"Well, theoretically that's correct..." Vivian set down the control board and switched off the television. "But in practice, it's impossible."
She immediately poured cold water on the excitement, as it goes reality always had a way of falling short of theory. And nowhere was that gap more apparent than in graphics hardware.
"I understand it, the problem is video memory, right?"
Ethan sighed.
After going in circles, it always came back to capacity. No matter the era, memory remained one of the most expensive components in digital devices.
"Even for just 256 colors, we'd need over 300 KB of VRAM at a resolution of 640×480."
"Exactly," Vivian nodded regretfully.
"And if we're going to make it a general-purpose graphics card that PCs can use, bandwidth becomes another issue."
At 640×480 resolution, even 16-color graphics were already pushing the ISA bus to its limits.
Things were manageable on the NovaBox's highly integrated architecture. But if they wanted to turn it into a universal graphics expansion card, as Ethan envisioned, things became much more complicated.
It wasn't until standards like PCI appeared years later that these bottlenecks would be significantly improved.
"The amount of money we'd have to burn on NAND technology wouldn't be any less than what we're spending here..."
Ethan rubbed his forehead.
The semiconductor industry was simply a bottomless money pit.
Even though GAMENOVA had already become one of America's leading game companies and had secured a substantial share of the Japanese market, they still looked like rookies when standing before the abyss that was the semiconductor industry.
"I'll figure out the VRAM problem later. For now, let's push this standard forward. Where's Jim?"
"At SGI, he says he wants to build a workstation based on a RISC architecture."
"That's good. Then sixteen colors it is, so now let's get the first batch of VGA graphics cards into production as soon as possible."
Ethan thought for a moment. All they really needed was to tweak the specification slightly so it could support both 640-resolution 16-color mode and 320-resolution 256-color mode.
For game consoles of this era, that was already more than enough.
If they wanted to challenge IBM, they had to move fast. They needed to seize control of the standards before IBM even realized what was happening.
And to achieve that, releasing ISA (Industry Standard Architecture) graphics cards that could plug directly into PCs was absolutely necessary.
"This product line..." Ethan paused.
"Let's call it Radeon… NovaRadeon."
"..."
NovaRadeon
It certainly had a nice ring to it. Of course, hardware alone wasn't enough to make a universal graphics card successful, and there was a reason Ethan had transferred Carmack here so early.
The software side needed preparation as well, and the card had to provide standardized libraries for DOS. This would effectively become the foundation for future graphics APIs, the precursor to OpenGL.
After finishing these arrangements, Ethan finally prepared to leave, then he suddenly stopped.
"Wait, that's not right."
He tapped his forehead.
"I almost forgot why I came here in the first place."
"I need an MMC enhancement chip for a new game."
...….
Back in his office, Ethan pulled out several Famicom cartridges from his collection.
Before designing the MMC enhancement chip, he wanted to understand exactly how Nintendo's developers had managed to squeeze so much performance from such limited hardware. The first cartridge he inserted was Pinball.
[Insight gained from playing Pinball: Game Physics EXP++]
[Insight gained from playing Pinball: Game Physics EXP++]
If there was one Nintendo title Ethan desperately wanted to get his hands on, the answer would undoubtedly be Pinball, released in early 1984. It was one of the Famicom's earliest games and one of the NES launch titles.
Its developer, HAL Laboratory, was Nintendo royalty in every sense. Future masterpieces such as Kirby and Super Smash Bros would come from the same studio.
Even Satoru Iwata, the future president who would eventually succeed Hiroshi Yamauchi, was still honing his skills at HAL during this period and Pinball happened to be one of Iwata's early works.
Ethan stared suspiciously at the silver ball bouncing around the machine's mechanisms, continuously racking up points.
"This is strange."
"How did the NES pull off physics simulation at this level with such terrible hardware?"
The NES processor was merely a 1.8 MHz Ricoh 6502, so performance-wise, it was nowhere near the NovaBox, and with hardware that weak, even a simple 2D physics simulation shouldn't have run this smoothly.
"Did Iwata have some kind of secret technology too?"
With his current knowledge, Ethan was confident he could create physics effects vastly superior to those in Pinball.
But he couldn't achieve them with such minimal performance costs. After all, Double Dribble needed to display ten players moving at high speed simultaneously, so he simply couldn't spare much processing power for complex physics calculations.
[Ding!]
[Skill Acquired:]
[The Finest Deception]
[Master basic game physics. Hardware-related attribute growth multiplier increased by 1.4x.]
"Hm?"
For the first time, Ethan experienced a very strange feeling. His understanding of game physics had clearly improved, yet he still couldn't solve the problem before him.
"No…" He stared intently at the silver ball. He watched the exact moment it was launched by a flipper.
Suddenly, something felt off, perhaps because his understanding of game physics had deepened, Pinball's physics no longer appeared quite so flawless.
He began noticing strange flaws everywhere. The ball's movement felt too rigid. Originally, Ethan had assumed this was simply due to the NES's limited processing power causing lower simulation accuracy.
The NovaBox occasionally suffered from similar issues, but the more closely he observed, the more obvious the truth became.
This wasn't a hardware limitation. The problem lay within Pinball's so-called physics simulation itself.
Every movement of the ball appeared free, and it seemed to bounce naturally around the playfield, yet upon closer inspection, every state transition was exactly the same.
"Now I understand why it's called a deception..."
His entire line of thinking had been wrong. Pinball wasn't performing physics calculations at all. At its core, it was using lookup tables.
Ethan grabbed a notebook and began scribbling furiously.
"Slow roll. Fast roll. Free fall. Rebound..."
The ball's speed in every state had already been predefined.
"If it's on a slope, add a fixed value..."
"When it hits a bumper or flipper, use the normal direction..."
"Wait... even the collision detection is fake?"
The deeper he dug, the more astonished he became. What looked like a perfect physics simulation was actually an elaborate illusion prepared for the player. Pinball used the smallest possible amount of computation to create an effect convincing enough to fool everyone.
"This is the kind of information you don't ignore, it's absolutely incredible."
In an era where both performance and storage were precious resources, only these kinds of unconventional tricks and ingenious shortcuts could produce effects that seemed impossible.
Ironically, this was precisely the skill that someone like Ethan, raised on modern development methods, was most likely to overlook.
"I've been so focused on hardware lately that I've gotten stuck in my own way of thinking."
At last, he understood. No matter how a game works internally, no matter how its systems are implemented beneath the surface. Video games are ultimately made for players, so as long as the final result delivers the intended experience, the implementation itself is irrelevant.