"Er Zi, why aren't you tearing tape anymore? Are you done?" Da Dong looked curiously at Wang Xu, who was leisurely basking in the sun by the window.

"Hmm, it's done. The results have been accepted, and now we just wait for the output. How long has it been since you paid attention to me? I stopped tearing tape a week ago. Go on, what do you want? I can tell by the way you stick out your butt."

Wang Xu pouted. Da Dong was clearly making small talk because he wanted something.

"Hold on, hold on, we're eating. Watch your language," Zhou Feng immediately stopped Wang Xu's nonsense.

"I was just asking if you had any projects we could work on. We're in our second year now, and professional courses have started. Optics is boring. I see you're always messing with this and that, and it all ends up benefiting the Department of Materials Science. Can't you come up with some projects for our class too?" Da Dong was the class's Youth League secretary last year. Although he had stepped down this year, he was still very enthusiastic about class matters.

By the way, at Shuimu University, the Youth League secretaries for each class change every year, and student union positions also have high turnover. It's rare for one person to hold a position for a long time. After all, they are all geniuses from various places; who is to say who is much worse than whom? It's not about whether everyone is convinced, but there needs to be some balance, right?

"Don't mention it, I actually do," Wang Xu suddenly had an idea.

"Tell us," the other roommates, hearing about a project, also gathered around, holding their lunchboxes, either standing or sitting in a circle.

What? Why were they all eating in the dorm instead of finishing in the cafeteria and then coming back? Well, because at this time, the Ninth Cafeteria simply had no seats. People could only stand to eat, and even then, they couldn't find a table to put their lunchboxes on, so it was better to eat back in the dorm.

Shuimu University had opened several new cafeterias in recent years, with numbers going up to sixteen. The new cafeterias were bright and clean, but they were far away, and lazy people didn't want to make the trek.

"Let's build a Supercomputer for fun. Although this kind of thing is more for the Department of Electronic Engineering and Computer Science Department, who are we? Automation! We're related to everyone, so we can do it," Wang Xu said.

"Pfft~" Well, these brothers, some spat out rice, some spat out water, and the tabletop immediately became a mess.

"Is it necessary? How disgusting," Wang Xu complained about his unreliable roommates while wiping off the rice grains that had splattered on him.

"You really can talk big, you really dare to say anything," Da Dong also wiped off the water he had sprayed, talking as he wiped.

"A Supercomputer? The country invested over a hundred million and still hasn't made one. You just move your lips and plan to build a Supercomputer. Although you've invented some things, don't be so outrageous, okay? I didn't know before, but you really can brag," Zhan Wenbo, Zhan Laoda, obviously also thought Wang Xu was talking nonsense.

Indeed, since the Yinhe 100-million-operations-per-second computer in '83, there had been many years without news of progress on domestic supercomputers. The term "Supercomputer" had only been translated from foreign news in recent years; before that, it was always called a giant computer.

The Supercomputer project was naturally one of them, and it was a very specific field. Wang Xu thought of doing this mainly because current computers were simply not powerful enough.

But current Supercomputers, not to mention how ridiculously expensive foreign ones were, domestic ones were simply unusable.

For example, the Yinhe, with 100 million floating-point operations per second, was an important scientific research product that broke the domestic monopoly. But in reality, everything inside was hand-soldered. Can you imagine a computer with 2 million solder joints and 400,000 flying wires, all hand-soldered?

This thing was essentially taking what should be done inside the CPU and directly creating it with circuits. It was like a junior classmate of Wang Xu's in later generations, who directly soldered an 8086 CPU using thousands of transistors on a table as his graduation project; the Yinhe was essentially something similar.

This kind of thing not only had no possibility of mass production or batch manufacturing, but its use value was also very small because its maintenance was extremely difficult. Any problem with a flying wire or solder joint would require an incalculable amount of time to troubleshoot.

So, the most important role of the Yinhe was not to calculate anything with it, but to tell the world that we had it, and then internationally, some conditions for purchasing giant computers would be lowered, breaking some blockades.

Didn't you see that as soon as the Yinhe came out, several giant computers were successfully introduced domestically? Although there were still many restrictions, including only specific uses and requiring foreign personnel to be present for use, at least they could be bought.

Moreover, current Supercomputers couldn't perform the operations Wang Xu needed. They didn't have a general-purpose operating system or commercial software; all calculations were done by custom-written programs.

Thus, Wang Xu had the idea of creating a simplified version of a Supercomputer. But this thing wasn't something Wang Xu could create alone. Although the system space sold technology, computer special effects and VR, for example, were important components of entertainment.

But Wang Xu couldn't afford these things yet, even if they weren't expensive, he'd have to save up for a while. On the other hand, even if he bought them, he couldn't build them; too many complex industrial fields were involved, so he could only proceed gradually.

But this didn't mean Wang Xu couldn't create something now. If one person couldn't do it, then a group of people could. Shuimu University had no shortage of talent. What was the purpose of Wang Xu frantically accumulating prestige? Wasn't it to utilize Shuimu University's resources as much as possible? What sheep would be more woolly and long-haired than the one right in front of him?

Personal computers had already entered the 32-bit era, but the capabilities of a single computer were still too weak. Minicomputers were too expensive, and CPU upgrades were limited by Moore's Law, so doubling performance wouldn't happen that quickly.

And the mainstream Supercomputers now were still vector computing models like Cray, which was a type of parallel computing. To put it simply, this thing was the computing model of graphics cards in later generations. In later generations, Cray machines were directly made using GPU graphics cards, such as the Titan machine at Oak Ridge.

Today's Cray machines, you can think of their core as a super-large GPU stacked from a bunch of RISC chips. These chips are like the cores in a GPU.

Then, with the addition of input, output, storage, and other devices, its performance is actually not as fast as a single graphics card in later generations.

We can almost say that after the miniaturization and integration of Cray machine technology, it becomes the basic GPU.

And although this thing is fast, it's very expensive, super expensive, and consumes a lot of power, making it completely unsuitable for small and medium-sized enterprises.

So, how can we create a computer with simple raw materials that can achieve high computing speeds?

Actually, later generations had the answer: distributed computing.

Distributed computing is a method that uses a special operating system to send separable computational steps to different computers for calculation, and then aggregates the results. This is essentially the predecessor of the "cloud" technologies that everyone is familiar with in later generations.

And parallel computing is actually consistent with distributed computing in its basic principles, except that it's not allocated to external computers, but to different CPUs internally, so it's faster, but the design is more complex, and the number of integrated CPUs is limited.

However, talking about speed without considering the application is just being a hooligan. First, you need to understand what you want to do.

If you want to be efficient in parallel algorithms, the content you're working on must support parallel computing, meaning the original computation needs to be decomposable into many independent simple operations.

For example, 1+2+3+4 can be easily split into two independent operations, 1+2 and 3+4, and then the sum of the results can be calculated. On one computer, it takes 3 clock cycles, while on a parallel computer with two CPUs, it only takes two, which increases the speed.

But some things, because subsequent operations require the result of the previous step, cannot be split this way, so parallel computing cannot significantly improve efficiency. For example, 1+2*3 will take two clock cycles to calculate on any machine.

Distributed computing also has similar problems. If your work can be divided into many similar subtasks, then distributed computing is ideal. For example, in animation rendering, materials can be sent to many computers to render individual frames.

After getting the rendering results, they are first combined, then divided into segments, and sent to multiple machines for video generation. By repeating this, a complete rendered animation can be obtained.

Audio processing is actually similar to video processing. Audio can also be divided into segments for various processing. Different audio tracks are inherently different subtasks, so audio processing can actually be done using a distributed approach.

It's just that audio processing is much simpler than video processing, so it doesn't need to be split in such a complicated way. A single DSP in later generations can handle it directly.

But current CPUs and DSPs don't have such powerful capabilities yet, so it will be a bit more troublesome, requiring some distributed or parallel techniques.

Use a high-performance PC as the host, responsible for human-computer interaction and task distribution. Multiple computers equipped with multiple sound cards act as computing units to process signals.

Also, equip a centralized storage unit, using RAID for intermediate data storage as a cache, and multiple tape drives as permanent storage devices. Yes, that's the general structure.

Doesn't it look familiar? That's right, many small 3D image rendering workstations in later generations use this model, but with sound processing units replaced by graphics cards.

Overall, building such a system now is somewhat not worth the effort, because various limitations will lead to extremely limited performance improvement. Moreover, this kind of thing already exists abroad and is quite developed; otherwise, where do you think the equipment in those top recording studios abroad comes from?

But the problem is, this thing has too strong a signal processing capability and can be applied to, well, you know. So, it was banned, not allowed to be sold, which meant Wang Xu had to build it himself.

Moreover, the accumulation of this technology is meaningful, and very significant, so much so that it can continue for decades.

The operating system, time-sharing system, interactive system, network routing, information exchange, and many, many other things contained within it, if developed smoothly, will become the future cloud, without even needing to change direction.

Wang Xu planned to use his initial capital, and then fleece Shuimu University, this big sheep, to create this super-simplified, basic audio-specific version of the future cloud.

After Wang Xu explained the detailed requirements to his roommates, he saw their contemptuous gazes.

"What kind of look is that?" Wang Xu said angrily.

"You call this a Supercomputer?" Zhan Laoda said indignantly, clearly upset about the disgusting lunch he had just had.

"Super Audio Computer, abbreviated as Supercomputer. Is there a problem?" Wang Xu stuck out his neck.

"Get lost!" Several roommates simultaneously raised their feet, intending to kick him.

The few of them messed around for a while. That's how college roommates are; everyone thinks they're the boss. Well, at least Wang Er thinks so, and they probably think so too. This tradition doesn't seem to have changed in later generations.

"Alright, no more fooling around. Just tell me if you're in or not."

"We can do it, but I don't think just a few of us can. Your plan has too much in it, even an operating system. It's too much; is this something a few students can do?" This was Da Dong's complaint.

"Yeah, do you really think we few can do this?" Even Zhan Laoda was a bit unconfident. This guy usually acted like he was the second coming of God, but when push came to shove, he still chickened out.

"What's there to not be able to do? If I didn't have time, I could handle this operating system by myself. It would probably only take a month or so," Wang Xu said, puffing out his chin proudly.

This really wasn't bragging. His skill infusion included the source code for this, specifically the device repair technology. It would just be a killer to type out so much source code. Moreover, the source code was only applicable to special equipment and couldn't be copied directly, so to implement it, they would still need people to build it.

"You just boast. You need to come up with a plan; just talking won't do. If the plan is feasible, we can go to Uncle Qiang and report it higher up. I think with your current favor within the department, applying for a project shouldn't be a big problem.

But since it involves so many people, and possibly across departments, it's best to bring in the Youth League Committee and the Student Union. Their organizational and coordination abilities can be a great help," Da Dong, despite his young age, had a broad perspective.

That year, he was a provincial-level 'Three Good' student and was directly admitted to Shuimu University. Then he served as the Youth League secretary for a year. If he hadn't been lazy, he would probably be in the department's Student Union by now.

Oh, right, Uncle Qiang is the current Youth League secretary, named Wu Guoqiang. Because he is highly trusted by everyone, they affectionately call him Secretary Qiang, or simply Uncle Qiang.

"Hmm, alright, I'll put together a plan right away and get things moving. I've been busy with the Department of Materials Science recently; I can't let our folks stay idle," Wang Xu said with a sigh, earning another round of laughter and scolding.

In the following days, Wang Xu spent his time on this Supercomputer project. Although calling it a Supercomputer might be an exaggeration, others didn't know, but Wang Xu understood that once the technology accumulated, it would truly be the blueprint for a Supercomputer.

—

(Chapter One, four thousand words. Too lazy to split it into Chapter One and Two, so please bear with a longer read.)