RLV C5 Rocket Concept Claims 74% Payload Edge Over Starship

A paper rocket takes on a flying one

Germany's national aerospace research center has published a heavy-lift rocket concept that it says uses its mass far more efficiently than SpaceX's Starship. That claim has spread widely through technology media this year. The design is called the RLV C5. It pairs a winged, reusable booster from the German Aerospace Center's (DLR) long-running SpaceLiner program with an upper stage that is thrown away after each flight. It is meant to put more than 70 tonnes into low Earth orbit.39 The headline number is a payload fraction of 74 percent for the European design against about 40 percent for Starship.56

The idea is clever, and the numbers are real outputs of a peer-reviewed study. But "more efficient" answers a narrower question than the headlines imply. And the Starship that DLR analyzed is no longer the Starship that is flying. To see why both points matter, it helps to look at what the study actually measured and how far SpaceX has moved since.

What the DLR team actually did

The study appeared in the CEAS Space Journal under the title "Comparison of SpaceX's Starship with winged heavy-lift launcher options for Europe." Its authors are Moritz Herberhold, Leonid Bussler, Martin Sippel and Jascha Wilken.3 Its most original part may be the Starship analysis itself, not the European rocket. The researchers did not take SpaceX's published specifications at face value. Instead, they pulled telemetry out of public webcast footage of Starship's first four integrated flight tests, second by second, and used it to build and check their own performance models.9 A text-recognition algorithm read the on-screen flight data.1

That reverse-engineering produced a fairly sober picture. The team estimates that the fully reusable Starship of those early flights could carry about 59 tonnes to low Earth orbit. That is roughly what a Falcon Heavy can lift when none of its boosters are recovered.36 For the next-generation vehicle, with Raptor 3 engines and larger tanks, the models project about 115 tonnes in reusable mode and up to 188 tonnes if the rocket is expended. That would beat the Saturn V.95

Where the efficiency comes from

The RLV C5 gets its advantage from two design choices. First, it burns liquid hydrogen and liquid oxygen, which gives more performance per kilogram of propellant than the methane-oxygen mix in Starship's Raptor engines.4 Second, it recovers its booster in a very different way. Starship's Super Heavy booster fires its engines to slow down and hover before the launch tower's mechanical arms catch it. That maneuver uses propellant that could otherwise go toward reaching orbit.1 The SpaceLiner-derived booster would instead glide back on wings, and a large subsonic aircraft would catch it in mid-air. No fuel has to be held back for landing.36

The bigger factor is that the European design does not try to reuse everything. Starship weighs more than three times as much as the RLV C5 at liftoff. Much of that extra mass pays for full reusability: heat-shield tiles, landing propellant, structural reinforcement and aerodynamic surfaces.9 Because the RLV C5's upper stage is expendable, it carries none of that hardware into orbit.

A statistic that is easy to misread

Outlets do not all describe the core figure the same way, and the difference matters. The original framing, carried by Phys.org and ScienceDaily, is that about 40 percent of what Starship delivers to orbit is useful payload, compared with 74 percent for the RLV C5.93 Some later write-ups shortened this to the rocket delivering "up to 74% of its mass as payload."1 Read literally, that would be physically absurd for any chemical rocket measured at liftoff. The number is a ratio of payload to everything that reaches orbit. It is not a ratio of payload to launch mass.

This matters for the comparison. A partly expendable rocket will almost always look better on this measure than a fully reusable one, because the reusable vehicle has to carry its own recovery equipment all the way to orbit and back. The metric shows a basic tradeoff. It does not show that one design is inherently better. Moving cost per kilogram to orbit, which is what customers pay for, also depends on how often hardware flies again and how cheaply it can be turned around. The study's payload-fraction figure does not settle those questions.

The DLR authors seem to understand this. Several reports say the researchers present the two vehicles as different answers to different problems, not as head-to-head rivals.23 They describe Starship's sheer capacity as suited to lunar bases, Mars missions and huge satellite constellations. They pitch the RLV C5 as a cheaper way for Europe to get its own super-heavy launch capability, and possibly as a step toward a fully reusable launcher later.910 Herberhold's team concludes that the RLV C5 "offers an effective path for Europe to independently develop partially reusable super-heavy launch capabilities."3

The timeline problem in the coverage

Reports disagree about when this work came out. BGR dates the DLR paper to May 2025.1 Phys.org covered it in March 2026 as newly published.9 Later stories in May and July 2026 described Europe as having "unveiled" the rocket.25 ScienceDaily's citation gives a 2025 publication with a volume-18 listing.3 Whatever the exact date, this is not a new announcement. A research paper has been picked up again in several waves of coverage, and the repeated "Europe unveils" framing makes it sound more official than it is. Nothing reported points to funding, an industrial prime contractor or a development program behind the RLV C5.

The concept's history also deserves weight. The SpaceLiner booster idea has been under study since 2005 and has never gone into production.1 BGR's skeptical take points out that a real program would need a new heat shield, new engines and proven gliding performance. Those were the same areas that slowed the Space Shuttle and drove up its cost.1

Meanwhile, Starship kept flying

The sharpest weakness of the comparison is that the Starship data is old. The DLR models rest on the first four integrated flights. Since then SpaceX has introduced its Block 3 vehicles with Raptor 3 engines. On boosters, those engines allow most of the engine shielding to be removed, saving more than a ton per engine.14

The 2026 flight record is mixed but moving forward. Flight 12 in May was the first Block 3 launch and the first from Starbase's second pad. The ship deployed its payload and made a controlled splashdown, but the booster could not relight enough engines and hit the Gulf hard.1116 Flight 13 in July deployed 20 Starlink V3 satellites, and for the first time the ship survived tipping over after splashdown with its heat shield intact. The booster again failed its landing burn.1115 Then on Flight 14 in late September, Starship reached orbit for the first time. It released 26 operational Starlink V3 satellites at around 275 km, and both stages came down under control. One upper-stage engine shut down early, and SpaceX cut the planned six-orbit mission short.11 Reports differ slightly on the date, with one listing September 28 and another September 29, and one describes the mission as only a partial success.1118

The next step is getting more ambitious. Elon Musk had said that a successful Flight 14 would clear the way for an attempt to catch the ship itself on Flight 15. He put the odds at "at least 50 or 60 percent," though SpaceX had not publicly confirmed that plan as of late September.11 The schedule is unclear. One tracker listed an October 9 target, while other listings point to a no-earlier-than date of October 19, or to November.1311 Prediction-market traders gave only about a 5 percent chance of launch by the end of October.19 SpaceX also says its next HLS milestones are a long-duration flight and an in-space propellant transfer test, both targeted for 2026.12

The verdict

The reasonable reading is that the DLR study matters more as an independent audit of Starship than as a European launcher plan. Its outside estimate of about 59 tonnes reusable payload for early Starships, rising to about 115 tonnes for later versions, is a useful check on SpaceX's own claims.3 The RLV C5 makes a fair point: Europe does not have to copy Starship's all-in reusability to get super-heavy lift.

Still, the 74-versus-40 comparison pits a rocket that does not exist against an old version of one that now reaches orbit with operational satellites on board. Europe's current flagship, Ariane 6, is fully expendable. Its newer launch companies are still at the stage of first flights and early tests.1 As a design philosophy, the RLV C5 looks credible. As a competitor to Starship, it would need money, a program and roughly a decade of work that no one has yet committed to. In that time, SpaceX will keep flying and improving its rocket.