Emberveil was last checked on 25 August and we recorded its Android build as "closed beta signups open, invites pending". That was true then and it understates what happened next. The client now runs.
What the devlog says, newest first
15 September, "One Month of Emberveil", celebrating the realm's first month: 340+ bug reports closed, 97 client revisions, a nine hour DDoS weathered, and "put the client on a phone".
5 September, "We put Emberveil on a phone. Now we need it broken." The client runs on Android on its own test realm, with closed beta applications open, invites going out in waves and the first wave deliberately small. The title is the whole development philosophy in a sentence: they shipped it to a device they do not own and asked for it to break.
20 August, addon directory open. Submissions are reviewed before publishing, and "the download always comes from the same repository as the code". That is a supply-chain detail most projects in this scene do not make.
19 August, Lua API documentation live, and the post notes it is "useful even if you are building an addon for the original 1.12.1 game", which is a deliberate nod to the wider Classic addon ecosystem rather than a walled garden.
15 August, launch party with Out Of Bounds Radio, five DJs and the station playing in game from the moment the realm opened.
13 August, client download open with character nickname reservations.
6 August, writing contest, Season 1 titled "Echoes of the Ashes", asking for stories about life in Azeroth after the Third War, with winning entries becoming part of the official lore and the best ideas potentially inspiring future quests, NPCs and world events.
Why this one is different from most entries in this tracker
Most of what we track is a feature list. This is a devlog with dates, numbers and a stated intent, and two of those numbers are the kind of thing operators normally omit.
The nine hour DDoS is one. A month-old realm taking a nine hour attack and reporting it in a celebratory post rather than burying it is either confidence or inexperience, and the same post reports 340 bug reports closed against 97 client revisions, which is a ratio that suggests they are shipping faster than they are stabilising.
The writing contest feeding into official quests is the other. It is a stated intention, not a shipped feature, and we have no way to check whether quests actually came from it. Worth watching rather than crediting.
The underlying claim
The site's own framing: classic patch 1.12.1 running inside a fully functional Unreal Engine 5 client, "the thing people swore couldn't be done". Same quests, same dungeons, same roads, on an engine twenty years younger.
That is the project's whole proposition and it is a UE5 rebuild, which makes Emberveil the most interesting engineering entry we track. It also means it sits in a category where our evidence is weakest: we are describing what the site says about a technical achievement, and we have not run the client. The launcher is a 7z archive and we did not download or execute it.
Also live alongside the devlog: a Dungeon Finder and Group Browser, a database browser at its own subdomain, an armory, an addon wiki, and locale paths for Spanish, Russian and Simplified Chinese.
What has not changed
The devlog's newest post is dated 15 September, which is 18 days ago. Nothing has shipped publicly in that window, and for a project whose headline is a monthly engineering pace, an 18-day silence is the first thing worth watching.
Our population tier stays tiny on the basis of the site's own positioning as a small realm. We have no live counter from it.
SOURCES: Emberveil's own homepage and /news devlog, read October 3 2026 in a browser with a control domain probed in the same batch. Every date, post title, count and quoted phrase above is transcribed from those pages. We did not download the launcher, did not execute anything, and did not play. Related: how we review a private server, the correction policy, and the grading scale.