Gordon-Freeman's blog

By Gordon-Freeman, history, 6 weeks ago, In English

So i have been practicing 1400 rated problems since i got to pupil i have solved so far 27 problems , i got stuck and had to read tutorials for about 12 of them and the rest i solved by myself , Now , is this ratio good? should i move on to 1500 rated problems? or should i solve more 1400 problems , also i have solved 29 problems rated 1200 after that i switched to 1400 rate because 1200 rate got too easy for me and i could find the solution to any 1200 problem but i don't feel the same about 1400 rated problems by the way i skipped problems that was solvable only using dp or graph since i didnt learn them yet and people suggested me to leave them and foucse on binary search and greedy instade , despite that i can't say i can solve any 1400 rated problem at the same time i can solve between 50%-60% of them.

So what do you think should i switch or stay at 1400 until i find them too easy?

»
6 weeks ago, # |
  Vote: I like it 0 Vote: I do not like it

That's fine. You can practice 1400-1600 if you want to but don't hesitantly move from this range until you can solve 50%-60% of them. Also, it's good you look at the tutorials (after exhausting all your ideas). Besides, those problems are at a new level and may require you to know some tricks (which you can get from editorial or solutions). Just my POV though

»
6 weeks ago, # |
  Vote: I like it 0 Vote: I do not like it

Are you doing problems randomly or following some ladder

  • »
    »
    6 weeks ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    both, i mix probems from TLE cp sheet with other random problems because i think the sheet lack some math problems so i get pick those randomly but otherwise the sheet have really good greedy problems

»
6 weeks ago, # |
Rev. 2   Vote: I like it 0 Vote: I do not like it

In my experience, you have to solve tasks based on a rating of about 60 tasks, that is, if it becomes a little too easy for you to solve 1400 tasks .And reading the tutorial of tasks is normal, when I solve tasks, I try to follow this plan — to make an idea in a maximum of 25 minutes and write code in 30 minutes , well, if I can't do this , I'll read the solution , but if I don't understand the solution , I look at the codes of others , if it didn't help , it means that the task is shit and I skip it. As for dp, you need to learn at least the base , because <=1400 because I have met quite a lot of dp tasks . As for graph, you can learn it later , believe me, you can reach an expert without knowing the graph.

P.S — Solve new problems

  • »
    »
    6 weeks ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    that's exactly what iam doing (which helped me improve) but you said to solve about 60 problems, isn't that too much? i mean i manged to get very good at all rating below 1300(solving about 170 problems combined) the difference in now i am not that much comfortable with 1400 rating and as i said i can solve about 50% of them so I don't know if i should moveon to the next rating or not

»
6 weeks ago, # |
  Vote: I like it 0 Vote: I do not like it

Focus on greedy if you are under 1400, maybe even 1600. Most of the time you won't see graph or dp problems (If they appear you probably won't be able to solve them, but it's probably for the best not to put all your effort learning them and practicing them at least for now). But in my humble opinion greedy helps with the logic for the majority of problems one may encounter. Also I'd say problems rating ain't that important, just solve problems that ain't absolutely easy for you and keep with the grind.

  • »
    »
    6 weeks ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    that exactly what i wanted to hear, but one more thing you said to solve problems that ain't too easy for me, how i determine that for a rate? i mean as i said i can solve about 50-60% of 1400 rated peoblems dose that mean its easy and i should move to 1500 or should i get better with 1400 problems

    • »
      »
      »
      6 weeks ago, # ^ |
        Vote: I like it 0 Vote: I do not like it

      If you can't solve half of those problems how can they be easy?, If you can't solve that half because you don't know some topics it is time to learn them. Anyways with too easy I just refer to reading them and clearly knowing the solution.

»
6 weeks ago, # |
  Vote: I like it 0 Vote: I do not like it

switch

  • »
    »
    6 weeks ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    The guy above you said to stay at 1400 since being able to solve half of them isn't enough , whats your opinion?

    • »
      »
      »
      6 weeks ago, # ^ |
        Vote: I like it 0 Vote: I do not like it

      I think you shouldnt practice in your comfort zone.

»
6 weeks ago, # |
Rev. 4   Vote: I like it 0 Vote: I do not like it

i suggest u try and solve 1600-1900 problems as most of them dont require any advanced topics and also this is what i used to train at whilst being pupil and it helped me have a consistent 1500+ performance after 18 contests in pupil

note that it showed performance after around 40 problems in the above said range

also another thing its ok to see the tutorial for around 80% of them

  • »
    »
    6 weeks ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    I just looked at your profile and.. WOW expert by solving 358 problems only? that's just very impressive i will definitely do what you've said , do you have any other suggestions? like what topics to focuse on mostly , and what topics i should ignore. thanks.

    • »
      »
      »
      6 weeks ago, # ^ |
        Vote: I like it +1 Vote: I do not like it

      dont learn any algos / ds until u encounter them in more than (5 — 10)% of the problems u are solving , and also u can try solving really hard problems for you for example problems 600-800 above your rating u'll most likely see the tutorial for more than 90% of them but it builds strong intuition and gives many ideas to start thinking from also it kills the fear u have and simplifies everything for you

      • »
        »
        »
        »
        6 weeks ago, # ^ |
          Vote: I like it 0 Vote: I do not like it

        okay i will do that and hopefully i reach expert like you soon. thanks for the advice

»
6 weeks ago, # |
  Vote: I like it +8 Vote: I do not like it

In normal case, if u can solve 50% of a set of problems then it's suitable for u. So I suggest u keep practising and move on when u can solve it easily. But problem rating is not an accurate thing so I suggest u practising problems in a rating range rather than a number(for example,me 2100~2400). Good luck!

»
6 weeks ago, # |
  Vote: I like it 0 Vote: I do not like it

stay at 1400, learn graph