CSP2020 summary

---

$CSP-S1$

This is the preliminary part, if there is nothing to write, just write it

I stopped for a week before the test to do the preliminary contest (although I mostly did the test), the simulation contest I did was actually not done very high, because many strange questions were not well understood, and some computer repairers were very engaged. Human mentality. Fortunately, the exam questions are pretty good, especially the computer repair questions are relatively simple, the latter is more inclined to programming ability, but also got a few questions right, which is to improve the program where there is a little more wrong, an unexpected good Score $95.5$。

The main reason is to pay attention to the questions in the reading program. The time complexity questions also need to consider different situations to get a good score. The following are some masking skills. I don't know how to perfect programs, but it feels like writing code myself.

---

$CSP-S2$

Travels $Part$

Because only one day of the test, there are no hotel adventures and refreshing battles, so write less.

I stopped classes for one month before the exam (I regret not going back to see the culture class in the middle, and I couldn’t do anything), and then the mock competition was basically fierce in the first two weeks and then tired and efficient. It should be very efficient in the future. Avoid long-term suspension of classes.

A week before the exam, I was lucky (dangerous), and I was frightened. On the day of the exam, I gathered a photo and took a photo together before entering the exam room. The test room environment is good (much better than the provincial competition), and after the test started, the invigilator also wrote the wrong password and made the entire test room come with a time penalty (no.

$T1$ title is very long, very troublesome, if it is normal exam Zhibuding skipped, but the normal force yourself to write the exam. Started to read the wrong data range and wrote a meeting$T$ Fei's code, only to see that the following data is so big, and then it took a long time to write it. When I wrote, I was panicked because it took more than an hour to write. After testing a large sample, I lost it first.

$T2\; is$ relatively simple, bit operations can be done quickly, but a$vector$ (I know that I don't need it until I finish the exam), the large sample is lost.

$T3$ looks like a routine but can’t do it. It’s Zhang$DAG$ clearly not merged segment tree. Then I found that you can split a multi-operation program into a bunch of operations and then add a multiplication at the end. This is a good job, so I did it once and did it once, and lost the large example.

$T4$ doesn't know anything, but there are multiple games and games. Then$yy$ got a$70$ points of practice seems to be on the first write, want to be able to optimize the combined fruits like after finished with a queue, but more sleepy and I think it will be very troublesome to check the Xiandiu.

$T1$ didn’t have time to write the match,$When\; T2$ checked, it was found that the maximum data was to be judged, so I changed it and lost it. $T3,T4$ seems to be fine.

When I came out, I was sleepy and went home after eating.

Later $Part$

Tested by Xinao Test Bank : 100+100+100+60=360
measured by Luogu data: 40+100+100+15=255
measured by data: 40+100+100+50=290
measured by Luogu official data Up: 90+100+100+60=360
Official test: 70+100+100+60=340

CCF Do I love you $Is\; it$

Summary $Part$

In general, the mistake was quite big (fortunately because of $CCF$ uses feet to make data)

First $T1$ was stuck. In fact, I still haven't found the wrong place. Anyway, it's not where everyone is wrong. Then$T4$ does not know where the problem is, but the method is correct. Anyway, ideologically there are$370$ , although a lot of points.

The main mistake is that the code power is not enough and there is no time to write the match, even in this time, there is no time, so the code power still needs to be improved for the complicated part (mainly because the inscription is less, go back and write more Some big simulations and big data structures).

Then the other aspects are pretty good, such as $T2,T3$ all have scores for thinking. There is$T4$ actually came up with a method in the examination room (although it is far from the correct solution), but the state is still not good, and the examination needs to be adjusted.

Then gone