Stars Reach Review: Is It Worth Buying in October 2026?

Buy or wait: the decision by player type

Most "is it worth it" pages end with "it depends". Here is what it depends on. Find your row in the table. The sections below give the evidence for each verdict.

What 504 Steam reviews actually say

The score has barely moved since September. At launch it was 51% of 255 reviews (read 24 August). By 24 September it was 59% of 490, and on 11 October it was 59% of 504: 297 positive and 207 negative, which Steam labels Mixed. Only 14 reviews arrived in those 17 days. That says as much about how few new buyers there are as about the game itself.

In launch week the negative reviews were mostly about performance and servers: rubber-banding, stuttering, lag and desync. Many also said the game had entered paid Early Access too early. The studio agreed in public. On 21 August Playable Worlds said it was not "where we want to be yet in terms of server stability and general game performance", and it shipped a broad hotfix the next day. On 4 September Raph Koster wrote that the team had "published nine updates in the last two weeks" and that "the game plays much more smoothly now". That is the studio's own claim. The flat review score suggests players who left have not come back to test it.

The criticism has since moved from "it lags" to "what am I meant to do, and is anyone here?". SmashPad's review (2 September 2026) is the only full review on Google's first page. Its headline calls Stars Reach "a lonely sandbox MMO". It criticises an onboarding that leaves you to look things up outside the game, combat that "felt like an idle game", and towns it describes as "desolate and devoid of life". The reviewer admits sandbox MMOs are not their genre. The specific complaints still match what players post on Steam.

Is Stars Reach dead?

No. The servers are up, the team is shipping, and the studio has publicly brushed off sunset fears. But the player base is very small, and the numbers should be stated plainly.

SteamCharts averaged 121.88 concurrent players in August 2026, 39.40 in September (−67.68%), and 23.12 over the 30 days to 11 October. Its all-time peak is 227. MMORPG.com reported a launch-week peak of about 697. The two trackers sample differently, so we quote both and do not try to reconcile them. Valve's own API showed 8 players online at 09:00 UTC on 11 October.

The studio is still working at a steady pace. Its build notes are dated 2 September for Ground Control 8.5.0 (live 3 September), 11 September for 8.5.1, 15 September for Better Together 8.6.0, 24 September for Crafted in Color 8.7.0, and 2 October for hotfix 8.7.2. Massively Overpowered reported on 3 September that the studio brushed off sunset fears as it released Ground Control. On 10 October (Pacific time) Raph Koster posted a developer note on Steam listing the work for the next two updates. It covers creature tactics and damage-type resistances in combat, Ranger upgrades, a Leadership profession, a new building control scheme, expanded storage, and GovBot changes.

Players are clearly worried. The most-replied Steam threads are about population, price and whether the studio will survive, not gameplay. A thread titled "89 concurrent players" drew 124 replies, and a Reddit post asking "Any news on the game sunsetting yet?" drew 44 comments. Those threads show what players fear, not what the studio plans. No source we can verify says the game is closing.

What changed since our August verdict

Our August verdict was "wait", mostly because of server stability. The verdict is still "wait", but for different reasons. The official build notes since then fixed several of the specific things new players got stuck on:

• Homesteading opened up (8.5.0, 3 Sep). Every player now gets one free Homesteader on first entering a Space zone. Fabricators are free and unlimited from the Homestead Console and the Tool Shop, with no durability.
• Character saving was hardened (8.5.0). The studio fixed save and persistence problems that made items vanish, plus another server memory leak behind lag that built up over time.
• Grouping arrived (8.6.0, 15 Sep). A real party system with /party commands, party chat and waypoints. 8.7.0 added a roster sidebar for up to 10 members.
• Crafting got simpler (8.6.0). Recipe Activation was removed. A learned recipe can be crafted whenever its qualifying skill is in practice. Hotfix 8.7.2 also stopped finish recipes from asking for more input than the previous step produced.
• A first Galactic Map (8.6.0), showing which sectors connect.
• Paints, dyes and finishes (8.7.0, 24 Sep). Twelve paint and dye colours from the starbase Tool Shop (50 uses for 1,000 Klaatus), craftable metal, stone, wood, plastic and rubber finishes, and close to 100 new paintable building pieces.
• A mission rework (8.7.0). Multi-step and branching mission chains, a simple carry limit of three Mission Board missions, and fewer missions failing on inhospitable planets.
• Terrain shields (8.7.0, on by default since 8.7.2 on 2 Oct). Homesteads and zoning lots now block terrain flowing into them. This addresses the griefing that a 93-reply Steam thread complained about.
• Creature melee returned and creature pathfinding improved (8.7.0). An in-game performance monitor (F9) now shows ping, packet loss and server load.

What genuinely works

The environmental simulation is the standout, and the Steam listing describes it as the reason the game exists: rivers flow and flood, forests burn, and mining a cave without ceiling supports can bury you. These are systems, not scripted scenes, and the changes persist on a shared server. Terraforming is powerful enough that the studio had to add terrain shields to protect homesteads, which shows how real it is.

The advancement design is the second. Playable Worlds' 2024 design post, by Dave Georgeson, set out the idea: experience for actions that are useful or risky, not for repetition. The Steam page still gives the example of a camp earning XP when other players rest at it. In the live build most trees also earn XP for doing their own work, such as crafting a recipe or dealing damage, and patches have added solo routes like Xenobiology XP for scanning creatures (8.4.0). Skills attach to tools, and skills you stop practising fall out of practice rather than disappearing. It is a real answer to the grind problem. The economy and governance still need other people, which is why population matters more here than in most games.

The third is the building game. Free Fabricators, a free first homestead, and since 8.7.0 a full finish-and-paint system make building the most complete part of the game today. The studio's own community events, such as the 3 October Cityscape Tour of player towns on four worlds, are built around it.

Click to play · loads YouTube — Stars Reach Early Access: Everything You Need To Know, by Controlled Chaos Gaming

Stars Reach vs No Man's Sky vs Star Wars Galaxies

These two comparisons come up most often, and they answer different questions. No Man's Sky asks "do I want to explore space?" Star Wars Galaxies asks "do I want a player-run society?" Stars Reach is aimed at the second question, even though its procedural planets look like the first.

Click to play · loads YouTube — The Next Star Wars Galaxies Is Finally Here! | Stars Reach First Impressions, by Potion Hoarder

Which comparison applies to you

If you came here from No Man's Sky, the honest answer is to keep playing No Man's Sky for now. It is finished, it is built for solo play, and its strengths do not depend on who else is online. Stars Reach turns the No Man's Sky loop around: the best gear is made by other players, planets are governed by groups, and the economy assumes a market. With about two dozen players online on average, much of that design has nobody to connect to.

If you came here from Star Wars Galaxies, the case is much stronger. The same lead is rebuilding the same ideas: classless skills, an economy where crafters matter, player towns with real governance, and progression that rewards being useful. The ideas have been improved with twenty years of hindsight. Atrophy lets you learn everything without keeping it all active, so a bad build is never permanent. The world also reacts in ways Galaxies' world never did: fire, flood, cave-ins, and populations that collapse when overhunted. Two things are missing. One is the Star Wars licence, which let you be an ordinary person in a universe you already loved. The other is the population a Galaxies-style society needs, which Stars Reach does not have yet.

In short: No Man's Sky is the better game today. Stars Reach is the more interesting project. Galaxies veterans are the one group for whom being early may be worth paying for.

Why waiting is unusually cheap here

Usually, waiting to buy a game means falling behind. Three commitments on the Steam store page remove most of that cost for Stars Reach:

First, a progression wipe before 1.0. The studio says "some player progress will be reset" before Early Access ends. Founders Pack items, Kickstarter backer items and other purchased or granted account entitlements stay tied to your account. Second, the game goes free to play at 1.0. Third, the studio plans to "sunset founders packs" at the same point, so the packs are the only thing that waiting really costs you.

Early Access is planned to last "at least 6 months" from 18 August 2026. The earliest possible free date is therefore around 18 February 2027, and no date has been announced. Our free-to-play guide covers the founder tiers, what each one adds and what survives the wipe.

What would change this verdict

We will re-read the reviews, SteamCharts and build notes against each of these and update the page when one of them happens:

1. The Steam score leaving Mixed, or recent reviews rising clearly above the all-time 59%.
2. Concurrency settling at a level, rather than falling month after month: August 121.88, September 39.40, last 30 days 23.12.
3. The combat rework Raph Koster described on 10 October shipping, and Steam threads moving on from combat complaints.
4. A published 1.0 date, wipe date or free-to-play business model, none of which exist yet.
5. Evidence of the interlocking systems working in practice, such as player towns, trade between planets and crafter reputations, at the population the game actually has.