Rewrite the Technological Landscape

Thinking of earning money from Koreans and Neon people to develop themselves, everyone was inexplicably excited, but thinking of the game engine requirements Meng Qian gave them, everyone felt Alexander.

To put it bluntly, the core of the game is the game engine. Many people call it the game engine. The prerequisite for developing a game is to develop or purchase a game engine. Of course, there are also free open source game engines that can be used.

As for whether games can be made without a game engine? In fact, it is also possible. Games in the 1970s and 1980s were made without a game engine. At that time, programmers were purely typing out a little bit of code.

But after 1992, everyone realized the benefits of the game engine. It is essentially a framework, including main modules such as renderer, physics system, collision detection system, sound effects, scripts, computer animation, artificial intelligence, network system, and scene management. .

Its biggest advantage is that the same game engine can be used to develop similar games in the future, which greatly saves costs.

However, it is very difficult to develop an excellent game engine, especially a 3D game engine, which is why most game companies in later generations would rather spend money to buy game engine licenses than develop their own, and the development is too tiring.

Several mainstream game engines can obtain considerable game engine licensing income every year. Since Dafeng Kechuang will continue to develop in the game field in the future, there is no reason not to lay this foundation. The origin of the game engine is 1992, and it has not It has been developed for less than 9 years. If Meng Qian does not grab this market, he will really be sorry for his rebirth.

The helpless thing is that the previous team did not have the ability to develop an excellent game engine, and Meng Qian himself still needs to learn. As I said before, when I was working on the engine of "The Legend of God", I wrote it while learning, but "The Legend of God" After all, the engine is relatively simple. After all, Meng Qian didn't rely on 2.5D, and only made a 2D game.

Coupled with the simple and rough gameplay and interface, this kind of game engine cannot be compared with those mainstream 3D game engines, which is why there are so many legendary engines circulating in the future, but not so many mainstream 3D game engines circulating on the Internet.

The difficulty of development is not at the same level. At the beginning, Meng Qian, the engine of "The Legend of God", built a big framework in about 2 months. Later, he and the company team continued to develop for more than 1 month on the basis of those 2 months. It took about four months to dry out.

Meng Qian later went to Koryo to learn about it. The development cycle of Koryo online games in this era was generally only about 6 months, but the overall quality of Koryo online games was indeed better than that of "Shen Biography".

"The Legend of God" has always been rated by players as inferior to Koryo online games in terms of game quality. There is no way, the level of the team is there, and Meng Qian is not a professional game maker.

Fortunately, there were no competitors at that time, and the quality was almost inferior, and Dafeng Kechuang has been optimizing.

However, games of the quality of "The Legend of God" and "Legend" can still catch the eyes of game players. After the rapid improvement of the game quality in the future, the requirements for the game engine will become higher and higher, especially to achieve At the level of a major game company, their development cycle usually starts at 3 years.

So, after the level of the team started to rise and some more experienced programmers were recruited, it was time to start the development of the game engine.

What Meng Qian wants to tell the employees today is a thing in the game engine, artificial intelligence

The game cannot do without simple AI, especially whether it is a fighting game or the Chinese ancient myth version of DOTA, it cannot do without man-machine battle.

Although Meng Qian needs to learn a lot in the development of game engines, artificial intelligence is his major.

"After rebirth, I can finally tell people about artificial intelligence." Meng Qian felt like the seal had been slightly lifted.

In 2001, there was no depth algorithm. The depth algorithm was proposed by Geoffrey Hinton, a professor at the University of Toronto in Maple Leaf Country in 2006. However, in this life, the name of the proposer of the depth algorithm is destined to become Meng Qian

For Meng Qian in this life, other fields can fail, except in the field of artificial intelligence. This is a matter of the dignity of an algorithm architect.

In China in 2001, there were not many mature programmers, not to mention mature algorithm engineers, so Meng Qian planned to start training at Dafeng Science and Technology Innovation. them.

How they go in the future depends on themselves.

"Let's first understand the history of artificial intelligence. This matter begins with the Nobel Prize in Medicine in 1981. The winner discovered that the information processing of the human visual system is hierarchical, which has inspired new thinking about the nervous system.

In 1957, humans first proposed a machine that can simulate human perception, called the perceptron. The algorithm principle of the perceptron is to calculate the appropriate weight coefficients (w1, w2, ..., wn), so that the system can evaluate a specific sample (x1, x2,...,xn) can produce expected value d.

But the perceptron can only solve linear problems, but cannot solve nonlinear problems. After that, there is the rise of BP neural network.

BP neural network is a multi-layer feed-forward neural network. Its first stage is the forward propagation of the signal, from the input layer through the hidden layer, and finally reaches the output layer; the second stage is the backpropagation of the error, from From the output layer to the hidden layer, and finally to the input layer, adjust the weights and biases from the hidden layer to the output layer in turn, and the weights and biases from the input layer to the hidden layer.

The transfer function used by the BP network is a nonlinear transformation function, the Sigmoid function, and the unipolar S-type function is defined as follows: f(x)=1/1+ex, and the unipolar S-type function is defined as follows: f(x)=1ex/ 1+ex.

I will give you a simple operation of the code implementation of the BP algorithm.

First define a BP neural network class, set network-related parameters, instantiate the neural network, build a BP network with 3-dimensional output, 1-dimensional output, and 3 hidden layers. When initializing the BP neural network, start initialization The weight, weight momentum, error initial value of each layer of network nodes, and the introduction of learning and training data”

When it comes to artificial intelligence, Meng Qian is very familiar with it, and he talked directly to the employees for 2 hours.

Since then, the game project team has started a crazy game engine development road.

While the technical department is working hard, another thing has to be put on the agenda, and that is the animation and film and television process starting from the game as an IP.

But in this matter, Meng Qian encountered difficulties.

In terms of movies, starting from the ten-film plan in 1995, a large number of American movies flooded into the Chinese market and began to suppress Chinese movies. In terms of animation, a series of neon such as "Dragon Ball", "Slam Dunk", "Detective Conan" and so on Manga already has a large number of fans in China, and it won't be long before the three major migrant workers Man Hokage Pirate Reaper will become popular.

Even Goryeo released "Blue Life and Death" last year, and the invasion of Goryeo dramas is imminent.

Now that the entire China is in a backward situation in the field of film and television animation, Meng Qian wants to use "The Legend of God" as an IP to gradually transform into film and television and animation. How about resisting pressure from three parties?

Meng Qian found it difficult.

Therefore, Meng Qian chose a more conservative approach when it comes to cultural output.

The four major cultural phenomena in the world in later generations are American movies, neon animations, Korean dramas, and one more, of course, is Huaxia.com.

Now Huaxia games are far from leading the world. Since games can't drive the development of the entire cultural industry, Meng Qian added the important weight of online articles.

It may be more reliable to build a foundation for Chinese culture output based on online literature + games.

So, on the morning of March 28, 2001, Meng Qian's office welcomed four representatives. Behind them was an immature website alliance.

Thank you Wandering Still 007 for your reward!