News
How SpaceX Falcon Heavy undercuts its competition three-fold
Following the stunningly successful debut of SpaceX’s giant Falcon Heavy rocket, the spaceflight fan community and industry have been abuzz with attempts to estimate Falcon Heavy’s true price as an expendable or partially expendable launch vehicle. Thankfully, CEO Elon Musk appears to have been interested enough to fill in the knowledge gaps concerning the rocket’s full range of prices and took to Twitter to answer several questions.
Among several other intriguing comments that I will cover later on, Musk revealed that a fully expendable Falcon Heavy would cost approximately $150 million, while a partially expendable FH would sport 90% of the performance while expending the center stage and landing the side boosters at sea rather than on land. In that latter mode of operation, a Falcon Heavy launch would cost about $95 million, whereas unlocking the final 10% of performance with a fully expandable configuration would be priced around $150 million. While $90-150 million is undeniably a huge amount of cash in any sense, Falcon Heavy delivers far more performance for multiple times less than the available competition.
- ULA’s Delta IV Heavy rolls out to the pad for a launch in 2016. Note the people in the bottom left for a sense of scale. (ULA)
- ULA’s Delta IV Heavy rolls out to the pad for a launch in 2016. Note the people in the bottom left for a sense of scale. (ULA)
- The fully-integrated Falcon Heavy rolls out to Pad 39A. For vertical integration, think of this… but vertical. (SpaceX)
- DIVH and FH are approximately the same size, although FH is far denser. (SpaceX)
The only real competition for Falcon Heavy is the United Launch Alliance’s (ULA) Delta IV Heavy rocket, a triple-core launch vehicle with nine total launches under its belt since its 2004 debut. Aside from one test launch for NASA, all of DIVH’s operational flights have been tasked with launching uniquely heavy military payloads to uniquely high orbits – both of which require an exceptionally capable rocket. Designed as a fully expendable vehicle, ULA’s Heavy is capable of launching ~29,000 kg to low Earth orbit (LEO) and ~14,000 kg to geostationary transfer orbit (GTO), whereas the fully reusable Falcon Heavy has a max payload of about 23,000 kg to LEO and 8,000 kg to GTO.
However, if Musk’s claim of 10% performance loss as a partially expendable launcher holds true, the story changes quite a bit. In its fully expendable configuration (call it the Delta IV Heavy config), Falcon Heavy is a beast of a rocket, quoted at ~64,000 kg to LEO and 26,700 kg to GTO. Subtract 10-25%, and Falcon Heavy still trounces the Delta rocket, all while costing well under $150 million, and probably closer to $100 million. According to a late-2017 report from the US Government Accountability Office, Delta IV Heavy costs as much as $400 million per launch, although ULA CEO Tory Bruno responded to Musk’s claim of $400-600 million earlier this morning with a figure of $350 million for the rocket.
Hey @elonmusk , congrats again your heavy launch. Clarification: Delta IV Heavy goes for about $350M. That’s current and future, after the retirement of both Delta IV Medium and Delta II. She also brings unique capabilities, At least until we bring Vulcan on line.
— Tory Bruno (@torybruno) February 12, 2018
Such a high price is not exceptionally surprising, if only for the fact that Delta IV Heavy launches as infrequently as it does. With an average cadence of one launch every 18 months or 1.5 years, the technical expertise and facilities required to design, build, and operate the DIVH must remain employed regardless of whether the rocket launches. Although Delta was previously a family of rockets, thus enabling some of its designers and builders to cross-populate, the final non-Heavy Delta launch occurred just a handful of weeks ago. Short of layoffs, this means that ULA’s Delta expertise are now solely working to build and operate a rocket with approximately seven launches scheduled between 2018 and 2023 – in short, $400 million is quite plausibly on the low end of the rocket’s actual cost, backend included. Both ULA and the Department of Defense are aware, however, that Delta IV Heavy is the only rocket currently capable of launching some of the missions desired and required by the National Reconnaissance Office (NRO), and are thus at least partially willing to swallow the vehicle’s high cost. SpaceX’s Falcon Heavy is bound to introduce some much-needed competition into the stagnant market after its highly successful introduction, but it will likely be a year or more before the new rocket is certified to launch the same highly sensitive and expensive payloads as ULA’s Delta IV Heavy.
How are SpaceX’s prices so low?
Still, this does not answer the “how” of SpaceX’s prices. What can even begin to explain Delta IV Heavy’s 200-400% premium over Falcon Heavy? The best answer to this crucial question was by no coincidence also one of the main reasons that Elon Musk created SpaceX. From the very beginning, SpaceX pursued a slim and flexible organizational structure, prioritized hiring brilliant and motivated engineers with hands-on experience, and encouraged the practice of thinking from first principles. Dolly Singh, head of SpaceX’s talent acquisition in the mid-2000s, described the rocket startup’s atmosphere like so:
We searched for candidates with a proven history of building and breaking things…candidates who had been tinkering with hardware systems for years…I knew the people who filled my open positions would be put to the test every day and would be asked to meet heretofore impossible targets. We looked for people with a history of defeating the odds, who had made careers of overcoming obstacles.
https://www.instagram.com/p/BVarZZSgfIP/
Birds of an organizational feather
In essence, this organizational philosophy has led SpaceX to become vertically integrated to the extent that is effective without comparison in the global aerospace industry. Vertical integration is a term used to describe the practice of bringing aspects of development and manufacturing in-house, whereas a company not attempting to integrate vertically would instead contract and subcontract out their design and manufacturing needs wherever possible. Musk is hard set on this philosophy: if SpaceX can do it in-house more cheaply than a contractor, they will become their own supplier. Companies like ULA – a cooperation between Lockheed Martin and Boeing – have the better part of a century of experience as heavyweights in the US military-industrial complex, a relationship that has quite literally changed processes of acquisition and created alternate realities of pricing.
Thick with armies of lobbyists, those military-industrial complex titans have help to direct the US down a path that has solidified truly insane concepts as the status quo. A cost-plus contracting framework almost universally applied in the procurement of military technology means that companies are nearly awarded for delays and cost overruns. Possibly even more absurd, the euphemistic strategy of “concurrency” espoused by those same titans has somehow convinced the upper echelons of US defense procurement that it is a good and preferable strategy to fully fund and build technologies en mass before any testing has been. Unsurprisingly, these two philosophies have led to years of delays and huge cost overruns as contractors and their subcontractors are forced to repair or modify extremely complex technological systems once bugs and problems are inevitably discovered down the road. The F-35 Lightning II – developed by Lockheed Martin – is perhaps the most famous example with near-weekly tales of abject failure – gun systems that are years late and inaccurate to the point of uselessness, extremely buggy and flawed software that the jet literally cannot function without, an oxygen system that frequently gives its pilots hypoxia and grounds the entire F-35 fleet, among dozens of other incredible missteps – and all for the most expensive fighter aircraft yet developed in the US. Tyler Rogoway, one of the best practicing defense journalists, has covered the debacle of concurrency and cost-plus contracting for many years and is a recommended read for anyone interested in the above industries.
- While it may look damn cool, the F-35 is easily the biggest government procurement debacle in all of human history. (Lockheed Martin)
- Operated by the same company responsible for the F-35, Atlas 5 is a highly reliable and equally expensive rocket. (ULA)
Now, back to spaceflight…
Parting from this partial diversion, the purpose of this brief history of military procurement is to provide some level of context as to why NASA and its spaceflight contractors act as they do, where they derived their organizational structures and philosophies, and why SpaceX is different.
Famously, a NASA study in 2010 estimated the cost of SpaceX’s Falcon 9 development to be approximately $4 billion under variables representative of NASA’s own R&D and engineering culture, or $1.7 billion using a more commercial, fixed-cost strategy. When SpaceX offered to cooperate with the addition of their internal data on Falcon 9’s cost, the same model’s estimate plummeted to less than $600 million, representing a truly extraordinary overestimate of SpaceX’s development costs, while SpaceX’s data showed approximately $300 million of investment in the first version of Falcon 9. Simply put, NASA’s cost estimates were off by more than an order of magnitude (PDF) – SpaceX successfully developed an unprecedented orbital-class rocket for mere pennies to NASA’s dollar.
Famously, a NASA study in 2010 estimated the cost of SpaceX’s Falcon 9 development to be approximately $4 billion, while SpaceX’s own data showed approximately $300 million of investment in the first version of Falcon 9. Simply put, NASA’s cost estimates were off by more than an order of magnitude.
More recently, Elon Musk has stated that SpaceX invested $1 billion or more in the development of reusability for Falcon 9, and this large investment can almost entirely explain why Falcon 9’s pricing has remained essentially unchanged over its seven years of life, even if it was already the cheapest rocket in its performance class. Despite the recent introduction and rapid routinization of operational reuse, SpaceX has not publicly changed the launch price from its $62 million base. Although there have been slight acknowledgments of small discounts from customers flying on reused boosters, the general theme is that reused rockets have not meaningfully lowered the cost of purchasing a launch. In practice, the cost of refurbishment and reuse of the first several Falcon 9 boosters was likely on par with the cost of a new booster, but the real reason for the lack of magnitudes of cost reduction lies in SpaceX’s desire to recoup some or all of the capital it invested in reusability. As the company matures its reuse expertise, the cost can be expected to plummet – Cargo Dragon’s reuse, for example, reportedly saved SpaceX 50% of the cost of a new capsule, and Falcon 9 is almost certainly far easier and thus cheaper to refurbish and refly.
The quote is from a 2015 hearing held by the Armed Services Committee: "I don't know how to build a $400 million rocket. Rather than ask how am I less expensive than ULA, I don't understand how ULA is as expensive as they are."
— Robin Seemangal (@nova_road) February 12, 2018
While payload fairings have turned out to be harder to recover than anticipated and Falcon 9’s second stage is likely to remain expendable for the foreseeable future, those components only comprise about 30% of the rocket’s price. If SpaceX can cut the cost of reuse to maybe 10-20% of the cost of a new booster, the remaining 30-60% of a new launch’s $62 million translates to approximately $20-35 million of profit for each reused launch. If, say, the company aims to fly flight-proven boosters on half of their launches in 2018, that translates into as many as 15 launches and as much as $500 million – or half of the $1 billion investment – recouped in a single year. With the introduction of Falcon 9 Block 5 in a few months, SpaceX will soon be flying an iteration of their workhorse rocket that is far faster, easier, and cost-effective to reuse. Ultimately, depending on how much of their initial investment SpaceX intends to recover, the huge profit margins they can derive from reuse could be redirected to drastic price cuts for the customer. More realistically, the company will likely lower its prices enough to ensure that their launch business is brutally competitive, and thus use those profit margins to begin heavily investing in BFR (Big F. Rocket), BFS (Big F. Spaceship), and the company’s loftier interplanetary goals more generally.
- Starship and a Martian city, circa 2017. (SpaceX)
- SpaceX’s 2017 BFS (now Starship) delivers cargo to a large lunar base. (SpaceX)
In fact, given that SpaceX President Gwynne Shotwell has quite consistently targeted early 2019 for the beginning of prototype BFS testing, SpaceX is probably already putting a significant proportion of their profits into Mars-focused R&D. As 2018 progresses, barring any unseen speed bumps, the funds available to SpaceX are bound to explode, and huge progress will likely begin to be made on actual hardware intended to enable colonies on the Moon and Mars.
Follow along live as launch photographer Tom Cross and I cover these exciting proceedings as close to live as possible.
Teslarati – Instagram – Twitter
Tom Cross – Twitter
Eric Ralph – Twitter
News
Tesla makes big Full Self-Driving change to reflect future plans
Tesla made a dramatic change to the Online Design Studio to show its plans for Full Self-Driving, a major part of the company’s plans moving forward, as CEO Elon Musk has been extremely clear on the direction moving forward.
With Tesla taking a stand and removing the ability to purchase Full Self-Driving outright next month, it is already taking steps to initiate that with owners and potential buyers.
On Thursday night, the company updated its Online Design Studio to reflect that in a new move that now lists the three purchase options that are currently available: Monthly Subscription, One-Time Purchase, or Add Later:
🚨 Check out the change Tesla made to its Online Design Studio:
It now lists the Monthly Subscription as an option for Full Self-Driving
It also shows the outright purchase option as expiring on February 14 pic.twitter.com/pM6Svmyy8d
— TESLARATI (@Teslarati) January 23, 2026
This change replaces the former option for purchasing Full Self-Driving at the time of purchase, which was a simple and single box to purchase the suite outright. Subscriptions were activated through the vehicle exclusively.
However, with Musk announcing that Tesla would soon remove the outright purchase option, it is clearer than ever that the Subscription plan is where the company is headed.
The removal of the outright purchase option has been a polarizing topic among the Tesla community, especially considering that there are many people who are concerned about potential price increases or have been saving to purchase it for $8,000.
This would bring an end to the ability to pay for it once and never have to pay for it again. With the Subscription strategy, things are definitely going to change, and if people are paying for their cars monthly, it will essentially add $100 per month to their payment, pricing some people out. The price will increase as well, as Musk said on Thursday, as it improves in functionality.
I should also mention that the $99/month for supervised FSD will rise as FSD’s capabilities improve.
The massive value jump is when you can be on your phone or sleeping for the entire ride (unsupervised FSD). https://t.co/YDKhXN3aaG
— Elon Musk (@elonmusk) January 23, 2026
Those skeptics have grown concerned that this will actually lower the take rate of Full Self-Driving. While it is understandable that FSD would increase in price as the capabilities improve, there are arguments for a tiered system that would allow owners to pay for features that they appreciate and can afford, which would help with data accumulation for the company.
Musk’s new compensation package also would require Tesla to have 10 million active FSD subscriptions, but people are not sure if this will move the needle in the correct direction. If Tesla can potentially offer a cheaper alternative that is not quite unsupervised, things could improve in terms of the number of owners who pay for it.
News
Tesla Model S completes first ever FSD Cannonball Run with zero interventions
The coast-to-coast drive marked the first time Tesla’s FSD system completed the iconic, 3,000-mile route end to end with no interventions.
A Tesla Model S has completed the first-ever full Cannonball Run using Full Self-Driving (FSD), traveling from Los Angeles to New York with zero interventions. The coast-to-coast drive marked the first time Tesla’s FSD system completed the iconic, 3,000-mile route end to end, fulfilling a long-discussed benchmark for autonomy.
A full FSD Cannonball Run
As per a report from The Drive, a 2024 Tesla Model S with AI4 and FSD v14.2.2.3 completed the 3,081-mile trip from Redondo Beach in Los Angeles to midtown Manhattan in New York City. The drive was completed by Alex Roy, a former automotive journalist and investor, along with a small team of autonomy experts.
Roy said FSD handled all driving tasks for the entirety of the route, including highway cruising, lane changes, navigation, and adverse weather conditions. The trip took a total of 58 hours and 22 minutes at an average speed of 64 mph, and about 10 hours were spent charging the vehicle. In later comments, Roy noted that he and his team cleaned out the Model S’ cameras during their stops to keep FSD’s performance optimal.
History made
The historic trip was quite impressive, considering that the journey was in the middle of winter. This meant that FSD didn’t just deal with other cars on the road. The vehicle also had to handle extreme cold, snow, ice, slush, and rain.
As per Roy in a post on X, FSD performed so well during the trip that the journey would have been completed faster if the Model S did not have people onboard. “Elon Musk was right. Once an autonomous vehicle is mature, most human input is error. A comedy of human errors added hours and hundreds of miles, but FSD stunned us with its consistent and comfortable behavior,” Roy wrote in a post on X.
Roy’s comments are quite notable as he has previously attempted Cannonball Runs using FSD on December 2024 and February 2025. Neither were zero intervention drives.
Elon Musk
Tesla removes Autopilot as standard, receives criticism online
The move leaves only Traffic Aware Cruise Control as standard equipment on new Tesla orders.
Tesla removed its basic Autopilot package as a standard feature in the United States. The move leaves only Traffic Aware Cruise Control as standard equipment on new Tesla orders, and shifts the company’s strategy towards paid Full Self-Driving subscriptions.
Tesla removes Autopilot
As per observations from the electric vehicle community on social media, Tesla no longer lists Autopilot as standard in its vehicles in the U.S. This suggests that features such as lane-centering and Autosteer have been removed as standard equipment. Previously, most Tesla vehicles came with Autopilot by default, which offers Traffic-Aware Cruise Control and Autosteer.
The change resulted in backlash from some Tesla owners and EV observers, particularly as competing automakers, including mainstream players like Toyota, offer features like lane-centering as standard on many models, including budget vehicles.
That being said, the removal of Autopilot suggests that Tesla is concentrating its autonomy roadmap around FSD subscriptions rather than bundled driver-assistance features. It would be interesting to see how Tesla manages its vehicles’ standard safety features, as it seems out of character for Tesla to make its cars less safe over time.
Musk announces FSD price increases
Following the Autopilot changes, Elon Musk stated on X that Tesla is planning to raise subscription prices for FSD as its capabilities improve. In a post on X, Musk stated that the current $99-per-month price for supervised FSD would increase over time, especially as the system itself becomes more robust.
“I should also mention that the $99/month for supervised FSD will rise as FSD’s capabilities improve. The massive value jump is when you can be on your phone or sleeping for the entire ride (Unsupervised FSD),” Musk wrote.
At the time of his recent post, Tesla still offers FSD as a one-time purchase for $8,000, but Elon Musk has confirmed that this option will be discontinued on February 14, leaving subscriptions as the only way to access the system.








