Silicon Magnate

You start in a basement in 1975 with limited money, space, equipment and credibility, and grow into a major hardware and software company. The setting is a fictionalised technology world: no real companies, brands or products.

The starting year is content-driven, so a run can open in a later decade instead. Research then treats the earlier eras as history the company already has behind it.

You are not just making products. You are building and running a company: the floor space it occupies, the people on the payroll, the research it can afford, and the stock sitting in a corner that nobody has tested.

Core loop

Nine steps, then again

Build the workspace
Build or improve the space you have.
Place equipment
Furniture, benches, QA stations, storage, server racks.
Hire and assign
Bring people in and put them on the right work.
Research
Unlock the technology the next product actually needs.
Design a product
Pick the family, the targets and the market.
Manufacture it
Find out how good it really turned out.
Sell it
To clients placing orders, or against an accepted contract.
Read the results
The rating, and the reason for it in a sentence.
Fix, support, or move on
Patch it, support it, or start the next one.

Step nine feeds step one. The building is usually what stops you first.

The decisions that define a run

Every one of these is a fork

None of them have a correct answer. They have consequences, and the consequences compound.

Chips or software first?
Which product can realistically ship from a basement?
Build fast and simple, or slow and good?
Compete on price, performance, reliability, features, compatibility, or support?
Invest in research, equipment, staff, marketing, QA, or support?
Rush to market, or delay for quality?
How many units to make?
How many to sacrifice to testing before you find out what you actually built?
Specialise, or expand?
Patch the current product, or move to the next generation?
Where do desks, benches, QA stations, support desks, storage and server racks go?
Is the building itself now the limit?

Hardware

The chain starts with an integrated circuit

Integrated circuits are the intended first hardware product: simpler than CPUs and GPUs, good for small teams and low-risk projects. They are also the bottleneck for everything that comes after.

Better integrated circuits let CPUs and GPUs reach higher performance without becoming unstable. Pushing a chip past its safe IC limit is allowed; it raises heat and cuts reliability. Buying ICs from another company skips the first half of the chain at a price.

Integrated Circuits, through successive packaging types from the earliest to modern, and Memory, through successive generations. Each specific product is unlocked by researching the technology behind it, and arrives with its own feature scope, target quality and price.

Performance
Reliability
Heat
Cost

A design becomes physical stock. Stock is counted and stored, tested in batches, and shipped against orders and contracts. Fabrication choices trade cost against defect risk: cheap, standard, premium, or production you own outright.

This is deliberately a supply chain rather than a factory floor. The interesting failures are a warehouse full of the wrong chip and a batch nobody tested: not a bottleneck on a conveyor belt.

An untested batch that shipped is the main way a hardware company earns its own support crisis.

Software · Designed, in development

Three ways out of the basement

Productivity and business software

The easiest path from a basement. Start with a narrow tool and grow it.

Why it works early: Small scope, few people, no fabrication.

Developer tools and utilities

Tooling for other people's work, and for your own hardware.

Side effect: Strengthens your hardware through drivers, diagnostics and compatibility.

Operating systems

A major milestone, and the one most likely to go badly.

Requires: Proper staff, QA and support. Risky and slow without them.

How software is judged

Features
User Experience
Stability
Compatibility
Support Load

More features attract more buyers, and can hurt stability. That trade is the whole design conversation.

Facility and property

You can only build on what you own

Buying property is the only way to gain floor space and floors. There is one starting office plus three purchasable expansion plots, each with its own price, floors and basements.

Floors arrive walled, and a new floor is unusable until stairs or an elevator reach it. Finished stock competes with desks for floor area, so a company that manufactures heavily ends up competing with itself for space.

What the building actually constrains.
Expansion	One starting office, three purchasable plots
Each plot has	Its own price, floors and basements
New floors	Arrive walled; unusable until stairs or an elevator reach them
Competing for space	Desks, benches, QA stations, support desks, storage, server racks, and finished stock
Later facilities	Labs, server rooms, QA areas, support departments, enterprise sales offices, campuses, industrial sites

Staff

Capability comes from skills, not job titles

You are the founder, not a hired role, and you can work anything. That is what makes a one-person company possible at the start.

Hardware Engineer: Hardware
Electronic Assembler: Hardware
Hardware QA Engineer: Hardware
Logistics Coordinator: Hardware
Software Architect: Software
Software Developer: Software
Software QA Engineer: Software
Marketer: Company
Cleaner: Company

Any employee can be cleared for a second role. The label is not the capability; individual skills are.

A technology tree of dated families, where the cost of each step reflects how long the industry actually took to get there. The size of your company decides whether you beat that timeline or fall behind it.

The eras it draws on: semiconductor improvement, personal computing, business computing, productivity software, developer tooling, operating systems, graphics acceleration, enterprise computing, networked software, digital distribution and recurring software revenue.

Outcomes

You never see the formula

Simple arcade-style scoring underneath. Every product is rated Poor, Average, Good or Excellent, and you always understand the reason in a sentence or two.

This chip is cheap and reliable, but not very powerful.
This CPU is fast, but runs too hot.
IC Quality improved. Future CPUs and GPUs can safely reach higher performance.
This software is easy to use, but lacks features.

Those are real output lines from the game. The rating tells you where you landed; the sentence tells you why, and what to do differently on the next one.

Hardware brings back

Defects, returns, heat problems, supply issues and reliability complaints.

Software brings back

Bugs, crashes, compatibility problems and patch issues.

Every launched product creates support work. Support capacity is therefore a production decision you make before launch, not after.

Status

What is built, and what is designed

Silicon Magnate is in active development. Here is the honest split, and we are not announcing a date.

Built foundation

Playable now, in the internal build.

Hardware loop: Integrated Circuits across successive packaging types, and Memory across successive generations.
Office and employees: Workspace construction, equipment placement, hiring and assignment.

Designed, in progress

Specified and being built, not yet playable.

Hardware depth: CPUs, GPUs and Computers, plus the manufacturing supply chain.
Software depth: Productivity and business software, developer tools and utilities, operating systems.
Support: Post-launch support and failure handling.

Our games

A basement, and not much else

Nine steps, then again

Build the workspace

Place equipment

Hire and assign

Research

Design a product

Manufacture it

Sell it

Read the results

Fix, support, or move on

Every one of these is a fork

The chain starts with an integrated circuit

Built and playable

A supply chain, not a factory floor