2026-09-03T00:00:00Z 0hphttps://0hp.com/ https://0hp.com/games/eve/feed.xml https://0hp.com/news/eve-online-hud-rebuild-opt-in/ 2026-09-03T00:00:00Z 2026-08-29T00:00:00Z

EVE's second design blog splits its central dial apart, adds a minimap and slots for consumables, and promises an opt-in phase before anybody is moved onto it.

<p>EVE Online, which is a long-running science-fiction game about spaceships, is taking apart the most established screen in it. And <a href="https://store.steampowered.com/news/app/8500/view/1842212951296996">its second design blog</a> is unusually explicit about the order. An opt-in phase, then a long opt-out phase, then a full switch only once adoption is high. A first version is aimed at the next expansion, with no date attached.</p> <p>The screen in question is the HUD, which is the set of bars and buttons drawn over the game. At its centre is the medallion, a round dial carrying your ship's speed, its health, its heat, and its capacitor, which is the energy pool that powers guns, shields and engines. A module is a piece of equipment fitted to a ship and switched on and off in combat. To overload one is to run it past its safe limit for extra power and some damage. And the minimap is the small radar view showing what is near your ship.</p> <p>The studio is Fenris Creations, which is the company behind the game. It took that name when it <a href="https://www.eveonline.com/news/view/a-new-era">announced on 6 May 2026 that it was becoming independent again</a>.</p> <h2>What changes, and in which release</h2> <table> <thead> <tr> <th>Element</th> <th>Change</th> <th>Stage</th> </tr> </thead> <tbody> <tr> <td>Speed</td> <td>Moves out of the medallion into its own controls</td> <td>First</td> </tr> <tr> <td>Health</td> <td>Shield, armour and hull stay around the capacitor, with clearer proportions and possible ticks</td> <td>First</td> </tr> <tr> <td>Capacitor</td> <td>Keeps its circle, gains a gauge showing how fast energy is gained or lost</td> <td>First</td> </tr> <tr> <td>Rack heat</td> <td>Moves beside the medallion, pointing at the rack it belongs to</td> <td>First</td> </tr> <tr> <td>Overload (running a module past its safe limit)</td> <td>Becomes its own interaction; Shift-click is preserved</td> <td>First</td> </tr> <tr> <td>Modules</td> <td>Clearer active and inactive states, a moving cycle indicator, discoverable shortcuts</td> <td>First</td> </tr> <tr> <td>Minimap</td> <td>On-grid, metro and system views</td> <td>Second</td> </tr> <tr> <td>Boosters</td> <td>Dedicated consumable slots with shortcuts</td> <td>Third</td> </tr> <tr> <td>Customisation</td> <td>HUD as movable components, hideable parts</td> <td>Fourth</td> </tr> </tbody> </table> <p>One row there needs a word: boosters, which are consumable drugs that raise a pilot's stats for a while.</p> <h2>Our take: what it means for players</h2> <p>The rollout plan is better than the redesign, and the redesign looks good.</p> <p>An opt-in phase, then an opt-out phase the studio itself expects to run long, then a switch only once adoption is high. That is how you change an interface people have spent years learning without taking those reflexes away from them. Compare it with the industry norm of shipping a new screen on patch day and learning from the forums afterwards. Nobody's Friday night fleet gets ruined by a redesign they did not ask for, and the studio still gets its modernisation.</p> <p>The research is credited where it belongs, and that should be noted. New players were brought in to be watched. Veterans were interviewed. There were sessions with the player-elected council the blog calls the CSM, plus a workshop and a booth at Fanfest, which is the game's own convention. The single most useful finding in the whole blog came from watching newcomers fail to realise the security status element was a button at all. A designer does not find that by thinking harder.</p> <p>Our one reservation is sequencing. Customisation is the thing veterans have asked for longest, and it is the answer to the screen space a minimap and consumable slots will eat. It is explicitly last. Screen space is a real cost in a game played in several windows at once, and shipping the parts that take room before the parts that let you hide them is the wrong order for the people most likely to opt in early.</p> <h2>What's next</h2> <p>A third blog is promised, covering what comes after this work. The first round of HUD changes is targeted at the next expansion with no announced date, so nothing here joins <a href="https://0hp.com/calendar/">the calendar</a> yet.</p> https://0hp.com/news/eve-online-python-3-migration/ 2026-09-03T00:00:00Z 2026-08-28T00:00:00Z

EVE has started moving off Python 2.7 onto Python 3. The measured scale: 3,300 lines that stop the build, and 20,000 more that work fine and behave differently.

<p>EVE Online, which is the long-running space game Fenris Creations makes, is being moved off Python 2.7. That is sixteen years after it last changed language version, and six years after that version stopped being supported. <a href="https://store.steampowered.com/news/app/8500/view/1841579228677996">The studio's own blog</a> puts real numbers on the job, which is why it is worth reading rather than skipping as housekeeping.</p> <p>Python is the programming language the game is largely written in: the blog counts 2.4 million lines of it. Version 2.7 reached the end of its life in 2020, so new tools are built for Python 3 instead.</p> <p>A note on the studio's name, because it changed this year. <strong>Fenris Creations</strong> is the company behind the game. The blog credits it as one of the most significant contributors to Stackless Python, the version of the language EVE runs on. The name comes from <a href="https://www.eveonline.com/news/view/a-new-era">its own announcement of 6 May 2026</a>, which says the company behind EVE is once again becoming independent. We use the name the sources use.</p> <h2>The measured size of the problem</h2> <table> <thead> <tr> <th>Measure</th> <th>Number</th> </tr> </thead> <tbody> <tr> <td>Lines of Python in the codebase</td> <td>about 2.4 million</td> </tr> <tr> <td>Files scanned under both versions</td> <td>about 20,000</td> </tr> <tr> <td>Files that already worked under both</td> <td>95.9% on the first scan</td> </tr> <tr> <td>Lines using syntax Python 3 rejects</td> <td>about 3,300</td> </tr> <tr> <td>Old-style print statements</td> <td>about 1,500</td> </tr> <tr> <td>Long number literals such as 123L</td> <td>about 800</td> </tr> <tr> <td>Deprecated exception clauses</td> <td>about 600</td> </tr> <tr> <td>Uses of the ancient not-equal operator</td> <td>50</td> </tr> <tr> <td>Lines that work under both and behave differently</td> <td>about 20,000</td> </tr> </tbody> </table> <p>That last row is the whole project. The example given is division. In Python 2, one divided by two is zero. In Python 3 it is 0.5. When the numbers involved are damage, currency or coordinates, each of those lines needs a person to decide, not a tool.</p> <h2>Our take: what it means for players</h2> <p>Nothing, this year, and that is exactly why we are writing about it.</p> <p>A twenty-three year old world running on a dead language version is a slow-motion risk to everybody who has a character in it. Every year on Python 2 is another year of libraries falling out of reach, and more of the toolchain maintained in-house instead of spent on the game. Doing the migration is unglamorous, expensive and invisible when it works, which is why studios put it off until something breaks. Starting before something breaks is the correct call and deserves saying out loud.</p> <p>The measurement discipline is the part worth copying. Running every file under both versions to find out how far the codebase actually is, rather than estimating, turned an unbounded fear into a countable list: 3,300 lines to fix mechanically and 20,000 to read. A studio that publishes those numbers is a studio that can be held to them.</p> <p>Credit belongs to the players here as well as the engineers. The first stage was tested in late July on Singularity, the game's public test server, by the players, whom this game calls capsuleers, who volunteered to play a version of the game whose whole ambition was to feel identical. The studio thanks them and says they are the ones helping it get there. That is unpaid work protecting everybody else's assets, and it is worth naming.</p> <p>The honest caveat is that stage one was the easy part by the studio's own account, and no completion date is given for the rest.</p> <h2>What's next</h2> <p>More test rounds on Singularity, with no dates announced, and the agent mission backend being prepared alongside them. Nothing here is dated, so nothing joins <a href="https://0hp.com/calendar/">the calendar</a> until the studio names a day.</p>