Letter From The Chairman: Alpha 4.10, QDSM, and Squadron 42

CIG has published a new Letter From the Chairman, and this one is not just a victory lap around the hangar. Chris Roberts thanks players for their support, playtime, and feedback, then lays out where Star Citizen Alpha 4.10 fits on the road to Star Citizen 1.0, why instancing matters, how CIG views Star Engine, and why Squadron 42 is now targeting Q2 2027.

The letter opens with fresh player activity figures: Star Citizen reached 35,890 concurrent players at its peak this year. In the first half of 2026, CIG also saw a 7% increase in average daily players and an 11% increase in monthly active players compared to last year. More pilots in the black, more stress on the machinery, and therefore more reason to talk about the machinery.

Alpha 4.10, Siege of Orison, and the First Step Into Dynamic Server Meshing

According to Roberts, Star Citizen Alpha 4.10 is an important step toward Star Citizen 1.0 because it brings area and encounter instancing to the Persistent Universe through Siege of Orison.

Player hangars already use a basic form of instancing, but Alpha 4.10 introduces a broader version: the ability to stream arbitrary areas and their related content on demand, connect them seamlessly to the main shared universe, and manage them with dynamic server loading. Roberts describes instancing in 4.10 as the first use case of dynamic server meshing in Star Citizen.

Until now, each universe shard has run with a fixed set of servers, with each server assigned to specific areas that CIG internally calls territories. Roberts gives microTech as the example: microTech and everything on it is a territory watched over by a server. If too many players gather at microTech or New Babbage, the static system cannot spin up extra servers to divide that simulation load, so the single server responsible for that territory slows down and the player experience degrades. With dynamic server meshing, CIG intends to spin up additional servers as simulation load demands, subdividing territories into smaller ones.

For Alpha 4.10's instances, CIG monitors server load and, once a specified maximum load ratio is reached, spins up additional servers to manage extra instances dynamically. The Networking team refers to this as Quasi Dynamic Server Meshing, or QDSM, because it is not yet subdividing territories in the persistent universe itself. It is, however, the first case of Star Citizen spinning servers up and down dynamically based on load, and Roberts says it required many of the services needed for full dynamic meshing in the Persistent Universe.

Roberts also stresses that while other MMOs use instancing, Star Citizen's version has to interact seamlessly with a fully dynamic meshed shared persistent universe, which significantly increases the complexity. In other words, this is not just opening another room behind a loading door and calling it a day. Very tidy if it works; extremely spicy if it does not.

Siege Testing and a Return to Behind-the-Scenes Development

Roberts says that, with player help testing Siege of Orison, instancing is now performing well. He acknowledges that it took longer than expected and had more complexities and edge cases than CIG anticipated.

As part of the launch, CIG is also sharing a development video about the process of delivering Alpha 4.10. Roberts frames this as a step back toward more open development communication, saying that Star Citizen was made possible by backer support and that one of the reasons people supported it was CIG's willingness to share failures and successes along the way.

He says that over this year and last, CIG has shared less of the process of building Star Citizen, and that being more open helps players understand why things are not always perfect when the team is chasing what he calls the most ambitious game in gaming. The Alpha 4.10 video is positioned as the first step toward getting back behind the scenes and back to CIG's roots.

Star Engine, CryEngine, Lumberyard, and the Tech Stack

Roberts uses the letter to address what he calls misconceptions in community discussion, beginning with the idea that Star Citizen's technical issues are due to its engine choice.

He describes Star Engine as CIG's own engine, originally started as a branch from CryEngine 3. Since October 2015, CIG stopped taking updates and has instead refactored and added to the code base to the point that Roberts says almost nothing remains of the original engine. He lists the Entity System, Renderer, GPU Particle System, DNA and Facial animation system, Audio system, UI system, Network / Replication Systems, Physics, and all Gameplay Systems as entirely new and bespoke to Star Citizen and Squadron 42. He also notes that the Renderer now has hardware assisted Ray Tracing / Global Illumination for Squadron 42.

Roberts adds that a game engine is only a small part of what is required to run a massively multiplayer game like Star Citizen. CIG also operates dozens of distributed online services in the cloud, connected to multiple backend databases that must function as one synchronized system.

The move from CryEngine to Lumberyard, he says, was made for Amazon Web Services, where CIG ran its servers and platform for many years. Lumberyard began as a full branch of the same CryEngine branch CIG was based on, which meant CIG could switch with no changes on the core engine side after Amazon bought the rights to that version of CryEngine. Roberts says Amazon heavily modified CryEngine over the years into what people know as Lumberyard, and that there was good work there, but CIG did not take it because its tech roadmap has always been built specifically around Star Citizen and Squadron 42.

In 2020, CIG reached an agreement with Crytek that gave the company a perpetual license for the branch of CryEngine it was on, covering all current and future games. Roberts says this gave CIG full control over Star Engine going forward, which he frames as important given the company's technology investment.

Roberts states directly that there is no game engine that can do everything Star Citizen can do, and that there is no game that can do everything Star Citizen does at its scale and fidelity.

He attributes that to the demanding requirements of a seamless first-person universe where players can explore any corner. Every planet or moon is fully realized and circumnavigable on foot, by vehicle, or by ship. Vehicles operate as complex collections of components that let a ship fly and sustain the player, with components that can be physically interacted with, broken, damaged, repaired, or replaced. Players can move fluidly from on foot to a ground vehicle to a spaceship, take off, fly to a visible moon or spaceship, walk around inside that ship, land, get out, and explore the planet, all seamlessly, with no loading screens or delays. Roberts notes that some ships are the size of whole game levels in most FPS games.

Funding, Team Size, and the Cost of Building the Unusual

Roberts says that building the tech stack behind Star Citizen requires a major investment of both time and money. He says CIG's Core Tech Group is 146 strong, with 106 engineers. This team covers the engine, tools, and backend technology used by the project.

He argues that most developers do not have the resources to build this kind of bespoke technology, which is why many license Unreal or Unity and build games with smaller scopes. Addressing headlines around Star Citizen passing the one billion dollars raised threshold earlier this year, Roberts says the reality is that building something like Star Citizen requires funding at this level.

As a comparison, he points to Rockstar, the main developer for GTA6, saying it spent over two billion dollars in staff and office costs in the UK from 2019 through the end of 2025. He notes that, like CIG UK Ltd, Rockstar Games Ltd publishes financials through Companies House, making those costs public. Roberts says this is not meant as criticism of Rockstar and adds that he loves their games, ambition, and commitment to gameplay, fidelity, and scale. His point is that new technology and games players have not seen before take this kind of funding.

Star Citizen, Squadron 42, and the Myth of the Skeleton Crew

The second misconception Roberts addresses is the claim that Star Citizen has a skeleton crew while all resources and money are being spent on Squadron 42.

He says that, outside the Core Tech Group, CIG has a further 611 staff in development across four studios. For the last couple of years, development has been split approximately 50/50 between Star Citizen and Squadron 42. Before that, Roberts says the split was more heavily weighted toward Star Citizen.

Roberts says one reason Squadron 42 has taken as long as it has is that CIG repeatedly reassigned resources to Star Citizen to get patches out, which pushed Squadron back. After CitizenCon 2023 in LA, where CIG showed functionality being developed for Squadron, the majority of the gameplay engineering team spent a large part of 2024 bringing many of those features into Star Citizen at the expense of Squadron.

Toward the end of 2024, CIG decided that balance could not continue and shifted resources back. In the final months on Squadron 42, Roberts says the split is now 55% to 45% in Squadron's favour, so the game can be made bug free, polished, and optimized. He adds that finishing Squadron 42 sooner means the majority of development staff can then focus on bringing Star Citizen to 1.0, while also acknowledging how difficult it is to split time and resources between two hugely ambitious projects.

Squadron 42 Moves to Q2 2027

Roberts then turns to Squadron 42's release. At CitizenCon 2024, CIG announced that Squadron 42 would ship in 2026, with the plan to release in Q4 around IAE.

That plan has changed. Roberts says he does not want to launch his spiritual successor to Wing Commander, a game with twelve years of work behind it, into the attention buzz saw of GTA6. He notes that Squadron 42 development began in 2014, while 2013 was focused on opening studios, hiring people, and getting the Hangar module out for Star Citizen. While GTA6 is not launching on PC this November, Roberts says most outlets and content creators will be covering it heavily, leaving little room for other coverage.

After looking at early 2027 release windows, CIG has decided that Q2 is a better time for Squadron 42 to release.

CIG will allow a small number of original backers and long-time Star Citizen content creators to play Squadron 42 on October 9th and 10th. Roberts says further announcements about release plans will be made during that time, with a specific release date to follow shortly afterwards.

Roberts acknowledges that many players will be disappointed not to play Squadron 42 at the end of this year, but says the change is best for the project because CIG wants it to have the attention of the wider gaming world.

The Billion-Dollar Milestone and a Thank-You to Backers

Roberts closes by thanking backers for their support and addressing why there was not a Letter From the Chairman at the $1 billion crowdfunded milestone. He says he initially felt the moment should not be about the money, because Star Citizen and Squadron 42 are more than a financial headline, and points to CIG's voluntary financials as evidence that the company invests the money it raises back into development of both games.

On reflection, Roberts says he was looking at it the wrong way. He describes the billion dollars as a testament to the backers and community members who gave him and CIG their faith and patience, stood by through thick and thin, and held the line. He calls it proof of the power of a collective to make something unprecedented happen, saying no publisher or private equity investor would have had the same patience and willingness to invest long term.

Roberts ends by saying there are amazing things ahead, that Star Citizen already feels special, and that this is because of the community. He signs off by wishing players enjoyment in SC Alpha 4.10 and saying he will see them in the 'verse.

Source: Letter From The Chairman