Sorry for asking this 69 IQ question for yet another time, but I really have no idea what situation I'm in.
I started solving CF problems in a fixed difficulty range every day recently. From what I've heard, as a thumb rule you should solve X+200 (where X is your rating) difficulty problem, so I tried to solve problems rated from 2300 to 2400.
After solving a few problems, I discovered that I could solve about 90% of the problems I open. And it only costs me around 20 minutes to come up with an idea. So I think they might be too easy for me.
However, I feel that I'm usually still able to learn some new idea after I solved each problem. I enjoyed maths and construction problems more than I ever did before. Also, it's really tough for me to try 2500-2600 problems, as I can only solve 20% of them. I guess these are signs that their difficulty is probably OK.
So can anyone describe what situation I'm in and give me some help?
Randomly generate a number between 8 and 35, multiply it by 100, and that is the rating you practice for the day.
Damny, what happened to your contribution?
I was wondering the same thing. For some reason, it went from like 50 to -5 in just a week or two.
ratism is working so good that you dropped from expert to spec and you started losing your social credits incredibly fast. codeforces don't need to work on this because we like to be ratists.
the amount of contribution you get from a single blog/comment is halved every 6 months, and you seem to have a decently upvoted blog and a lot of small comments right around the 6 month mark.
Do some more contests. If you can solve ~90% of 2400 problems, you can get to red really easily.
Not necessarily if you are a slow solver like me. If his claim that he can get an idea in 20 minutes is true, then sure. However, I can solve 90% of 2400s, but it almost always takes at least an hour and on average 2 hours, not fast enough to solve in real contests.
Maybe (I don't really know you, so you can just ignore my opinion if you want) you are deceiving (prolly it's too harsh a word, but) yourself: take https://codeforces.me/submissions/HaccerKat/contest/1975 for instance, you can't solve a 2100 problem for 2h (I've had that experience too, I can't solve a 1300 for 1h). Maybe (again just a guess) you're solving 2400s from div2s, which are genuinely just way easier (at least for me), than div1 2400s.
Again, you can just ignore my opinion if you want, it's just some personal thoughts:)
Ok, it seems that Misuki has alr told my thoughts, so you are more than welcome to freely ignore this comment:)
This misses the point of the comment. There is a difference between in-contest solves and upsolves. Sometimes I would miss a problem in contest. Then, I would wait for my mind to clear up and be able to upsolve without editorials. This can easily happen if I was sleep deprivated (that did happen in the contest you mentioned) in contest (07:35 contests for me), which is something I am trying to fix.
The point is that for me, I often approach problems wrong the first time, even on easy problems. This would lead to very slow solves. But unlike some people, I can eventually puzzle things out, sometimes requiring breaks. For this reason, I am changing my practice strategy to something more time restricted to help with approaching problems right the first time.
And also I did take into account div. 1 2400s when writing the original comment.
Yes, there is a difference for sure. As for sleep deprive, contests here are 22:35 so it's a problem for both for us.
For "approaching problems wrong the first time", this happens frequently with me as well, I guess you should just do "time restricted solving" on 2400s for the time being or so?
You tell me master.
Maybe try to filer out div.2 problem? these are prone to be overrated compared to div1(+2) problems
your rating + 200 is actually quite accurate if done correctly (which i dont think you did) u can usually tell a hard 2400 problem from an easy one e.g. 2200-2300 div 2 D are usually harder than most 2500-2700 div 2 F also the higher the division the more overrated the problems become infact i doubt anybody whos 2300- is able to solve div1 2400 problem in less than an hour
Well, I can't really feel the difference between Div.2 and Div.1 problems in same difficulty.
Example: I just opened two 2400* problems, 1988E - Range Minimum Sum from Div.2 and 1924C - Fractal Origami from Div.1. I came up with the solution of the latter one almost instantly but I can't just do the former one in 40 minutes of thinking. Are the problems I chose too much biased or is there something wrong with your theory?
Beside from that example I feel other problems from Div.1 and Div.2 have almost the same actual difficulty for the same rating.
not so true look at the leaderboard for the div1 problem some high gms and even lgms like radewoosh didnt solve problem C compared to problem E which there are only afew gms who didnt get it in the div2
I have done the latter when I was a *1700, just "bingo!" and it's solved. In my opinion, given the same rating prediction, maybe problems from the higher division require more instinct rather than preliminary knowledge. This is because a standard problem at *2300 showcasing a data structure shouldn't work as a Div1C but can appear as a Div3G.
I think 1988E is just easier than 1942C
I do virtual contests with Codeforces Anytime (https://codeforces-anytime.firebaseapp.com/).
This is good, because the difficulty is gradual in contests, you get motivational boosts and practice with easier problems, and you always encounter a critical problem that takes you out of your comfort zone.
I've 1842 contest rating in Leetcode, and astounished by codeforces, started it recently. Currently I'm in Div 3 and only being able to solve 4 problems from each contest, almost solving 5th every time, but it's hard, man! :)
I used to do a bit of virtual contests before but now I realized I'm wasting time for racing on easy problems and most contests doesn't even have a problem within my narrow difficulty range.
I want to ask something ,could you please Mindeveloped help me and Um_nik , that i have no friends , but i heard that doing competative programming with freinds keep you motivated , is it true , or it is posiible to do cp alone to become a good programmer , beacause i have no friends. And could you please help me in , how to think after reading editorials , because sometimes it is very difficult to understand what is written in editorial.
what??
could you please tell me
means could you please me the right path
doing competative programming with freinds keep you motivated
yes
it is posiible to do cp alone to become a good programmer
yes, iff you have sufficient motivation and time to do it
And could you please help me in , how to think after reading editorials , because sometimes it is very difficult to understand what is written in editorial.
needing help reading editorial = problem too hard, try easier one
Are you basing yourself on cf's problems ratings or clist's? I personally think clist's one is more accurate.
Now about problem difficulty. I myself have spent some months experimenting with different problem difficulty ranges to practice. The conclusion I reached for myself is that I find both more enjoyable and more productive to practice in range that I can solve the problems by myself ~30-40% of the time.
I like this difficulty range because I can feel good about myself if I solve a problem by myself because they are usually somewhat hard, and if I am unable to solve them, they aren't super hard to the point of me having trouble with figuring out the solution, I usually just missed one or two key solution points.
I would suggest you to also do some experimentation like you said you were doing on the post, I believe everyone has their sweet spot so maybe copying what others do maybe won't be a good idea.