Zhao Ye thought about imitating the keyboards of later generations? But there is no such thing now. You can play with it as you like.
However, he soon compromised and decided to design it according to the keyboards of later generations. It could be slightly adjusted, but the position of the alphabetical keyboard could not be moved.
Why?
Because he is used to it! He can't design a keyboard that would be awkward for him to use! As for whether it is awkward for others, it has nothing to do with him. Anyway, they have never seen it, let alone used it, so there is no such awkwardness question.
While thinking about it, Zhao Ye didn't stop until he had labeled the letter keys, numeric keys and some basic function keys. Then he said: "That's about all I can think of now. In the future, Just modify it as the application is added or subtracted.”
"This, how do you use this?" Professor Zhang looked at this graphic and felt so weird.
"Of course it's for input. Don't worry about this first. What you need to care about is encoding English letters, Chinese characters, Chinese and English symbols, and mathematical symbols. This is the focus of your work."
"coding?"
"Yes, coding! Let me tell you, our calculator is used for calculations. It has its own display device. So naturally, we have to convert the binary numbers in the calculator circuit and memory into It can be used to show numbers, symbols, Chinese characters, letters, etc. to others.”
"Please tell me in detail." Professor Zhang said, pulling out his pen and preparing to record it.
"First of all, this code is divided into outer code, inner code, and glyph code.
Let’s talk about this foreign code first. As the name suggests, it is a keyboard symbol that we use to input Chinese characters, numbers, symbols, and letters into the calculator. For example, we can’t make each Chinese character into a button, right? We need to use special methods to achieve our needs. I have some ideas about this. Let’s go into details later. I want to use a way to split Chinese characters to achieve it.
Secondly, there is the internal code. All Chinese characters, numbers, letters, and symbols are, in the final analysis, just a pattern. If we want to realize this pattern, we use dot matrix display technology. How to put it? In an area, use To draw a figure one by one, each figure must correspond to a fixed binary array. When we display, we only need to tell this array to the control circuit of the display screen, and it can draw based on this array. The corresponding dot matrix comes. This array is the internal code.
Finally, there is the glyph code. Just like what I just said, we told the display screen an array, and it will draw a graphic. So how to draw this graphic? This requires us to write a glyph code in advance, which is to multiply it by ten Within the ten area, which points should be lit and which points should not be lit, so that the corresponding characters can be displayed."
"Is this possible? Not to mention letters and symbols, which are counted after all, but we have one hundred thousand Chinese characters. Do we need to encode each one? How many storage units are needed?"
"Now we have just started to do this, we don't need to do that. Now in our design, one unit has been expanded to eight storage units. I named each unit a byte. A byte is an eight-bit binary number, such as We use up to two bytes to store a Chinese character or symbol. In this way, we can have 65536 codes available, which is enough."
This is of course enough. Later generations have always used it this way, and there is no shortage of it.
"Of course, you don't need to do so many at the beginning. Just get in the commonly used characters first. For example, start with three or two thousand, but the letters and symbols should be more comprehensive." Zhao Ye then added.
"Is this similar to Morse code?" Professor Zhang asked.
"Eh? Don't tell me, it's true. Research it and see if you can use it."
"Okay, I'll start working on this as soon as possible. I can ask some experts in writing."
"That's right, that's right, we just need to work together as a team and just do the connection work ourselves."
Of course Zhao Ye knows that this cannot be done by Professor Zhang alone. The coding is actually very simple. Now that they are the first to do this, they are the standard!
Correspondingly, the glyph code is the difficult part. It is very annoying to design one by one!
After sending this person away, Zhao Ye hurriedly went into his lounge, fearing that he would encounter Professor Ci and the others. Although their work was the core, Zhao Ye had no plans to start "making up lessons" for them now. At present,
Generally speaking, their work fell into the category of IC design in later generations. But now they consider things that are much more complicated than those architects in later generations.
After all, the industrial chain has not yet been formed at this time, and everything from design to processing requires their attention and design.
How can this be eaten in one bite?
Zhao Ye has already "left homework" for them. He must first understand his previous design thoroughly before talking about it. This cannot be completed in a day or two.
But this is also the basis. If you can't even "look at the pictures and speak", then how can you design it yourself?
Zhao Ye chose the number not so that he could sleep, but because he had to go to space and think about registers.
In the entire processor, this thing is the most important thing. Zhao Ye feels that it is also the most complex thing among the integrated circuits of the processor.
In his design, the processor is eight-bit arithmetic. Simply put, the arithmetic circuit must read and process eight-bit binary data at a time.
Then the corresponding register is at least eight bits, and in order to store the data during operation, there is an essential instruction set, as well as addresses, pointers, etc. Zhao Ye thinks that at least 12 registers are enough.
These 12 registers can only solve the most basic operating needs. If you want to improve the function and level of the processor, the registers must also be increased accordingly.
The biggest problem Zhao Ye faces is not the logical design, but the physical manufacturing.
The storage unit of a register, that is, a bit, requires at least one transistor, one capacitor, a material to store charge, a shielding layer, and corresponding circuits and control components.
For an eight-bit register, there are eight such systems. For 12 sets, you need to multiply this by 12, which is 96 sets.
In the previous calculator, the processing circuit did not have registers at all. It used external storage, which is equivalent to the memory of today's computers. There is another one. Last time, it did not use the instruction set. It was all planned directly.
The circuit performs four arithmetic operations.
Now the complexity of calculations has greatly increased. If we reuse the technology from the last time, it will undoubtedly slow down the speed, and there will be many inconveniences in the design of the processor. For example, the instruction set will be difficult to play a big role.
However, the process accuracy of his last lithography machine exceeded ten microns, close to twenty microns. When converted, it is between 0.01 and 0.02 millimeters. Using such precision to make registers, Zhao Ye felt that one memory unit would have to have one.
It’s about the size of a grain of rice!
One bit is that big, and 96 bits are almost 220 square millimeters. This is still a theoretical floor area. In reality, it is at least twice as big as this, because you can't have every register next to each other! You also have to provide circuits
Make room.
Not to mention, there are also transistors! Eight-bit computing circuits cannot be less, Zhao Ye estimates that there are about 10,000. Even the most optimized mathematical design model can never be less than this number.
.
Calculating this, Zhao Ye’s processor must be at least as big as a pocket book!
We still have to shrink! History has proven that shrinking all the time is the right way!
After Zhao Ye thought about his response strategy, he found an office and started writing. He was going to "retreat" again. Before entering, he had to clearly arrange the work here. So it was necessary to write a job description.
Already.
He wrote down the work instructions for each system and said that the instructions were not accurate. This should be called work instructions.
He clarified the direction that everyone needed to tackle, and some even gave a rough explanation of experimental ideas and methods.
These people are basically used by him as assistants. Again, there is no way, there is no time for them to grow up! What they need to do is to cooperate with themselves, complete the work first, and leave time for learning and progress.
to future work.
After writing the work description, Zhao Ye found several people from the project coordination team and expressed his thoughts.
Professor Ci, Professor Zhang and Professor Li each looked at the job description written by Zhao Ye. The more they read, the more stressed they became.
It's not that it's stressful to experiment according to this, on the contrary, they feel that if they do it according to this, the probability of success is still very high.
Their pressure comes from the gap with Zhao Ye!
Look at these things, many of them don’t even have an idea. They already have concrete plans! Is the gap between this person and other people really that big?
However, Professor Ci thought for a while and said to Zhao Ye, "Yezi, take this out and think about it again? Our time is not too tight. I think for the sake of the growth of the team, should we let them work hard on their own first?"
Trying hard, but you really don’t have any solution ideas, so you’re using this as a reference?”
"Professor Ci, I understand what you mean. But I don't think it is necessary at the moment." Zhao Ye waved his hands and said. Seeing that Professor Li on the side also wanted to speak, Zhao Ye quickly added: "Teachers, your idea is correct.
, but our time does not allow it. I promised my superiors to produce the finished product within half a year. My method is to use it directly for experiments and then to industrial production. The time of half a year is already very tight."