That's right.

Two or three years ago, I used to have 60+ contribution lmao

I also think that it's a good thread.

By the way, how does general intelligence affect thinking skills?

Well that's a very hard question, and I don't think anyone knows the answer, at least for codeforces thinking skill. There are surprisingly few studies on $$$IQ$$$ vs competitive math/programming skill.

First of all, most people don't know the definition of general intelligence, so just so we're on the same page, I will say it here. The general intelligence factor (or g factor) is basically a statistical factor (or underlying variable that can't be measured) that explains why some people will, on average, do better than others on pretty much every cognitive test in existence.

Like if you give someone a language test and they do really well on it, you'd expect them to do well on a math test too, even though, on the surface, there really isn't much in common between the language test and the math test. If you give someone a reaction time test and they do well (say, over-average) on it, well you can say that they are more likely than not to do well on pretty much every other mental test out there. You can argue that maybe doing well on tests isn't exactly the same as being 'smart', but the factor that accounts for this is called general intelligence.

Since codeforces can be considered a cognitive test, we can say that, given that two people know how to code but they don't have any competitive programming experience, the more intelligent one is more likely than not to do better than the less intelligent one. Okay, you probably could've figured that out by yourself. But, since general intelligence is so all-encompassing, this also applies to two people with the same amount of experience at any level of experience.

It's also important to note that just because general intelligence plays less of a role in higher-skilled contestants doesn't mean it plays less of a role in contestants with more experience.

I suspect that the skill gap between two people with differing $$$IQ$$$s widens rather than shrinks as the experience level goes up. Imagine someone who is like $$$70\,IQ$$$ — they might really have a tough time learning how to code, if they could learn at all. Their rating is basically gonna plateau very quickly at a really low level. Now imagine someone with $$$150+$$$ $$$IQ$$$ — they might reach $$$LGM$$$ with enough practice, but that would still take a lot of practice. The gap between the $$$70$$$-$$$IQ$$$er and the $$$150$$$-$$$IQ$$$er pretty much only increases, and I don't see why this would change for any pair of $$$IQ$$$ scores.

Also, if one's rating (output) vs time spent (input) could be a function, it would more or less be an increasing function but with a decreasing derivative, maybe like $$$y = IQ \cdot \sqrt{x}$$$. This would mean that, not only is someone with higher $$$IQ$$$ and the same level of experience gonna do better than you, but as you gain more and more experience (and they do too) it will take more and more additional time for you to close the gap, if they were to just stop practicing. So, to answer your question, I would say both. I would be surprised if it weren't both.

Since it's relevant, I'd like to also add this graph (scroll down a little bit to problems solved vs rating as a percentile) https://carnegiemellon.shorthandstories.com/competitive-programming-talent-vs-tenacity/index.html that kind of shows this widening, even with all of the interfering factors.

Btw, you say you're of average intelligence, but why don't you actually find out for yourself? I would really appreciate it if you tried a few subtests from https://cognitivemetrics.com/test/CORE this test (ideally, Quantitative Knowledge, Arithmetic, Figure Weights, Antonyms, Information: all 5 of those would yield a pretty accurate estimate of general intelligence (the items are good, but I'm not so sure about the norms, but they seem alright), but to be fair since I'm assuming English is not your first language, antonyms might be deflated for you).

You don't have to pay, unless the mods there made it so you actually have to pay now. I wouldn't be surprised if they did that, since some of them have done much worse things wrt scamming people. Everyone will eventually face judgement for what they have done. But the test seems alright. It will probably say you will have to pay but the charge will be $$$0.00$$$ and you just put in your email and get your score. Anyway, you don't have to take it but I think you should figure out if you are actually around $$$100\,IQ$$$ since you say it so much.

just save two variables minimum and maximum score for prefix i and the transitions for this are also easy

I practically did exactly that but I dont entirely get why the answer could be max(dp[n][0/1/2/3]) lol. I thought that it would be max(dp[n][0/2]), but it didnt work so I just changed that. But that makes me think the code would be vulnerable to hacks lol

Bro just see From: lol

You can first check whether the binary-search range exceeds $$$O(\log^2 n)$$$. If it does, you only need to perform the search $$$O(\sqrt{n}/\log n)$$$ times; otherwise, each binary search costs only $$$O(\log\log n)$$$. Thus it is $$$O(q\sqrt{n}\log\log n)$$$. You can use some other techniques to achieve $$$O(q\sqrt{n})$$$