New Employees With an Annual Salary of 1 Trillion Won - Chapter 182

—————
This chapter was translated by Lunox Team. To support us and help keep this series going, visit our website: LunoxScans.com
—————

Chapter 182. Starting Line (2)

Professor Hinton hurriedly returned to Canada.

However, Jenson Huang still remained and continued research with us to complement the GPU’s shortcomings.

He immediately called Nvidia’s core engineers to Korea and began directly experimenting with various patent technologies we possessed.

“Branch Manager, applying Nexpin’s composite control patent reduced heat generation by an average of over 20%.”

“Originally it was technology used in small HDD control circuits, so stability has already been sufficiently verified.”

“Moreover, noise has definitely decreased. We can’t say we’ve surpassed competitor products, but now we’ve caught up to almost the same level. It’s all thanks to you, Branch Manager!”

This was an achievement gained in less than 2 weeks.

I put down all other work and focused solely on GPUs.

At what angle should air flow pass for heat to escape most efficiently.

Which circuits cause problems.

And when and how to cut off current to perfectly control heat generation – I found these out one by one.

Of course, this was possible because of Nvidia’s full support.

They put out everything they had to complement the shortcomings.

Design drawings, circuit diagrams, operating principles – they didn’t hide a single piece of data I requested.

Thanks to this, I could perfectly understand the structure, and modification work progressed surprisingly quickly.

“Now that we’ve established all the modification directions, we can proceed with the modification work.”

“Let’s start with Korean inventory first. We’ll bring in all global inventory soon.”

“Logistics costs will be considerable.”

“Still, it’s much better than remaining as bad inventory. If we can sell them this way, we can breathe easier.”

Jenson Huang’s expression was definitely better than before.

It wasn’t just because we could complement the GPU’s shortcomings.

During those 2 weeks, he had meetings with various game companies with Tiger Fund’s help.

“Then shall we go directly see the factory where modification work will be conducted?”

“May I come along too?”

“Of course you should see it. It’ll probably be quite different from what you’re thinking.”

“As long as modifications can be done quickly anywhere, I’m satisfied with that.”

We took a helicopter and headed to Nexcon Factory with Jenson Huang.

Soon after arriving near the factory, a massive construction site spread out below came into view.

New buildings were rising from all directions around our factory, and Jenson Huang, who got off the helicopter, looked around with eyes mixed with amazement.

“Is a new factory complex being built? Buildings are going up from all directions.”

“They’re iPod dedicated new factories.”

“All of them?”

“iPod sales volume exploded, so existing factories had production limitations. So we’re constructing new factories, and construction will be completed soon.”

The iPod was achieving worldwide success.

And that fervor still hadn’t cooled down but was getting even hotter.

Sales volume increased rapidly every month, and we had to produce new generations and various lineup models,

so simple expansion was impossible to handle.

“I heard Apple was doing well these days, but seeing the factory scale makes it feel real.”

“New iPod models are already being produced in completed buildings. Thanks to that, older factories have some room.”

“Then you’re conducting GPU complement work in factories that produced previous generation iPods?”

“That’s right. Even though they’re older factories, most facilities and equipment are still top-level.”

I read a slightly disappointed look on Jenson Huang’s face.

The term “older factory” seemed to bother him.

But when we arrived at the factory, his expression changed as if asking when that happened.

“Is this really an older factory? I’ve seen countless factories for GPU production, but this is the first place this clean.”

“It was designed from the beginning to match semiconductor production environment. After all, good environment leads to lower defect rates.”

Perfect dust-proof facilities.

Automation equipment installed in various places and control modules installed throughout.

And all employees were wearing dust-proof suits while concentrating on delicate work.

Because we had this environment, we could become iPod’s largest production base.

For next generation iPods, almost the entire volume was scheduled to be produced here.

“I really like it. If our GPUs are made in this environment, nothing could be better.”

“I was just about to mention that. We have plans to build a new Nvidia-dedicated customized production factory. To catch problems from the production process, wouldn’t it be better to make them directly in Korea?”

Jenson Huang couldn’t answer easily.

Until now, Nvidia had entrusted all production to foundry companies in Taiwan.

In a situation where he hadn’t completely verified our capabilities, it seemed difficult for him to give a hasty definitive answer.

“It’s difficult to give a definitive answer right now.”

“I have no intention of rushing either. First, I’m thinking of building a factory solely for producing high-performance computational GPUs.”

“You mean general-purpose parallel GPUs. That’s definitely possible.”

This time Jenson Huang didn’t hesitate.

High-performance computational GPUs still had nowhere to be used except by us.

Since we planned to bear investment costs and produce and use them directly, there was no reason to refuse.

“Let’s discuss that part again later, and first proceed with GPU modification work.”

“But… have you already finished line rearrangement?”

“Yes. First, I directly designed line optimization. With current lines, we can conduct modification work on over 800 GPUs per day. If we maximize efficiency just a bit more, 1,000 is definitely possible.”

Jenson Huang blinked rapidly.

Then he gripped my hand tightly as if moved.

“Is work really possible that quickly?”

“They’re all skilled veteran employees. Plus with automation facilities, we can greatly reduce process time.”

“I thought it would take at least several years. But at this speed…”

“No matter how late, we can modify all 300,000 units Nvidia possesses within this year.”

Actually, the speed could be even faster.

Due to the nature of repetitive work, work speed increases as people become familiar.

And automation equipment and line efficiency would continue improving, so we could finish all modification work within a few months, not this year.

“First, we plan to start modification work with GPUs going into PC Bangs. We can check unexpected problems fastest and re-modification is much easier.”

“You mean conducting actual usage tests through PC Bangs as well. Thank you so much.”

I don’t know how many times I’m hearing thanks.

But since I felt his sincerity, I answered with a smile.

“Now don’t worry about inventory GPUs and focus on new GPU design.”

“We’re already discussing high-computation parallel GPU design with engineers. The system Rollbook made to increase rendering speed helped us quickly establish direction.”

“It’s just mimicking parallel methods, but I’m glad it helped. I’ve also instructed Daehyeon Semiconductor to establish a parallel GPU dedicated team.”

Two people rather than one.

And multiple companies rather than one company working together could increase development speed.

Everyone knew this fact, but actually creating that structure was never easy.

Only when thick trust and interests align perfectly can everything be shared.

“If we collaborate with Daehyeon Semiconductor, we might be able to create prototypes within this year.”

“If Dr. Hinton comes to Korea with his research team, development speed will be even faster.”

“It’s been a while since I felt my blood boiling.”

“You don’t need to think too hastily. This isn’t a project that started expecting big success from the beginning. The goal is to get results through repeated successes and failures.”

Now was the stage of planting seeds.

What I had to do now was just consistently pay attention so those seeds would grow well.

Of course, I plan to generously apply fertilizer and growth stimulants to increase growth speed.

***

I headed to Daehyeon Semiconductor with Jenson Huang.

He naturally joined the newly established GPU team.

I left him there and moved to the conference room to sit face-to-face alone with President Gelsinger.

“How is it, are you satisfied with life in Korea?”

“Thanks to you, Branch Manager, I’m very satisfied. I’m so happy that I can do things here without any restrictions that I couldn’t do in the United States.”

Though he sat in the president’s position, he was an engineer to his bones.

He was deeply involved in every project, and many engineers followed him.

That didn’t mean he was indifferent to management either.

While proceeding with various projects, he was making multiple efforts to increase Daehyeon Semiconductor’s sales.

“Daehyeon Semiconductor’s sales have increased significantly since you took office as President.”

“What could I have possibly done? Thanks to Tiger Fund, we were able to secure major contracts with various companies, which led to increased sales. Especially thanks to Apple’s iPod, we were able to significantly reduce our deficit.”

The iPod contained various semiconductors.

Among them, Daehyeon Semiconductor was supplying a considerable portion of key core components such as memory and power management chips.

Naturally, as iPod sales increased, Daehyeon Semiconductor’s sales rose together.

“It’s fortunate that the deficit has been reduced.”

“Still, the DRAM market situation remains poor. The price of DRAM, Daehyeon Semiconductor’s main product, is still holding below 1 dollar. Although it’s showing signs of recovery, it will take quite some time to fully normalize.”

The worst price crash period that came after the IT bubble.

DRAM was being traded at historically low prices, and as a result, Daehyeon Semiconductor couldn’t avoid massive deficits either.

“The entire semiconductor industry seems to be struggling.”

“Only companies like us and Samjin Electronics are barely holding on. Major semiconductor companies in the United States, Taiwan, and Japan have already reduced DRAM production, and some have even gone bankrupt.”

It was a structure where the more you produced, the more losses you incurred.

So production reduction or bankruptcy was an inevitable result.

If Tiger Fund hadn’t acquired Daehyeon Semiconductor, it would still be floundering in that swamp by now.

“The DRAM market will be reorganized with companies that can survive this recession.”

“There really aren’t many companies with the stamina to endure. Even companies that have such stamina are often turning their attention to other businesses instead.”

“Then Samjin and we can divide the DRAM market between us.”

Cyclical industries ultimately approached a winner-takes-all structure.

Companies without the stamina to endure the recession get eliminated without even reaching near the finish line.

That’s why only Samjin Electronics, which had built up stamina in advance, or Daehyeon Semiconductor, which received Tiger Fund’s full support, could survive.

“When the upward cycle returns, the surviving companies will generate tremendous profits.”

“Simply enduring isn’t enough. What’s the current technology gap with Samjin Electronics?”

“…About one generation difference. However, there aren’t many companies worldwide with Daehyeon Semiconductor’s level of technology.”

One generation difference.

In terms of time, it was about a 1-2 year gap.

It wasn’t that big a gap, but I didn’t like that we were the followers.

“I’d like to narrow that gap. I don’t want to stop there – I want us to become the leading group and run ahead.”

“We’re already rapidly narrowing the gap. Daehyeon Semiconductor is investing more in R&D than any other company, and within a few years, we can catch up to a level almost equivalent to Samjin Electronics’ technology.”

I wasn’t satisfied with President Gelsinger’s answer.

He still seemed unaware of what weapon he was holding.

Tiger Fund’s financial power – it was time to make him clearly realize what that overwhelming weapon meant.

—————
This chapter was translated by Lunox Team. To support us and help keep this series going, visit our website: LunoxScans.com
—————