Computing Lab!

Ye Fei sat in front of the computer and quickly typed on the keyboard.

He is now using computers and mathematics to develop a new algorithm.

This algorithm, Ye Fei called it "graph computation method", the full name is "Turing computability algorithm".

Graph computing method mainly uses most of the mathematical knowledge, and a small part is computer knowledge.

After starting to cooperate with Professor Zhou Kang, Ye Fei also went to learn computer knowledge.

Although not as strong as Professor Zhou, it was enough for him to develop a graph calculation method.

Graph computation uses computability in set theory, topology on sets, pre-ordering, and partial ordering.

After a long time, Ye Fei stopped his movements, took a long breath, and said, "It's finally developed." ”

"Let's start testing!"

After that, he installed the algorithm on ABTest and tested it.

This is a software that specializes in computer algorithm testing.

There are seven steps in algorithm testing, which are interface joint commissioning, degradation scheme and performance optimization, recommendation effect testing, functional testing, algorithm data, requirement rules, and algorithm model.

"Wait!" Ye Fei saw that the test would not end in a short time, at least for an hour.

After an hour, the test is over!

Ye Fei looked at the test results and showed Success, indicating success.

He laughed, "Good, the test was successful." ”

"Ye Fei!" Professor Zhou ran in sloppily and said excitedly: "Mine is done, we can start the next step of Turing equivalence." ”

Ye Fei nodded and said, "Okay, wait for me for a while." ”

"What are you doing?" Professor Zhou asked.

"Oh, I took advantage of nothing to do and made an algorithm with Turing equivalent research."

"Algorithm?" Professor Zhou asked, "Computer algorithms?" ”

"yes!" Ye Fei nodded.

"I need to do algorithms!" Professor Zhou was almost out of self-confidence after being hit by Ye Fei during this time, and seeing that Ye Fei needed himself, his self-confidence returned.

"I can help you, let's sum up how to do this algorithm."

"No need." Ye Fei shook his head: "I have already developed it." ”

"R&D?" Professor Zhou was surprised: "How can you develop an algorithm without me?" You don't know how to use a computer. ”

"My algorithms have very little knowledge of using computers, most of them are math." Ye Feidao: "The computer part is also very simple, and the computer knowledge at the undergraduate level is enough, so I can do it myself." ”

"Impossible!" Professor Zhou shook his head and said, "That's an algorithm, although a lot of mathematical knowledge is used, but mathematical knowledge needs to be converted into computer language. ”

"What do you do about that?"

Computers are not human beings, and anything that enters a computer must be translated into computer language.

This lets the computer know what you're typing.

Something that is equivalent to input requires a translation mechanism.

So, in algorithms, a lot of mathematical knowledge and a lot of computer knowledge are mixed together.

"The algorithm you developed is not very good, is it?"

In Professor Zhou's view, the algorithms developed by computer knowledge at the undergraduate level are definitely not very good, and they can run a small game at most.

"The performance is not bad!" Ye Fei said: "The number of floating-point operations has reached five trillion times. ”

"How much?" Professor Zhou was surprised: "Five trillion times? Just you? ”

The concept of 5 trillion floating-point operations is to let the computer run 5 trillion times a second if the hardware can afford it.

Home computers, the average speed is between 20 billion and 100 billion.

Xia Guo's fastest supercomputer, Shenwei. Taihu Light, the sixth largest supercomputing in the world, has 1.254 billion floating-point operations.

This is equivalent to 50 times the fastest speed of a home computer and 1/250,080 speed of the light of Taihu Lake.

But don't forget, this is the world's sixth-fastest supercomputer, which is the crystallization of the wisdom of countless scientists in Xia Kingdom, and it took more than ten years to develop.

And Ye Fei only took ten days to develop an algorithm with five trillion floating-point operations.

This is outrageous in anyone's opinion.

"That's right!" Ye Fei nodded and said, "If you don't believe it, you can test it yourself." ”

Of course, Professor Zhou didn't believe it, and he personally installed the algorithm on ABTest to test it.

An hour later, Professor Zhou looked at the data tested in the algorithm software in amazement.

The average number of floating-point operations is 5 trillion and the highest is 6 trillion.

He looked at Ye Fei in shock and said, "Ye Fei, you are a monster!" ”

"Not only does your algorithm have 5 trillion floating-point operations, but even the size is only about 100Mb."

The number of floating-point operations reached 5 trillion times, which is very common in Professor Zhou's opinion.

Many large companies have long developed the number of floating-point operations of the algorithm of the home version of the computer program to ten trillion, tens of trillions, or even trillions of times.

It's just that the technology of the hardware at this stage cannot support such a high algorithm speed, so it is not promoted.

What really surprised Professor Zhou was that the algorithm developed by Ye Fei was only 100Mb in size.

That's pretty amazing.

This means that the number of floating-point operations is higher and the size is much larger.

It's the same size, and the number of floating-point operations is much smaller.

Professor Zhou immediately thought of the scope of application of the algorithm.

Home computers, unmanned cars, robots, ......

Professor Zhou's eyes lit up and said: "Ye Fei, don't mix in the mathematics circle, come to my computer circle, with your talent, you will quickly become famous in the computer circle." ”

"Forget it!" Ye Fei shook his head and said, "I don't know computers, I can only do mathematics." ”

"Yes, you said that you use a lot of mathematics for this algorithm, and very little for computer knowledge." Professor Zhou said in disbelief: "Is that really the case? ”

"You don't know if you look at the backstage for yourself."

Professor Zhou opened the background of the algorithm to check, and after a while, he exclaimed, "How can this be? ”

"You're going to pack all your math knowledge into one package so that the algorithm can run?"

Algorithms use a lot of math, but they're all spread out and can't be packed into a single package.

A large amount of mathematical knowledge requires a large amount of computer knowledge, and the computer knowledge packages mathematics into packages one by one, translates mathematical knowledge, and connects packages between packages.

And now, Ye Fei has packed all his mathematical knowledge into one bag.

This is equivalent to translating all the mathematical knowledge together, and there is no need to connect packages.

Because there is only one package.

However, this is contrary to the common sense of algorithms, and even to the common sense of computer programming.

Ye Fei said: "That's how it works, in my opinion, numbers are the most primitive language in this world." ”

"It runs through humans, machines, races, and doesn't need to be translated."

"Moreover, numbers have a soul, and they are smarter than humans."

"They don't need computer language to guide them to what they're going to do, they know what they're going to do."

"In that case, there is no need for translations and a lot of packs."

Ye Fei's algorithm can indeed work, but Professor Zhou is puzzled by how Ye Fei did it.

If it is really developed like Ye Fei's algorithm, a lot of code will be omitted.

First of all, there is no translation mechanism, and there are few packages, and there is only one.

Anyone who knows a little programming knowledge understands that the less code, the smaller the code size.

Small means that the running path is few and short.

If the mathematical logic is strong, then the algorithm runs very fast.

Professor Zhou suddenly found that he didn't know anything about computers, and he had learned all the computers he had learned from dogs for decades.

After watching it for a long time, he still didn't understand how Ye Fei did it.

He shook his head regretfully.

Professor Zhou asked, "What are you going to do with this algorithm?" ”

"What can I do?" Ye Fei asked.

"You patent it, lease it out, or sell it."

"How much money can I make a year by renting it out? How much can it sell for? ”

"Rent out at least 2 million a year, and sell it for more than 10 million, only a lot more."

"So much?" Ye Fei said in surprise.

I've heard people say in the past that the computer industry is very profitable.

Now that he is in trust, it is indeed very profitable.

I have worked hard for half a year, and I have earned less than a million from school.

I didn't expect that the algorithm I developed when I was bored would earn more in a year than what he earned from school just by renting it out.

Ye Fei said excitedly: "I don't read much, don't fool me, I will really believe it." ”

"Hehe......" Professor Zhou smiled: "If you can't make so much money by then, I will make up as much as you want." ”

Ye Fei took a deep breath, suppressed his excitement, and said, "Then sell it, rent it out, it will take five years to earn 10 million." ”

"Are you sure?" Professor Zhou said: "Although it is only 2 million a year when you rent it out, the money you make in five years is equivalent to the money you can sell. ”

"If you make double the money you sell in ten years."

"I think it's better to rent it out."

"And, that's how all companies with algorithms do."

"Almost none of the algorithms are sold."

"Even if there were, the performance of those algorithms is very poor, far worse than yours."

Ye Fei smiled: "This algorithm was just developed by me in ten days, if I spend more time, I can develop a better algorithm." ”

Professor Zhou suddenly felt so heartbroken.

Indeed, this was developed by Ye Fei in more than ten days, and he spent more time to develop a better algorithm.

In this way, it seems that it is more cost-effective to sell.

Professor Zhou said: "Then let's talk about selling algorithms!" ”

"I'm not good at selling things, but I know people who are good at it, and I can find someone to sell it for you."

Professor Zhou said and left.

He won't take Ye Fei's benefits, Ye Fei came to his laboratory to help, if he takes benefits, let Professor Lu think of him.

The most important thing is that he is very optimistic about Ye Fei.

He speaks straight and loves to offend people, but it also depends on who he is, and he is very happy to be close to talents.

For a genius like Ye Fei, he is more willing to get close.