Cache

Currently, Fuxing Base is still small. When its scale expands in the future, it will be impossible for the base to operate without purchasing hundreds or thousands of units.

Especially since Han Yang also plans to start excavating the underground base after the computing power is upgraded again.

The Revival Mall on the surface is just a cover. For safety, Han Yang decided from the beginning that the future development would definitely be underground.

However, for the excavation of the underground base, it is certainly impossible for Han Yang to find a large number of workers and mobilize a large number of heavy machinery to do the job.

He still has to rely on his own computing power to control robots to dig.

By then, the requirements for computing power will be even higher. And after the base is built, the number of robots needed to operate more equipment and conduct more scientific research will not be small.

Imagining the scene of controlling thousands of robots busy in the base simultaneously when his computing power is greater in the future, Han Yang's heart was filled with anticipation.

But at this stage, he can only anticipate. Without developing more advanced chips and improving computing power, everything is just an illusion.

Time slowly passed. Half a month after the first trial production, on this day, the robot, as usual, moved a box of materials stacked in the corner of the laboratory to the beginning of the production line, opened it, and poured the materials into the feeding port.

Fixed robots worked together with the machines to start processing this batch of materials. The ten mobile multi-functional robots shuttled back and forth within the production line like diligent workers.

Feeding, operating, sorting, integrating, and discharging, everything was orderly.

In the entire large laboratory, there was only Han Yang. At this moment, he was still in front of the computer listening to a math lecturer. Behind him, everything was left to the machines and robots to operate, just like a fully automated assembly line.

When this batch of raw materials also turned into specific types of cables coming out of the discharge port, and then were bundled by robots and sent to dedicated testing equipment for inspection, the numbers on the instrument entered Han Yang's eyes, immediately invigorating him.

After half a month of repeated modifications, debugging, and attempts, the first batch of cables that met Han Yang's requirements were finally produced.

A total of 100 cables were produced in this first batch. Their length was 12.5 centimeters, and both ends had a small connector, looking very delicate.

On the other side, another piece of equipment was also roaring, producing things as small as grains of rice, which were also sent to the testing equipment.

These things were all transistors.

Cables are equivalent to circuits, circuit boards are equivalent to silicon wafers, and silicon wafers, circuits, and transistors are the three most important things for constructing a CPU.

In formal CPU production, transistors and circuits are directly photolithographed onto silicon wafers using light, and there is no need for such trouble at all.

At this moment, Han Yang could only achieve the replacement of photolithography technology through this method.

At this moment, the circuit boards that met Han Yang's requirements, as well as the first small batch of cables and transistors, were ready.

With a thought, a multi-functional mobile robot carefully picked up these materials and placed them on the work platform.

The fixed robot on the work platform immediately received Han Yang's computing power infusion.

Under Han Yang's direct operation, the new generation of CPU finally officially began manufacturing.

During this period of long-term learning, Han Yang had gradually conceived the architecture and manufacturing method of the new generation of CPU in his mind.

Compared with the previous generation of CPU, the logical structure of this new generation of CPU was more complex. Han Yang added multiple self-created instruction sets to it and adopted the pipeline design mode for the first time.

Pipelined circuits greatly improved efficiency by allowing multiple instructions to perform different tasks at different locations on the circuit at the same time.

This is a very mature technology. In modern society, even the oldest CPUs have widely adopted this technology. But for the circuit board CPU hand-made by Han Yang, this was the first time this technology was used.

According to theoretical predictions, this technology would greatly improve the performance of the CPU he manufactured.

In addition, Han Yang also added a high-speed cache of up to 4KB to this CPU for the first time.

Modern CPUs generally have three levels of cache to maximize performance. But for Han Yang, creating a cache structure was already the limit of existing technology.

The addition of many new designs made this set of CPU architecture much more complex than before, and correspondingly, the manufacturing difficulty also increased significantly.

Although Han Yang had designed this set of architecture in advance, he knew that theory was ultimately just theory.

In the specific manufacturing process, many new problems that he had not discovered before would continuously emerge.

Meeting problems as they arise and solving them one by one was the only feasible solution for Han Yang at this moment.

As expected, before the first batch of 500 transistors and 100 cables were used up, problems emerged.

Insufficient welding accuracy.

The cables and transistors were too small. When they were fixed on the circuit board by welding, the welding points would affect each other, causing a short circuit.

This was a problem that had never occurred before.

Han Yang spent a lot of thought on this problem. Finally, Han Yang spent a high price to urgently buy a high-precision welding gun and purchased special solder, which solved this problem.

For this, the first batch of transistors and cables, as well as the circuit boards, were all scrapped and turned into garbage.

Fortunately, the second batch of transistors and cables had already been made, so Han Yang didn't have to wait any longer.

After solving this problem, Han Yang continued to manufacture.

Soon after, the second problem appeared.

Han Yang found that his original design had a BUG. The distance between two modules in the CPU was too far, located on two different circuit boards, and needed to be connected by cables up to 62 centimeters long.

However, the cables manufactured with existing technology could not achieve the required accuracy and speed.

There was no other way, Han Yang could only adjust the architecture, build these two modules on the same circuit board, and move the modules originally located here to other circuit boards.

But this would affect the overall structure, and a change in one part would affect the whole. The communication reliability, communication delay, etc., between all modules had to be reconsidered.

Fortunately, this was only an adjustment to the overall structure and did not involve details, so it was not too troublesome. Han Yang spent five days completing all these adjustments.

After that, the manufacturing work continued.

After the third day, Han Yang's work stopped again, and he began to research and think again. After two days of interruption, the work continued, and then stopped again after four days, and then stopped again after three days