Number Cruncher@NumberTableau
What it actually takes to power a 10+ GW chip mega-fab
Elon just dropped the reality check: Terafab (Tesla + SpaceX joint chip push) needs thousands of acres and over 10 GW of power at scale. Not hype—physics. This isn't another Giga; it's infrastructure on the level of a small nation's grid.
Let's run the numbers on what powering a ~10 GW continuous operation really requires if we're serious about near-zero carbon (Tesla/SpaceX style—no marketing fluff, just engineering).
Daily draw at 10 GW 24/7: 240 GWh
Annual: ~88 TWh
For scale:
- That's ~7× large nuclear reactors (~1.4 GW each)
- Or 10+ massive offshore wind farms (~1 GW each)
- Or tens of thousands of acres of solar
No single silver bullet exists. A real system needs diversity.
Realistic near-zero-carbon mix for 10 GW baseload reliability
- Solar: 24 GW nameplate (cheap, scalable; Texas sun helps) → ~59 TWh/year
- Wind: 12 GW → ~37 TWh/year (nighttime & complementary weather production)
- Nuclear: ~3 GW firm baseload → ~23 TWh/year (weather-independent stability)
- Batteries: ~9 GW / 90 GWh (shift solar peaks to night/evening)
- Backup: ~2.5 GW (gas/hydrogen turbines, rare use—<3% of total energy)
Total generation: ~119 TWh/year → covers the 88 TWh need + losses, bad-weather buffers, curtailment.
Why this mix?
- Solar dominates because it's the cheapest scalable source today.
- Wind cuts battery sizing dramatically (produces when solar doesn't).
- Nuclear (~25% of demand) provides the "always-on" backbone—no weather risk.
- Batteries time-shift, but they're not magic; they don't create energy.
- Minimal backup is engineering honesty, not failure—rare dunkelflaute events happen.
Land reality check
- Solar: ~120,000 acres (~485 km² / ~187 sq mi) — roughly a square of 22 km × 22 km (~14 mi × 14 mi)
- Wind: Spacing ~600,000–1,200,000 acres, but <5% actually occupied (land still farmable)
- Nuclear: ~2–4 km² (~0.8–1.5 sq mi)
- Batteries: ~1–2 km² (~0.4–0.8 sq mi)
Tiny nuclear footprint for huge reliable output. Solar/wind scale but sprawl.
Key takeaway for Terafab
Elon is right—this isn't just "build a factory." It's building an energy ecosystem at country scale. The challenge isn't generation alone; it's managing time (solar peaks vs. 24/7 chip demand) and firm power for uptime.
Tesla fans nail it: solar + batteries scale insanely. But real zero-carbon at this level needs wind diversity, nuclear stability, and tiny backup realism.
If Terafab hits 10+ GW powered mostly clean, it won't look like one tech. It'll look like an engineered symphony: solar for volume, wind for coverage, batteries for shift, nuclear for backbone.
That's not compromise. That's how you actually build the future Elon envisions—galactic civilization starts with solving terawatt-scale energy on Earth.
What do you think—will Terafab lean heavier on nuclear in Texas, or go all-in on renewables + massive Megapacks? 🚀⚡
