Real-time Graphics
MasteredInvented or productized BSP trees, surface caching, adaptive tile rendering, and binary-space partitioning at scale.
How it developed
Decades of reading SIGGRAPH papers and reimplementing every algorithm from scratch.
A life, end to end
John Carmack
Programmer; Co-founder, id Software; ex-CTO, Oculus.
The programmer who shipped 3D gaming to the world, open-sourced his engines so the next generation could build on them, and then bet his second act on virtual reality and AGI — always optimizing for raw shipped code over status.
- Birth Year
- 1970
- Industry
- Software & Game Engines
- Country
- United States
- Key Achievement
- Wrote the rendering engines behind Wolfenstein 3D, Doom, Quake, and the early Oculus stack — and open-sourced most of them so the entire game industry could be built on his shoulders.
Life Timeline
The full arc, year by year.
Every story has the highlights. This is the boring middle, the doubts, and the moments that quietly changed everything.
- 1970
Born in Shawnee Mission, Kansas
Father was a TV reporter, mother a school administrator; an early Apple II at age 10 set the trajectory.
ChallengeNo formal CS education available to a curious 1980s kid.
LessonManuals plus persistence beat curriculum every time.
- 1981
Juvenile detention for an Apple II break-in
At 14, broke into a school to steal Apple II computers; got a year in a juvenile home.
ChallengeA felony at 14 that nearly closed off the conventional path.
LessonThe thing you can't stop doing is the thing you should be paid for.
- 1989
Met John Romero at Softdisk in Louisiana
Hired as a contract programmer; the team that would become id Software coalesced over weekend hacks.
ChallengeStuck producing low-rent monthly software for a magazine subscription.
LessonYour day job pays for the side project that becomes your real life.
- 1991
Founded id Software
Quit Softdisk with Romero, Tom Hall, and Adrian Carmack to build PC games full-time from a Shreveport lake house.
ChallengeConvincing distributors that shareware could be a real business model.
LessonDistribution innovations matter as much as the product.
- 1992
Shipped Wolfenstein 3D
First true 3D action game on PC, distributed as shareware; spawned the first-person shooter genre.
ChallengeConvincing the team to abandon the safer side-scroller they were good at.
LessonThrow away the working thing to ship the better thing.
- 1993
Shipped Doom
Networked deathmatch, BSP rendering, and a soundtrack that defined a decade; sold by direct download.
ChallengeReal-time 3D on hardware that wasn't supposed to be capable of it.
LessonAlgorithmic insight beats hardware budget.
- 1996
Shipped Quake
First fully 3D-polygonal multiplayer game with hardware-accelerated rendering; created esports and modding culture.
ChallengeEighteen months of grueling crunch and internal conflict with Romero.
LessonShipping eats friendships unless you protect them deliberately.
- 1999
Open-sourced the Quake engine
Released the source code under GPL, seeding Half-Life, Call of Duty, and a generation of modders.
ChallengeConvincing the business that giving away the engine wouldn't kill sales.
LessonOpen source is the deepest distribution channel.
- 2000
Founded Armadillo Aerospace
Self-funded rocketry company chasing the Lunar X Prize on weekends from a Texas hangar.
ChallengeAerospace doesn't share the iteration speed of software.
LessonHardware humbles you; do it anyway.
- 2012
Joined Oculus VR
Met Palmer Luckey, became CTO, and ported Doom 3 to early prototypes that made VR feel real.
ChallengeLeaving id to bet on a barely-functional headset from a teenager.
LessonFollow the wave that excites you most, not the one you already conquered.
- 2014
Oculus acquired by Facebook for $2B
Stayed on as CTO through the platform's most formative years and the launch of Quest.
ChallengeWorking inside a large company after decades of small-team freedom.
LessonScale changes which problems are interesting; pick accordingly.
- 2019
Moved to Oculus 'Consulting CTO'
Stepped back to focus most of his time on artificial general intelligence research.
ChallengeWalking away from VR right as Quest hit mass market.
LessonKnowing when to leave is harder than knowing when to start.
- 2022
Founded Keen Technologies
Raised $20M to pursue AGI from first principles with a tiny team — same playbook as id.
ChallengeEntering a field already dominated by billion-dollar labs.
LessonSmall teams with clarity outpace huge teams with politics.
Skills Acquired
What they learned to do well.
Skills aren't talents — they're the residue of a thousand decisions. Here is what compounded over a lifetime.
Engine Architecture
MasteredDesigned engines explicitly to be modded and ported — Quake ran on phones a decade after its release.
How it developed
Hard-won lessons shipping six landmark game engines.
Performance Optimization
MasteredTreated every clock cycle as a moral obligation; rewrote inner loops in assembly when needed.
How it developed
Constrained 1990s hardware forced micro-optimization as a discipline.
First-principles Thinking
MasteredRefused to inherit architecture from prior engines; rebuilt the renderer from the ground up for each game.
How it developed
Wolfenstein, Doom, Quake — three engines, three full rewrites.
Writing Plain English
MasteredHis .plan files and tweet-storms became technical literature read across the industry.
How it developed
Decades of writing devlogs and changelogs intended for other engineers.
Long Focus Blocks
MasteredFamous for 12-hour Diet Coke-fueled coding sessions with door closed.
How it developed
Treated attention as a depletable resource and defended it ruthlessly.
Open Sourcing
MasteredReleased engine source code after each generation, fueling Half-Life, CoD, and modern engines.
How it developed
Believed knowledge compounds only when shared.
Failures & Challenges
The chapters most pages skip.
No journey is a straight line. The setbacks weren't detours — they were the route.
Daikatana collapse with Romero
Context
Personal split with co-founder John Romero damaged both their next decade.
Recovery
Stayed focused on the engine work; eventually reconciled publicly years later.
Lesson
Co-founder fights are real; don't let them outlive the company.
Armadillo Aerospace hibernation (2013)
Context
Self-funded rocketry company ran out of momentum without crossing milestones.
Recovery
Wound down gracefully and applied the lessons to Oculus hardware work.
Lesson
Pivot or pause before you burn out trying to force a market.
Doom 3 reception
Context
Technically dazzling but criticized for slow pacing and dark corridors.
Recovery
Folded the criticism into Rage and later engine work.
Lesson
Tech demos aren't games; remember the player.
Friction at Meta
Context
Publicly criticized Meta's organizational drag while still inside the company.
Recovery
Left on his own terms to pursue AGI rather than fight internal battles.
Lesson
When the org stops shipping, the engineer should leave.
Books & Resources
The library that shaped them.
The books on the shelf, the people they studied, the ideas they kept returning to.
Masters of Doom
David Kushner
The definitive history of id Software and the Carmack/Romero dynamic.
The Art of Computer Programming
Donald Knuth
Carmack has called Knuth the deepest influence on his engineering aesthetic.
Real-Time Rendering
Akenine-Möller et al.
He's recommended this as the canonical textbook for graphics programmers.
Surely You're Joking, Mr. Feynman!
Richard Feynman
Modeled his first-principles temperament after Feynman's.
Snow Crash
Neal Stephenson
The fictional ancestor of VR thinking he carried into Oculus.
Code Complete
Steve McConnell
Standard reference he's recommended to younger engineers.
Videos & Documentaries
Watch them in their own words.
Interviews, keynotes, talks, and documentaries — chosen for the moments that reveal how they actually thought.
Key Decisions
The forks in the road.
The bets that, made differently, would have written a different life.
Open-sourcing the Quake engine (1999)
Risk · High- Why
- Believed the community would extend the engine farther than id ever could.
- Outcome
- Spawned an entire ecosystem including Half-Life and Call of Duty branches.
- Long-term impact
- Reshaped industry norms around engine licensing and modding.
Leaving id for Oculus (2013)
Risk · High- Why
- Saw VR as the next platform shift and id as a finished chapter.
- Outcome
- Stabilized early Oculus tech and shaped Quest's mobile-first architecture.
- Long-term impact
- Gave VR credibility with skeptical PC engineers.
Walking away from VR for AGI (2019–2022)
Risk · Extreme- Why
- Believed AGI was the highest-leverage problem available to a small team.
- Outcome
- Founded Keen Technologies with a tight focus on architecture-first research.
- Long-term impact
- Modeled how veteran engineers can re-enter a frontier field late.
Refusing to scale id by headcount
Risk · Medium- Why
- Believed quality scales sub-linearly with team size in creative software.
- Outcome
- id stayed small and influential rather than becoming a publisher.
- Long-term impact
- Proved 10-person teams could outship 200-person studios.
Publishing the .plan files
Risk · Low- Why
- Wanted to share what he was actually thinking with anyone willing to read.
- Outcome
- Created a generation of engineers who learned by reading his daily notes.
- Long-term impact
- Set the template for the modern technical devlog.
What Can You Learn?
Take the lesson, not just the story.
AI-distilled takeaways, sorted by who you are and what you're building toward.
Focus is a depletable resource.
Protect long blocks; multitasking corrodes the deep work that compounds.
Throw away the working thing to ship the better thing.
Architectural sunk costs kill more startups than market timing.
Read the papers, then reimplement from scratch.
Understanding without reimplementing is borrowed; it evaporates.
Open source compounds.
Giving away the engine grows the surface area faster than guarding it.
Optimize for shipped, not status.
Output beats prestige in the long compound.
Small teams outpace big labs when the problem is clear.
Headcount past a threshold buys politics, not progress.
Document what you're doing in public.
Your devlog is a recruiting funnel and a future-self diary.
Questions People Ask
Questions people ask about this journey.
The questions most people have after studying this life. Tap one — every answer is built from John Carmack's own timeline, decisions, books, and lessons on this page.
Continue Exploring
Don't stop here.
Adjacent journeys, a collection that frames the craft, and one pick from a different world.
Study Next
7 min read
Elon Musk
An obsessive engineer betting on rockets, electric cars, and the impossible — applying first-principles thinking at planetary scale.
Read Journey
8 min read
Jeff Bezos
A hedge-fund quant who quit Wall Street to sell books out of a garage and ended up rewiring global commerce, logistics, and cloud computing around one obsession: the customer.
Read Journey
11 min read
Jensen Huang
The Denny's busboy who bet thirty years on parallel computing — and turned a 1990s graphics card maker into the most strategically important company of the AI era.
Read Journey
7 min read
Naval Ravikant
An Indian-American founder who built AngelList into the operating system for early-stage startups, and whose tweets and podcasts have become a generational manual for wealth, judgment, and happiness.
Read Journey
Related Collection
Startup Founders
Builders who turned a stubborn idea into a company. Study the persistence, the pivot, and the boring middle that no pitch deck ever shows.
Open Collection Related Story
My first pull request was a typo. It changed my career.
Rafael M. · Open Source Maintainer
A two-character fix to a README in 2018. Six years and three hundred PRs later, the project pays half my rent and most of my confidence.
Read Story Random Discovery
Brian Chesky
Co-founder of Airbnb
Three air mattresses and a designer's eye that turned strangers into hosts — a founder who treats hospitality as a craft and product as the story.
Open Journey