Always defeated by ad-hocs, can't solve constructives, bad performance, low iq.
What can I do.
Always defeated by ad-hocs, can't solve constructives, bad performance, low iq.
What can I do.
| # | User | Rating |
|---|---|---|
| 1 | jiangly | 3810 |
| 2 | Benq | 3676 |
| 3 | Kevin114514 | 3655 |
| 4 | maroonrk | 3463 |
| 5 | strapple | 3447 |
| 6 | Um_nik | 3387 |
| 7 | heuristica | 3322 |
| 8 | turmax | 3317 |
| 9 | tourist | 3307 |
| 10 | jiangbowen | 3291 |
| # | User | Contrib. |
|---|---|---|
| 1 | Qingyu | 156 |
| 2 | nik_exists | 150 |
| 2 | maspy | 150 |
| 4 | Um_nik | 143 |
| 5 | Errichto | 139 |
| 6 | adamant | 137 |
| 7 | AmShZ | 135 |
| 8 | maroonrk | 133 |
| 9 | BledDest | 132 |
| 10 | qwexd | 129 |
I have the same problem, and I want to ask you the same question but for reaching Master. What is your advice.
More hard trainings and a simple usage of brain.
The reality for every single rank XD.
I have the same problem, and I want to ask you the same question but for reaching Candidate Master. What is your advice.
Get good at the basics. I did the entire CP4 books 1 and 2 and then joined codeforces. I was able to get to CM in relatively short time (mind you I didnt just skimmed through the book, I would do all starred questions. I remember I did like 6/9 starred questions a day depending on the difficulty of the section and was able to finish both books in about 4 months).
43<<<
deleted
Depends on previous experience with competitive programming. I remember it took me like 4 months of crazy hard work (starting from little experience), so approximate it by a fulltime job:
40hrs * 4 * 4 = 640 hrs of hardwork
That said it is much more pleasant than doing problems on codeforces, because you learn a topic
Xand then solve problems related toX, so you feel like you are improving. I only notice that on some problems on codeforces unfortunately ;(. Probably because most problems are mostly ad-hoc nowadaysno at all an answer to your question but after some introspection I have come to realize that there are no true “ad-hoc” problems. Every adhoc problem I have solved myself has used the same old bag of meta-thinking strategies used in other similar problems. So if you have the motivation and time, the answer to how to get to any rating at all really seems to be “solve more problems and fit the pattern into your intuition for each problem you solve” during training and “actively try to fit patterns/meta-patterns of the problem you see to your library of solved problems”.
Personally though I feel my progress is limited by how accurately I can debug and implement my solutions and not related to my ability to mind-solve a problem, so I feel that to get higher than a 2100 level one needs to work on accurate and fast implementation as well for beyond this level I cant seem to rely on just being able to solve a problem in my head alone. But thats just my opinion derived from a sample size of 1 so take as much stock into it as you wish.
But i'm completely opposite... have problems solving problems but once I mind-solved, can implement it in 20 mins...
Great then can I ask if you plan your implementation every-time before starting to code? Or do you just go with the vibes and figure it out as you code?
Also, before implementing your idea, do you verify that your idea is indeed correct, and for instance figure out the precise transitions of the DP? I have the problem of getting to the idea really quickly but then being lazy about the details/missing them/rewriting code multiple times with bugs.
Yes i think about and write down all the details before i started coding so in most cases i can write my code very quickly without hardly any bugs. (But sometimes that's the reason why i solve problems slow because of thinking too much :D)
Man I so agree with the second point. I suck at implementing ideas.
Any tips on how to implement fast? At this point I am thinking of changing my practice regiment from 1-2 1800-2100 Qs to 3 1400-1700 qs but 20-30mins daily each so I am forced to increase my speed.
How to reach Master?
Always defeated by ad-hocs, can't solve constructives, bad performance, low iq.
What can I do.
xd
How to reach Expert?
Always defeated by ad-hocs, can't solve constructives, bad performance, low iq(94 by iqtestacademy.com)
What can I do?
well first of all you should find a better IQ test
How will it help me to reach expert? To be honest, I believe it might become quantitatively less but not significantly less. Even with a bunch of tests, average will be quantitatively same.
it won't but they're kinda fun. like this one
It was fun at the end but I got TLE. Solved 20. Got a raw: 17 and scaled: 13. What is my IQ then?
Since the scaled scores are set to have a mean of $$$10$$$ and an sd of $$$3$$$, you can convert to $$$IQ$$$ with this formula: $$$IQ = 5 \cdot (score - 10) + 100$$$.
That is 115 but this test has low g loading (~0.60) i.e. Not a Good correlation with g or simply not a complete and good estimation of your IQ.
wait wait wait Are you active on r/ct or the discord server?
I used to visit the subreddit pretty often, but it went downhill ever since the mods made that deal with the Indian scammer. There are still good releases every so often, so I visit every few weeks or so. How about you?
Are you talking about Polar? When I saw your name and profile picture here a month ago, I remembered seeing them on the Giftedness server.
Well I think it was a group effort by the mods. They made a deal with some Indian scammer to advertise his shitty "Jordan Peterson" IQ test: realiq.online. It is just the worst test you can make. And most of the items aren't even original. You can still find it in the resources tab I think. The worst part is that you have to pay for the results, so they undoubtedly scammed some newcomers out of their money. And if you post about it, you get banned. Truly scummy behavior on the mods' part. They must be getting paid for this effort. Or one of them has done some bad shit and the guy is blackmailing them. But that is all speculation. I don't know why they are advertising such a bad test.
But you must be thinking of someone else because I've never been to the discord server, except for like a few minutes once to find a test. Honestly, It didn't seem like a very nice place.
got a raw 22 and a scaled 18
Unfortunately, IQ is hard to increase. So if IQ were your problem, then you'd better start doing lsd while listening to tool like your life depended on it. But it looks like you got pretty close, so IQ might not be your problem.
Just curious, what IQ do you think corresponds to each rating?
not him but assume median effort and practice of ~2 year.
he/she rating will converges to as follows
Ain't no way average world IQ is blue LOL
you mean too high or too low?
I think too low. You are underestimating how hard it is to get to blue
It's true, but I believe that anyone who puts in the effort I described 18 months of training or 1,000 solves will reach blue. Some talented people might hit orange or even LGM (e.g., jingly), while others might struggle at green forever. But if someone has an IQ of 100, then blue is pretty much guaranteed. but not everyone can push that amount of effort (due to school work or jobs) that's the thing I think
Like me
well, the statistics are available online. you can check them.
I am telling you this, because it is impossible to get "pure" and "clean" data, because we are all living beings (otherwise you are a cheater :p)
you can join stuff like errichto server, and use ;plot centile to plot the entire codeforces distribution of ratings (i assume its just active users). tldr described in text: x axis is normal (not logarithmic), it's rating, y axis is also normal, it's the number of people below/equal; between ratings 350-1600 it is roughly a line; 1600 is roughly 90 percentile; 1900 is roughly 96.5 percentile; and 2100 is roughly 99 percentile.
how to interpret this correctly? blue is at least top 10% which would mean iq 120 (because of normal distribution). + i'd assume that codeforces has a higher concentration of intellegent people... (which would mean blue -> 120 < iq), but this is (highly) debatable.
If I have an IQ of 94 what am I??
Also, (almost) no one has >150 IQ.
So here is how it goes https://imgur.com/uLZuLGV
Remember that it isn't gonna be $$$1:1$$$ accurate, because $$$IQ$$$ and cf rating aren't perfectly correlated, but this should be like more or less how it goes.
Chad advice
Reaching GM worths just nothing.
fft_ntt: 2408->2361 Became IM :(
Reaching IM worths just nothing for me :((
How to reach Newbie?
Always defeated by ad-hocs, can't solve constructives, bad performance, low iq.
What can I do.
Have you tried breaking ur wifi?
Let me participate a contest for you.
I can assure you that you will be negatively rated after $$$1$$$ contest.
How to reach GM?
Always defeat ad-hocs, can solve constructives, good performance, high iq.
What can I do.
look at contests like CFR1001. diff of D is 2100 and the next task is *2900.
in contest 'Goodbye 2024', I only solve A~E(diff up to *1900), and I have a positive delta.
so you only need to solve easy tasks fast, you can win.
also you can do some vp to know some easy tricks, which would help you to solve template problems easier.
Same issue but my iq's lower than yours, and not only can I not solve ad-hocs, I can't solve anything else really...
Try AtCoder, especially ARC/AGC contains tons of ad-hoc problems focus on thinking and puzzle solving skill
Rename yourself to GM and reach 4000 rating. Just that simple.
How to reach Master
Be more confident. "low iq" mentality won't result in anything great.
but LGM when pretty boi
Fair
it can be empirically proven red is impossible under current circumstances
What do you mean by that?
I've been a GM a few times, so I think I have something to say
Div1 competitions are very different from Div2
Speed is not usually the deciding factor of performance, solving the problem is
Often there is exactly one problem that fits your rating range perfectly and you either solve it and gain rating, or fail and lose rating
Codeforces nowadays focuses on mostly observation + simple algorithm/data structure problems, and you're just going to have to live with it, that is your target "problem tag"
My advice here is to upsolve problems that are way out of your rating range. Problem ratings after 2400 are very skewed and random, some ideas from a solution of a 3000 problem might often occur in problems in your actual range of 2200-2600. Spend more time on harder problems to gain observation you need to solve simpler problems. That's it.
Practice makes perfect :)
Yes now I trains in rating range [2600,3200]，but ty for ur advice
Wrong. If you solve 3 problems in Div.1 (usual difficulty), your rank can fluctuate between 60 to 300, the former one is about 2700 performance, while the latter is only about 2200.
The average is 2450, which is completely fine.
Tackle <=2600diff of past Div1s and understand how to deal with medium-hard Div1 problem on CF. If you try 2-3 hours for single problem, after that, looking editorial is ok, but please not stop at understanding solution, and make an AC submission. If you solve enough number of problems, just participate many contests and wait for upward swing.
Ok, I'm GM now (not on this account (yet)). How do I become tyristik?
How to reach Specialist?
Always defeated by ad-hocs, can't solve constructives, bad performance, low iq.
What can I do
:p
solve more problems