America: The Godfather of Game Development
Chapter 502

A dedicated engine for "Need for Speed"

1,897 words6 min read0 views8h ago
Next

To be honest, Russell had seen others make empty promises to him before.

Having lived this long life, what kind of pie haven't I eaten?

I've never tried this kind of pancake before!

The digital design pipeline brought about by the new GPU and the digital twin racing video game alone are enough for him to elevate Lin Lixin to a god-like status and take good care of him.

Is there any exaggeration involved? Is it feasible?

Who the hell has Kung-Fu Master to care about that!

General Motors has no shortage of Chaniel Neil, right?

When Lin Lixin arrived, he was in his own company's Bentley.

When I left, I was driving Chevrolet Corvette ZR-1 Limited Edition.

For the next few factories at Ford, as well as those in Europe and Japan, Branca and Sam will be responsible for making one trip each.

Since bringing John Branca to their side last time, Gamenova has gradually built a more modern legal team. While it's not yet at the level of a top-tier legal firm like Pizza Hut in Nanshan or the strongest legal department in the Eastern Hemisphere, it's steadily improving and on the right track.

As for Lin Lixin's next job...

Now that we've finished the written stuff, it's time to get down to the logical stuff.

Gamenova headquarters, development department.

" Carmack, Will, let's get to work. We need to build a new engine."

Faced with a completely next-generation game like Need for Speed, Gamenova's existing technology stack simply can't compete.

Despite the fact that the development team possesses engine technology accumulated during the Quake era, which is still being continuously iterated upon.

However, many projects still choose to bypass this engine and build their workflows from scratch during development.

This is not to say that the Quake engine is faulty.

On the contrary, after such a long period of iteration and countless developers building the community environment,

The Quake engine ecosystem is already highly complete.

Even a novice developer who's just starting out can quickly get started and try to build their own version of "Quake" as long as their hardware and software fundamentals are decent. But that's where the problem lies.

The Quake engine was designed specifically for games like Doom and Quake.

It cannot be used to develop " Tokimeki Memorial " because the modules are too redundant and bloated, which would seriously drag down performance.

It also cannot be used to develop SimCity, nor can it be used to develop Ace Combat...

Different game genres often have drastically different requirements.

Even in later eras when engine technology and the PC ecosystem were highly developed, various engines still competed fiercely.

The fact that Unreal's new lighting engine is so powerful will not crowd out Unity's niche.

The Frostbite Engine's inherent strengths in AAA-level rendering and physical destruction are well-suited for EA's titles such as Battlefield, but not for lightweight indie 2D projects.

Godot's outstanding performance in independent game development cannot be carried over to the industrialized assembly line of AAA games.

RPG Maker allows beginners to quickly create their own RPG games, but it doesn't offer the same high degree of freedom as a true RPG engine.

A game engine often has its own clearly defined role.

It's difficult for any engine to claim it can handle everything. If the developers' goal is for players to exclaim "Holy crap!", then Unreal Engine is certainly excellent enough.

If an independent game developer is planning to start a garage-style game business, then Unity and Godot are actually better choices.

The concept of a "universal engine" is a complete fallacy.

Even if Lin Lixin were to use cheat codes, he couldn't possibly create something like this.

Being able to do everything means being good at nothing.

For Need for Speed, they needed a completely new toolkit built from the ground up.

Only in this way can we ensure that this work fully utilizes all the hardware's capabilities without any redundancy or waste.

But...

While Lin Lixin may not be able to create a perfect "all-purpose engine," he can push an engine specializing in Need for Speed to its limits.

The more fundamental the optimization, the easier it is to do it thoroughly.

Of course, it will be even more difficult.

Projects of this scale are clearly beyond what one or two people could handle in a week, as was the case in the 8-bit era. Lin Lixin took out his phone and dialed Activision:

"Get David here and have him bring Ben Gorcy over."

Ben Gorcy, though not as technically gifted as Carmack, was still a child prodigy.

Having Activision with Activision, my technical abilities have become quite sophisticated.

He's arguably the top graphics engineer in the Gamenova camp after Carmack.

"Oh, and Bethesda too. Have Christopher bring the entire physics engine team over."

After thinking for a moment, Lin Lixin made another long-distance call to Japan.

" Satoshi, contact Sega and ask President Nakayama to transfer that kid Yuji Naka over here."

Yuji Naka can be considered a genius programmer.

Looking at the entire Gamenova camp, his programming skills are among the best.

Of course, there was another reason why he was chosen to join the project team.

His strength lies precisely in "speed." What's the most important aspect of the Sonic the Hedgehog project?

The key is the sense of speed.

At high speeds, lens control, performance tuning, and loading of high-speed scenes all require continuous refinement and optimization.

As the head of am8, now known as snocteam, Yuji Naka has a very mature technical background in this area.

Racing games require exactly what he excels at.

Add to that Will, a genius at creating simulations.

Of course, there's also John Carmack, the most crucial god of 3D graphics, and Lin Lixin, who served as the final safety net and architect.

This dream team has finally developed some initial capabilities.

Hang up the phone.

Lin Lixin carefully examined the list, his brows furrowing slightly as he looked at Carmack beside him.

This father of 3D engines, the father of FPS, the god of computer graphics, seems to have a tiny problem.

Previously, due to differences in projects and various other reasons, he had always overlooked this point. " Carmack, if you were to write the engine for Need for Speed, what would you do?"

"Me? Uh... I built the matrix system and quaternions, wrote the mathematical foundation layer, and then wrote the 3D renderer and physics..."

"Stop, stop, stop! That's not what I meant."

Lin Lixin waved his hand, interrupting Carmack's "performance report".

"Okay, I'll give you—no, I'll give you half an hour—to write a prototype. Don't worry about how complete it is; I just want to see your thought process."

Carmack paused for a moment, then nodded vigorously and quickly ran to his workstation to get to work.

It must be said that Carmack's work is absolutely of high quality.

Sometimes, his ingenious problem-solving ideas were so brilliant that even Lin Lixin had to applaud them.

In just half an hour, although the thing was still completely unusable, it had actually taken on a rough shape.

"Um..."

Lin Lixin quickly reviewed the code, then pondered for a moment. It was pretty much what he had in mind.

Carmack's technical skills are absolutely beyond question, but his style of doing things is somewhat peculiar.

If Lin Lixin's thinking is "engineering-oriented," then Carmack's style is actually "artistic."

Or, to put it another way, a "masterpiece-level engine".

This is not a derogatory statement.

In a sense, this is precisely Carmack's greatest strength and advantage.

"I noticed that when you were working on the engine... you already had a game in mind, right?"

This is Carmack's line of thinking.

Strong target binding.

Just like the Quake engine, it was created for Quake from the very beginning.

What if we were making Doom?

Some people might choose to use the Quake engine, which has completely common and even inherited logic, for secondary development. But if it's Carmack...

He will choose to develop a separate Doom engine.

An engine can only ever serve one game.

Not even its sequel.

This means that even developing Quake 2 would require a complete overhaul of the engine.

Every part and every function within the engine is designed to solve a specific task.

Quake uses a BSP tree, and that's it; there's no room for expansion or change.

From an engineering perspective, it's somewhat inelegant.

The level of abstraction is too low.

However, from an optimization perspective, it is indeed perfect.

In Carmack's mind, the highest priority was always "being able to run".

Secondly, it must be "fast enough". Everything else is unimportant.

In other words, Carmack had already developed Need for Speed in his mind before reverse-engineering the game based on it.

Performance was pushed to its limits, development speed was incredibly fast, and there were no unnecessary layers of abstraction.

Every line of code, even every character, every constant, and every variable, is created for a final purpose.

That's good.

It's excellent, at least in the 8-bit and 16-bit era; from a geek's perspective, it's so elegant it's almost insane.

But that's precisely where the problem lies.

" Carmack, I don't know if you remember what I said. With the improvement of hardware performance and the increase in storage capacity, the era of personal development is over."

This statement certainly does not include independent game developers who sell purely creative ideas.

It targets large companies that industrialize the game industry.

Or rather, these companies that make AAA games and next-generation titles.

"You see, the engine you created is perfect, but it's not suitable for teamwork. When Industrial Light & Magic's modeling team joins, when other developers join, when Christopher and his team bring their physics engine..." "These designs, completely embedded in the code and sacrificing structure for performance, will cause huge problems."

Not everyone can analyze a large amount of highly optimized code at a glance like Lin Lixin.

For other developers, it would probably take them ten days to half a month just to study the code before they could even begin to get started.

"Moreover, Need for Speed is a game that sells its graphics. As graphics cards are upgraded, it will inevitably need a sequel. And by then, will this engine still be usable?"

The answer is no.

Lin Lixin walked up to the bookshelf and his gaze fell on a new book he had just bought a few days ago.

"Zen of assembly language"

A book on coding optimization, a representative work on machine-level thinking.

Lin Lixin originally bought this to improve his attributes, but it now serves as a reminder to him.

The author of this book is Michael Abrash, an expert in the field of computer graphics.

He currently works as a programmer at Microsoft.

But according to the original timeline, it wouldn't be long before he became a member of id Software, one of Carmack's many confidants, and the graphics technology developer for "Quake".

When it comes to computer graphics, Abrash may not be as good as Carmack.

But when it comes to optimization and engineering, he definitely has something to say.

End of Chapter
Prev
NovelSummaries
Next

Explore the wiki

Comments