16
Wang Fei has been having a great time lately.
During the day, Wang Xu would go to school, and Wang Fei would have free time, occasionally running to Lao Na's house to bother her, or going back to Wang Xu's home to play mahjong with Aunt Jiang and the others.
Mahjong was already very popular in the capital city during this era, but the rules were very simple, not as complex as Hong Kong mahjong.
Most of the time, it was just a regular win, a self-draw doubled, a flush or seven pairs also doubled, a discard win was paid individually, and a self-draw was paid by everyone. It was simple and fast, with a game lasting only a few minutes.
The stakes weren't high, just playing for one or two dimes, or even a five-cent base. Wang Fei quickly got the hang of it after Aunt Jiang called her to fill in when they were short a player, and then she became completely addicted to mahjong.
If it weren't for the fact that she still practiced her cultivation every day and had mastered the songs he gave her, Wang Xu would have wanted to enroll her in a class to break her addiction. Since it wasn't delaying any important matters, he figured she could play if she enjoyed it; she wasn't a child anymore.
However, every evening, Wang Fei would return to the small courtyard on time, bringing the dishes Aunt Jiang had cooked, and spend time with Wang Xu, just the two of them. This made Wang Xu very happy, giving him the sense of happiness one feels when coming home from work and seeing his wife.
Wang Xu didn't know that many people were thinking about him, nor did he know that more and more people were starting to pay attention to him. Even if he knew, he wouldn't care, because he was very busy right now.
He was currently busy repairing his recording equipment. He had delayed doing so before, mainly due to a lack of necessary equipment, but now that he was back at school, he could borrow the school's laboratory equipment.
Equipment like electric welders and multimeters were easy to acquire; after all, you could buy them if you had money. However, professional laboratory equipment such as oscilloscopes, signal generators, and spectrum analyzers were impossible to buy right now, even if you had the money.
Because imported ones required waiting, and domestically produced ones required even more waiting.
Imported ones required waiting for foreign exchange quotas, ordering, manufacturing, shipping, and customs clearance, which could take two or three months, or even half a year.
As for domestically produced ones, the electronics instrument factories hadn't undergone reform yet, and production was limited. If you weren't a state-owned enterprise or a university, you just had to wait. When it was your turn depended on their mood.
Fortunately, Shui Mu University's laboratories could be rented, as long as you were an internal member. For students, graduate students needed to apply with their supervisors, while undergraduates needed to apply with their department, which wasn't too much trouble.
It was just the beginning of the semester these two days, so there were fewer people doing experiments. He had to borrow it quickly, as it wouldn't be possible once more people came in a few days.
After some twists and turns, Wang Xu finally obtained a week's usage of an electrical engineering laboratory. Although it cost two thousand yuan in rent, it was still very cost-effective, and the school would even assign a lab teacher to help supervise.
Among undergraduates, he was the only one with such a privilege.
And this privilege was exchanged for a paper he wrote, a paper on an artificial intelligence algorithm for convolutional neural networks. In his previous life, it was published in 1998, and it didn't make much of a splash initially.
It wasn't until 2010, when convolutional neural networks truly exploded in popularity, that it was gradually unearthed and became a legendary paper. Of course, this also had a deep connection with the original author's 20 years of dedicated work in this field; many of his later papers were based on this one.
I won't go into the specific content, as it's not very relevant to entertainment. Copying song lyrics is one thing, but copying a paper might actually get the author killed by readers.
This was one of the must-read papers for anyone researching artificial intelligence in his previous life. It wasn't overly difficult, yet it was ingeniously designed, making it very suitable for Wang Xu, who only had an undergraduate degree. Publishing it as a Q2 SCI paper was absolutely no problem.
The original was a Q1 paper, but the author was a United States Ph.D. and a renowned laboratory head. Wang Xu, on the other hand, was just a small fry; to even publish in Q2, he needed to have a departmental supervisor listed as a corresponding author.
But even so, in Shui Mu in 1991, it was a big deal. You should know that this was an era where even Ph.D. students only needed one SCI paper to graduate, so you can imagine the weight such a paper carried in the hearts of the department's teachers.
(Characters like teachers and classmates are completely unrelated to reality, as they are not famous people, and using their names would not be recognized and could easily cause trouble.)
Universities at that time weren't as competitive as they would become later, so when an undergraduate suddenly came forward saying he had written an algorithm paper, and this student was only in his second year, with his major courses not even having started yet, the teachers were all very surprised.
However, they still connected Wang Xu with a returned doctor named Luo Congwen as his supervisor, asking Wang Xu to let him review the paper.
Yes, that's right, undergraduates at Shui Mu could also have supervisors. When there were special needs, a supervisor could be assigned, which also benefited the supervisor, as it counted towards their workload, especially for young teachers who hadn't started taking on graduate students yet.
The supervisor was very responsible. After reading the paper, he found it very novel, considering it an extension of a branch of artificial intelligence with a certain foresight—a very good paper. So, he helped him revise the format and submitted it.
He even asked his former supervisor in the United States for help, to see if the process could be expedited. This was his first time being a corresponding author for a publication, and he wanted to get off to a good start. If it went smoothly, it would be beneficial for applying to recruit graduate students and for promotion to associate professor next year.
As for the laboratory, well, this was both related and unrelated to the paper.
It was related because of that paper; Wang Xu successfully became a celebrity in the department, so renting a laboratory, especially one that didn't have a task yet, was no longer a problem.
It was unrelated because the purpose of renting the laboratory could be said to have no connection whatsoever with the paper; what good would an electrical engineering laboratory do for a paper on artificial intelligence?
Wang Xu's progress was very fast; this time, it was just about repair and incidentally adjusting the signal-to-noise ratio.
Recording equipment can be seen as a whole system, with sound going in and a finished product coming out. Therefore, systems engineering is a crucial technique when optimizing recording studio equipment.
Friends who play Hi-Fi might have some experience: even though they use all good components, the final effect is always underwhelming, sometimes even worse than the finished product put together by some skilled enthusiasts with one-tenth the budget.
This is the issue of system optimization, and it's also why Wang Xu needed advanced testing equipment like oscilloscopes.
Simply put, you use a standard wave generator to produce a standard waveform, then after it passes through the entire system, you look at the final waveform obtained. That overall deformation is the total error of your system.
Each piece of equipment along the way will cause some changes in the waveform, but if these changes cancel each other out, it might form a complementary result, where both intermediate results are wrong, but the final result is correct.
Of course, music won't be as precise as a standard wave, but this compensatory mechanism is similar. Wang Xu's goal was to use these professional devices, along with his mechanical-like ears, to repair this stack of equipment and adjust it to its optimal state.
As for the signal-to-noise ratio, that was more troublesome. Wang Xu could design some special circuits to filter out noise, but this filtering would inevitably affect some normal waves as well. Therefore, the design and manufacturing of recording and audio equipment often felt very mystical.
So, Wang Xu spent two full days designing several sets of filtering and amplification devices for different positions, and then connected them to the equipment. Only then did he achieve a result that he was basically satisfied with.
This was considered usable. Although it couldn't compare to the industry's top recording studios, it could be considered outstanding in China. But in Wang Xu's eyes and ears, this set of equipment was merely usable, because he knew of better ones.
