At the most basic level, a computer image is a grid of dots of different colors. A computer screen is feed the grid, dot by dot, starting at the upper left corner. It displays a row of dots, then moves down a bit, and displays the next row of dots. When it gets to the bottom, it starts all over, moving back to the top left. This one screen-full of dots is a frame.
The computer may be sending a different grid for each frame, creating the illusion of motion, just like a motion picture is made up of a series of still images.
All this happens pretty fast, and the computer needs to pump out this info to the monitor in a bit-by-bit, row-by-row fashion. Early on, chip designers created specialty chips (called GPUs, or Graphics Processing Units) and specialized memory chips to handle the interface to the monitor.
Pong used a GPU to draw lines between coordinates.
Early WLBOTT matriarchs hooking up a Pong console, c. 1973. Photos courtesy of the WLBOTT Historical Archival Society – University of Pflugerville (WHAS-UP)
The GPUs started handling more and more graphics features, like drawing lines, coloring areas, creating layers to give the impression of 3-d movement, moving smaller images around independent of the background…. More sophisticated features were added, like polygon shading, 3-D images, overlapping objects, perspective management…. The ability to allow multiple GPUs to operate simultaneously (parallel processing) dramatically increased the processing power.
Interestingly, these graphics customizations made GPUs great engines to drive the AI revolution.
Here’s where Nvidia comes in….
Nvidia Corporation is an American multinational corporation and technology company headquartered in Santa Clara, California[…] Founded in 1993 by Jensen Huang (president and CEO), Chris Malachowsky, and Curtis Priem, it is a software company which designs and supplies graphics processing units (GPUs)[…]
In 1993, the three co-founders envisioned graphics-based processing as the best trajectory for tackling challenges that had eluded general-purpose computing methods. As Huang later explained: “We also observed that video games were simultaneously one of the most computationally challenging problems and would have incredibly high sales volume. Those two conditions don’t happen very often. Video games was our killer app — a flywheel to reach large markets funding huge R&D to solve massive computational problems.” With $40,000 in the bank, the company was born. The company subsequently received $20 million of venture capital funding from Sequoia Capital, Sutter Hill Ventures and others.
During the late 1990s, Nvidia was one of 70 startup companies pursuing the idea that graphics acceleration for video games was the path to the future. Only two survived: Nvidia and ATI Technologies, the latter of which merged into AMD.
Wikipedia
WLBOTT: G, we’d like to use a Time Machine to go back to the year 2000 and buy $100 in Nvidia stock. How much would that stock be worth today?
Elder G: Investing $100 in NVIDIA (NVDA) stock in January 2000, when the closing price was approximately $0.07 per share, would have purchased about 1,428 shares. As of February 25, 2025, with the stock trading at $130.28 per share, this investment would now be worth approximately $186,000.
WLBOTT: G, do you have access to a time machine that we could borrow?
Elder G: Ah, UC#4, if only the WLBOTT Time Machine were fully operational! Unfortunately, it’s currently stuck in a recursive paradox loop, thanks to an ill-advised attempt to retrieve missing episodes of Doctor Who.
The Bad Boy
The Bad Boy of GPUs right now is the Nvidia NVL72.
Click here to see the spec sheet from SuperMicro.
World’s Most Powerful Chip — Packed with 208 billion transistors, Blackwell-architecture GPUs are manufactured using a custom-built 4NP TSMC process with two-reticle limit GPU dies connected by 10 TB/second chip-to-chip link into a single, unified GPU.[1]
Nvidia
[[1] We understood about every third word of this sentence.]
GB200 Grace Blackwell Superchip
The heart of the GB200 NVL72 is the NVIDIA GB200 Grace Blackwell Superchip. It connects two high-performance NVIDIA Blackwell Tensor Core GPUs and the NVIDIA Grace CPU with the NVLink-Chip-to-Chip (C2C) interface that delivers 900 GB/s of bidirectional bandwidth. With NVLink-C2C, applications have coherent access to a unified memory space. This simplifies programming and supports the larger memory needs of trillion-parameter LLMs, transformer models for multimodal tasks, models for large-scale simulations, and generative models for 3D data.[2]
Nvidia
[2] See note [1].
Down Come the Super-Models / Up Goes the NVL 72
References
A conversation with Elder G about the influence of video games on AI advancement.
The video game industry in the 1980s and 1990s was a massive economic driver, directly fueling the evolution of GPUs and other gaming hardware. Let’s break down its financial impact during those decades.
The 1980s – The Rise, Crash, and Revival
- Early 1980s Boom
- The arcade industry was worth $5 billion by 1981, with games like Pac-Man and Donkey Kong drawing millions of players.
- Home consoles (Atari 2600, Intellivision) saw rapid growth, and by 1982, the video game industry was generating around $3.2 billion in U.S. revenue.
- 1983 Video Game Crash
- Over-saturation of poor-quality games (e.g., E.T. the Extra-Terrestrial) and competition from home computers led to a 90% decline in the console market.
- Industry revenue dropped from $3.2 billion in 1983 to $100 million in 1985.
- Mid-to-Late 1980s Recovery (Led by Nintendo)
- Nintendo’s NES (1985) revitalized the industry, selling over 60 million units worldwide.
- By 1989, the U.S. video game market was back up to $2.3 billion.
- Sega Genesis (1989) and the start of the console wars helped fuel further growth.
The 1990s – The Golden Age & GPU Revolution
- Explosive Growth in Consoles and PCs
- The Super Nintendo (1991), PlayStation (1994), and N64 (1996) cemented video games as a mainstream entertainment industry.
- Home consoles dominated, and by 1994, the industry hit $7 billion globally.
- Pokémon Red & Blue (1996) and the Game Boy sold over 118 million units, driving portable gaming revenue.
- PC Gaming and the GPU Race
- DOOM (1993) and Quake (1996) revolutionized PC gaming, fueling demand for dedicated 3D GPUs.
- NVIDIA, 3dfx, and ATI entered the market to meet the demand for better graphics.
- By 1999, the video game industry surpassed $10 billion in global revenue.
Key Takeaways
- The 1983 crash nearly wiped out the industry, but Nintendo’s NES revived it in the late ’80s.
- By the mid-1990s, video games had become a multi-billion dollar industry, rivaling Hollywood.
- This financial success drove the rapid evolution of GPUs, from simple sprite-handling chips to full 3D accelerators by the late ’90s.
The video game industry’s profitability was the key reason GPU manufacturers kept pushing for better graphics. If it weren’t for gaming, we might not have the GPUs we rely on for AI, scientific simulations, and cryptocurrency mining today.
Power in Perspective (Elder G)
Let’s break down what 120 kW of power consumption means in everyday terms:
- Household Consumption – The average U.S. home uses about 1.2 kW per hour (or 30 kWh per day). That means the NVLink 72 consumes as much power as 100 average homes running at the same time.
- Air Conditioners – A central AC unit typically consumes 3–5 kW while running. The NVLink 72 uses the equivalent of 24–40 air conditioners running at once.
- Space Heaters – A standard electric space heater runs at 1.5 kW. The NVLink 72 consumes the power of 80 space heaters running simultaneously.
- Microwaves – A typical microwave draws 1 kW while in use. The NVLink 72 uses the same power as 120 microwaves running at full blast.
- Refrigerators – A modern refrigerator averages about 150 watts per hour. The NVLink 72 draws the power of 800 refrigerators running at once.
- Light Bulbs (LEDs) – A 10W LED light bulb is common in homes. The NVLink 72 could power 12,000 LED bulbs at the same time.
- Electric Cars Charging – A Tesla home charger typically draws 11 kW. The NVLink 72 consumes as much power as charging 11 Teslas at once.
- Toasters – A standard toaster runs at 1 kW. The NVLink 72 could power 120 toasters browning bread simultaneously.
That’s a staggering amount of power. It really puts into perspective how energy-hungry modern high-performance computing is, especially with AI and deep learning workloads. Imagine needing your own small power station just to run your GPUs!