying to figure out the optimal training schedule myself. I'm typically upsolving problems after a contest without reading any spoilers at least for a few hours, assuming that I'm not too tired on that day. Then I'm reading the editorials to get the general idea about the remaining harder problems, but I'm not trying to write code for them. This is both good and bad at the same time. The good thing is that I'm not inflating the "difficulty" of solved problems in my account stats artificially. The bad thing is that neglecting to write the actual code is making my training less efficient. And right now I'm considering to use some other platforms for this kind of training. I mean, read all AtCoder editorials, try to implement the ideas from their editorials and submit the resulting solutions.
I had the same problem but I continued to practice and one day I placed 83th place in Div 2 and reached expert :).
So don't give up and try to solve more hard problems and one day you will perform very well during contest.
One tip that I have not heard many mention is to reflect on problems once you've attempted them. After solving a problem or giving up on it and reading the editorial (which should not be done until after serious thinking) you should ask your self the questions: Why did I not solve this? How would someone come up with the intended solution? Could your thought process have been any better? and so forth.
Agreed and maybe watching video explanation from good coder might help you.
Quit codeforces try atcoder.
Any specific reason?
It is not about problem count. It is more about problem type.
Consider this: How many times have you seen a div 2 problem whose only solution is FFT/Suffix Array/CRT/Li Chao Tree/Ji Driver Tree/Segment Tree Beats?
More often than not, you will see div 2 problems that are more ad-hoc in nature (maybe requiring some simple math/observation).
The reason why some people can solve complex problems but fail to solve simple ones is because they haven't quite learnt to exercise their problem solving skills.
Complex problems can sometimes be nuked by some well-known googleable algo invented by some genius. However, simple (but creative) problems often take unpredictable forms and do not fit into pre-made algos so easily. This is where the strong players shine as they have strong problem solving skills as opposed to the weaker ones who look to nuke problems purely with well-known algos.
exercise and VP
try virtual contest as many as you can. in order to increase rating, speed really really matters.
for me problem B is more difficult than C as B requires one observation should I try to do more ADHOC problem will it help ? and because of this I waste lot of time on B
i suggest you to just repeat participating virtual contest and upsolving them. you know your problem is that you are not fast, not that you can't solve harder problems in contest.
Can you suggest how much to up solve after a VC. Should I try to solve all the problems or only up to a certain level depending on my rating ?
i think as long as you can understand solution, upsolving all problems are always better. but if you fail to understand editorial that problem is toooo hard for you currently.
Everybody has some different story. It may be possible that you need more practice. You should work on your speed also. Keep trying and trying, one day you will succeed. All the best btme
see my profile
see Qualified's profile
u solved just 150 problem to become expert?
He's SkyDreams on discord, AKA he's too scared to use his main. wkek
Same problem :((
All I can say is you are not alone :)
All I can say is you are not alone too :)
I was stuck like you for a really long time. Like you, I could solve higher rating problems but couldn't solve them during contest. But i learned few, things
1. VC increases your rating!
2. stop solving so much higher rating than your current rating. try +200 ratings atmost. Solve one particular rating problems until you feel really confortable.
3. It may be a controversial one but learn some ds (it never limits you learning something new).
4. Solve out of your comfort zone and stop worrying about rating!
5 and final one : Hang in there, it's really worth it, i'm saying from experience man!!
In my experience, all you need to know to get a decent rating of about ~1700 is know how to solve stupid greedy
In my opinion, just never giveup. And if you’re stuck, try new ways of practicing. Be sure that you will finally reach your goal.
Solve problems with interest and concentration.
after doing VCs now i am quite comfortable with 1400 rated ones(not much in contests though). for becoming comfortable with 1500 ones did u just start doing random ones from problemset? i am about to practice 1500 ones..just started recently...and right now do u focus more on solving B faster on doing C during a contest?
I filtered by 1500-1600 tags, switched off problem tags, and solved 3 problems each day, though I'm not consistent now due to exams. During contest, I solve problems with higher number of submissions first.
I'm not sure but try learning new things (algorithms & data structures). Also maybe try to solve the unsolved problems after the contest
Compare your growth with Chinese bamboo tree
A Chinese bamboo tree takes five years to grow. It has to be watered and fertilized in the ground where it has been planted every day. It doesn’t break through the ground for five years. After five years, once it breaks through the ground, it will grow 90 feet tall in five weeks! The question is did the Chinese bamboo tree grow 90 feet tall in five weeks or in five years? Says Les Brown, the answer is obvious, it grows 90 feet tall in 5 years. If at any time you stopped watering and fertilizing that tree, it would die in the ground.
Growth is like a A Chinese bamboo tree. It takes time to establish. Time more than you expect
Start solving from div2 C during the contest.
Solve more. Don't give up.
You have done like 10 2000-problems, which account for 1% of your solved problems. Your average difficulty is 1500 so stop being obsessed with the count.
I have solved a similar number of problems of a similar difficulty (maybe a little harder, but not by much) and I'm 2140...
Weird flex but ok
Point is that blogs like are these pointless.
Try Harder
If you can solve problems with similar speed during contests and outside of them, you should eventually be able to get the rating you expect. If you are worse or slower during contests, it means you perform worse under the pressure.
I used to have the same problem with you. I solved a lot of difficult problems in other platform but I still got a low rank when I take part in a contest.
Thanks to my teacher, he gave me some suggestions. He pointed out that solving problems in contest is not quite similar to solving problems in usual. That's because that you may see some algorithm tags when you are not in contest which may effect your thinking. When you are in a contest, you should analysis a problem from nothing. So this is very different.
In my opinion, I think you may take part in some more contests (if you are care about your rating you can choose Virtual participation) It may be hard in the beginning, but soon you will get better. Hope my suggestions will be helpful.
Did you solve problems by reading the editorial and then implementing the solution or did you actually solve them from scratch? There's a big difference. I'm not judging, just asking.
Just keep doing, Good things will come eventually. You can try one more thing don't look at standing during the contest. And can try in ACB in the contest. Also, keep doing virtual contests and upsolve them. You can check out my profile.
Go ahead for another 950 problems... and if it isn't enough, do another 950 and so on. I'm pretty sure that at some point it will worth buddy... GLHF!
I am also facing the same situation with you, It seems that I have been solved many difficulty problems (in another online judge) but can't get good score in contests. I think the touble with me is I can't thinking about the problems long time by myself, I always solve problems by reading the editoral. So I think we should try to solve problem by ourselves to improve our skill.