Blog

NVIDIA led the way in graphics processing unit (GPU) production, originally geared toward gaming and graphics. With virtual reality technology advancements on the rise, the software company finds that its main product is essential in the evolution of artificial intelligence (AI).

NVIDIA may not be a household name, but less than a year ago, this US-based computer chip manufacturer was the talk of the tech town.

In 2024, NVIDIA experienced a strong boost in its earnings, with its stock surging by 24%. The reason for this is the rising demand for AI, which NVIDIA is uniquely positioned to leverage. The company has been fabricating computer chips for video games that use heavy graphics for decades. But over the last few years, researchers found that these chips could power new algorithms that resulted in AI advancements—not just in virtual reality games. Other applications included scientific simulations, machine learning, data processing, and cloud computing.

The remarkable increase in demand for NVIDIA's chip led to what could be the largest single-day escalation in an organization's value. When the manufacturer unveiled its new AI-centric chip with better performance than its existing products, investor interest grew, momentarily pushing the company’s market value beyond US$1 trillion.

Introducing NVIDIA

A GPU is an electronic circuit that rapidly controls and alters memory. This process creates images faster for display. GPUs can make many calculations simultaneously, which helps with big data tasks that require significant processing power. These circuits are usually used in gaming and AI tasks.

Three computer engineers established NVIDIA in 1993. As computers advanced, the founders foresaw the increasing demand for improved visual processing capabilities. Their efforts bore fruit in the early 2000s when NVIDIA secured a contract with Microsoft to produce chips for its Xbox gaming console. NVIDIA has since expanded its presence in the growing video game industry, surpassing the television, music, and film sectors in worth and scale.

What NVIDIA offers

In 2023, NVIDIA surged to more than 100% because of the rapid expansion of the AI industry. Tech titans and start-ups rush to acquire the company's GPU chips, which can process the immense data volumes required to train advanced AI programs. It so happens that NVIDIA is the sole company that manufactures such chips.

The chip that skyrocketed NVIDIA's fame and valuation is the H100 GPU, which single-handedly reshaped the AI industry and increased the company's capitalization by US$1 billion. Suddenly, NVIDIA's worth exceeded that of Facebook and Tesla, and sitting just below major tech giants like Google, Microsoft, Apple and the Saudi Arabian government-owned oil bigwig Saudi Aramco. 

The H100 takes its name from Grace Hopper, an American computer scientist and mathematician known for her vital contributions to software development and computer programming. Essentially, the H100 shares the architecture of gaming GPUs but has been tailored to manage extensive processing and perform rapid calculations efficiently. These capabilities make the H100 compatible with AI model training.

How? AI can only interact with users, generate videos or images, and carry out various tasks through training using extensive datasets. The greater the data volume, the higher the proficiency of the AI model.

The learning process of AI models is built on trial and error, where it performs a task repeatedly until it achieves the desired outcome. Such an iterative procedure demands massive computing capability.

The H100 is a remarkable innovation because it boasts four times the speed of its earlier version, the A100. This means that training large language models is faster and users respond to requests 30 times more quickly. This improved performance is very important for businesses that want to train LLM models more efficiently.

It's no wonder then that only eight of NVIDIA's breakthrough chips cost US$30,000, and companies don't hesitate to purchase them by the thousands. In fact, X owner Elon Musk was reported to have bought 10,000 GPUs to jumpstart his new AI venture.

NVIDIA's emerging competitors

Currently, NVIDIA dominates around 80% of the GPU market used in AI facilities. Big Tech companies, including Google, Microsoft, and Amazon, are some of its clients.

But these tech giants are catching on by creating their rival chips. This harks back to Apple's years-long effort to manufacture its own chips to cut down costs and its dependence on outside suppliers. 

Semiconductor chip companies Intel and Advanced Micro Devices (AMD) also produce GPUs and central processing units (CPUs). AMD launched its Instinct MI300X GPU last December, said to process data 1.3 times faster than the H100 offering. Meanwhile, Intel also unveiled its Arc GPU. CPUs have always been Intel's strength, but the company strives to keep up with the growing GPU demand, which has outpaced that of CPUs.

However, NVIDIA is not resting on its laurels. Soon, it will launch the H200, which the company website boasts as "the world's most powerful GPU". It boasts unparalleled capability in accelerating AI and high-performance computing (HPC) operations. This new offering seeks to sustain NVIDIA's top position in the AI market.

Moreover, NVIDIA holds the advantage of owning CUDA technology, which performs rapid and precise calculations in graphics cards. This enables developers to create algorithms that operate on GPUs. Because of NVIDIA's considerable head start in the industry, experts believe the company will continue dominating the market despite emerging competition.

NVIDIA's gaming tech developments

Though NVIDIA is catering to diverse industries that are leveraging AI, the company retains its focus on gaming. It's currently experimenting with the possibilities of generative AI to develop distinct dialogues for Non-Player Characters (NPCs), which are characters controlled by AI instead of a player. Generative AI can also help NPCs better control their surroundings.

But the most common concern of using AI still presses on workers, especially artists, who are afraid of losing their jobs. NVIDIA has debunked this fear, saying that today's AI demands more creative writing. Rather than producing tired and superficial conversations, NPCS now require elaborate yet convincing narratives that shape their actions within the game. 

NVIDIA may continue to lead the industry, but in the long run, healthy competition boosts innovation, accessibility, and affordability. Competition also fosters consumer choice, allowing the public to have better products and services at the best price. But for now, NVIDIA is basking in the spotlight, which it has worked hard to reach.

 As one of the Top 20 EMS companies in the world, IMI has over 40 years of experience in providing electronics manufacturing and technology solutions. 

 We are ready to support your business on a global scale.

 Our proven technical expertise, worldwide reach, and vast experience in high-growth and emerging markets make us the ideal global manufacturing solutions partner. 

Let's work together to build our future today.