Pinned Tweet
The “Miners Multiple”: #Bitcoin Market Cap vs. the Sum of all Mining Rewards ($). Identifies bubbles & dips with high signal to noise ratio. Current value: 11. Was never below 1. More signal than the Mayer Multiple @TIPMayerMultple @PrestonPysh @TraceMayer (1/3)
GeertjanCap retweeted
A propulsion engineer at SpaceX told Elon Musk that the engine he wanted to build had never been successfully made in the history of rocketry. Not by NASA. Not by the Soviet Union. Not by anyone. Elon's response changed the direction of the entire company.
The engine was the Raptor. Elon wanted it to use a full-flow staged combustion cycle, a design where both the fuel and the oxidizer are fully gasified before entering the combustion chamber. It's the most efficient and most complex rocket engine cycle ever conceived.
The Soviet Union spent decades and billions of rubles trying to make one work. They came close with the RD-270 in the 1960s but never flew it. The engine kept destroying itself during testing. After the Soviet program collapsed, most aerospace engineers considered the concept a dead end. Theoretically perfect. Practically impossible.
When Elon told his propulsion team he wanted to build one, an engineer laid out the history. Every failed program. Every exploded test article. Every reason the global aerospace community had moved on from the idea decades ago.
Elon listened to the entire presentation. Then he said: "Good. That means nobody holds the patents and nobody has a head start. We're starting from zero, same as everyone who tried before. Except we're not going to stop."
In 2019, SpaceX successfully test-fired the Raptor engine. It was the first full-flow staged combustion engine to ever reach a test stand and survive. It is now the engine that will power Starship to Mars.
The graveyard of impossible things is full of projects that stopped one attempt too soon.
GeertjanCap retweeted
SpaceX's CFO just shared a very interesting new metric. A gigawatt of ground-based compute costs the industry roughly $50B. Half goes to chips and memory, which cost SpaceX the same as everyone else. The other $25B goes to land, power, buildings, cooling, and labor.
In orbit, that second half largely disappears. The sun supplies power, deployable radiators handle cooling, and there's no land to buy. The remaining cost is launch. If SpaceX can get payload to orbit for less than $25B per gigawatt, space becomes cheaper.
SpaceX's latest compute satellite averages 175 kW (250 kW peak) at roughly 3.3 tons. That means about 19,000 tons of satellites to deliver a gigawatt. At $200–500/kg in 2028, launch costs would be $4–10B/GW, well below the $25B threshold. Break-even is around $1,300/kg, a very realistic cost target by then.
But even at break-even, orbital compute has another advantage: eliminating the enormous ongoing electricity costs and infrastructure hurdles of ground-based data centers means higher operating margins and faster deployments from day one, even before launch costs fall further.
That's why Bret calls the next few flights of Starship's reusable upper stage "the biggest catalyst." He correctly sees the tipping point just one or two years away. The first compute satellites are expected to fly next year, ramp in 2028, and deliver meaningful gigawatts by 2029.
And the economics only improve from there. Ground based costs keep rising with power and labor, while Starship launches get cheaper with higher flight rates, and mass production drives down the cost of both ships and satellites.
Exciting times ahead for SpaceX!
GeertjanCap retweeted
I see people trying to compare Space Shuttle reusability to Starship. The Shuttle’s external tank was discarded, but the Solid Rocket Boosters required extensive refurbishment. After splashdown, recovery ships towed the boosters back to Cape Canaveral. Crews washed off salt, separated the segments, and hydroblasted residual insulation.
The bare steel cases were then shipped by rail to Utah, where they were cleaned to bare metal, inspected, and proof-tested. New insulation and propellant were installed, along with nozzles and igniters, and the finished segments returned to KSC for stacking and reuse.
Steel cases were designed for about 20 reuses.
This was not quite the reusability NASA originally envisioned for the Space Shuttle. Also why SLS does not attempt SRB recovery.
Yeah, I was going over the whole Starship heat shield earlier today. Overall, it looks great. The engines look great too. Ship 42 could probably be reflown with a few weeks of refurbishment.
However, many small fixes are needed to achieve immediate reusability (like an airplane), which requires zero expected refurbishment, apart from scheduled maintenance.
GeertjanCap retweeted
IF YOUR CHILD SAYS "I'M STUPID," DO THIS IMMEDIATELY.
Not "no you're not."
Not "don't say that."
1. Pause and get down to their eye level.
Insurance for this is possible but expensive.
Once you have >6-7% illness insurance becomes prohibitive.
--> a real problem
More Madness ...
A Dutch company needs to pay a 'sick' worker for another two years. (In Germany, it is just 6 weeks) Even if he/she is sick for non-work-related reasons.
It's a huge risk, so Dutch companies try to hire as few staff as possible ..
306
GeertjanCap retweeted
BREAKING:
Scientists confirm that an ADHD brain can hyperfocus for 4 hours straight on a complex hobby—but will physically freeze for 3 days over a chore that takes 90 seconds to finish.
Researchers baffled.
ADHD community: "Yeah, we know."
No clinical citation needed—every parent and kid staring at a pile of shoes by the door is the study. Here is why tiny tasks paralyze an ADHD brain, and how to trick yourself (or your kid) into starting.1/6
GeertjanCap retweeted
Thanks to Elon, we just watched a Roadster fly.
For anyone still wondering how the Rocket trim does it, here's the breakdown:
Underbody suction for grip: Powered fans draw air out from under the floor so atmospheric pressure presses the car against the surface. Tesla says the system pulls up to 2 Gs, boosts cornering beyond a typical 1.6 G tire limit to >2 Gs, and can hold the car upside down at zero speed.
Compressed air for flight: Two tanks (600L total) hold ~250 kg of air feeding 10 nozzles. Four rear jets generate >80 kN of thrust, hitting 0 to 60 mph in 0.81s during the demo, while perimeter thrusters provide upward lift and midair attitude control until the tanks run dry.
Configurations
– Performance: road ready, seats 2
Active aerodynamics & multi-mode suction system for next-level performance on & off the track
– Thrust: race ready, seats 2
Multi-mode suction system paired w/ rear-mounted cold-gas thrusters that enable 0–60 in <1 second
– Rocket: air ready, seats 1
Multi-mode suction system plus perimeter-mounted cold-gas thrusters supporting forward & upward acceleration plus lateral rotational movement. So...yes, it can fly
Range >300 miles
Top speed up to >230mph (Rocket)
0-60
Performance (<1.6)
Thrust (<0.9)
Rocket (<0.59)
Reserve now –
Tesla.com/Roadster
GeertjanCap retweeted
This is the most powerful single steam-turbine train ever built for electricity generation.
Built at Belfort in France by Arabelle Solutions, two of these enormous machines are being installed at Britain’s Hinkley Point C nuclear power plant. Each can generate an immense 1,770 MW of electricity while turning at just 1,500 rpm, driven by roughly 2.4 tonnes of steam every second from a 4,500 MW thermal EPR reactor.
As that steam expands toward the near vacuum condenser, its density collapses and its volume becomes enormous. The final low pressure blades are therefore an extraordinary 1.9 metres long, so large that controlling their weight, centrifugal loads and vibration required new design and manufacturing processes.
Its massive welded rotors and high strength steels are engineered to withstand decades of centrifugal stress, thermal cycling, vibration and wet-steam erosion.
Together, the two turbines will deliver around 3.2 GW to the grid, enough to power 6 million UK homes and meet close to 10% of the entire UK’s electricity demand from a single nuclear plant.
Well performing nuclear fleets can exceed 90% availability, equivalent to roughly 330 full power days a year.
Now let's consider the physical scale.
Britain’s largest operating solar farm, Cleve Hill, uses more than 560,000 panels across roughly 900 acres. At typical UK solar output, it would take around 79 days to generate what Hinkley Point C nuclear reactor can deliver in 24 hours.
Hornsea 2 has 165 offshore wind turbines spread across 462 km² of sea. At typical UK offshore wind output, the entire farm would need roughly 6 days to generate the same amount of electricity.
Hinkley Point power plant's permanent development footprint is only 167 acres.
Two turbines, gigawatts of sustained output, from a remarkably small physical footprint. There is simply no other major power generation technology that demonstrates power density at this scale quite like nuclear does.


















































