Playtest 0 · 3 October 2026, afternoon

Smoke test

First internet lobby ever. Four friends, one Oregon server, three maps in a row. Everyone joined first try; nobody crashed; the mantle didn't work online and other players' bodies looked wrong. Both got fixed before the evening.

Build
349ccb2
Lobby
A · US West (Oregon)
Lobby time
43 min
Maps
4 in rotation
4testers
4at once, peak
3matches
520kills
172by humans
23human-on-human
7,866shots fired
24 %accuracy
49revives
0 · 0crashes · kicks

Humans in the lobby: 28 min with 2 or more, 20 min with 4 or more.

Classes

Who did the killing

Share of human kills by which class's gun did it. The pick screen doesn't log picks yet, so this is the honest proxy for what people played.

Share of human kills by class: Skinny 32.5 % · Mid 26.5 % · Thicc 41.0 %Skinny: 32.5 % of human kills · 54 kills · 3,042 shotsSkinny 32 %Mid: 26.5 % of human kills · 44 kills · 2,212 shotsMid 26 %Thicc: 41.0 % of human kills · 68 kills · 2,578 shotsThicc 41 %
  • Skinny54 kills · 3,042 shots · 39 % of shots
  • Mid44 kills · 2,212 shots · 28 % of shots
  • Thicc68 kills · 2,578 shots · 33 % of shots

Weapons

The gun table

Shotguns and the sniper live at the top of kills-per-shot by design; the three main rifles are judged on accuracy and kills per hundred shots.

Weapon statistics for Smoke test, human players only
WeaponClassShotsAccuracyHeadshotKillsKills / 100RangeShare of kills
ChauchatThicc2,34623.4 %10.4 %642.719 m
AK-69Mid2,09324.0 %7.6 %422.014 m
PEW-90Skinny2,98025.4 %14.4 %411.413 m
Double BarrelSkinny6245.2 %0.0 %1321.04 m
AYO-12Thicc21810.1 %13.6 %31.48 m
MID-16 BurstMid9626.0 %0.0 %22.124 m
Pocket Pistoleveryone3411.8 %0.0 %12.99 m
Desert BeagleThicc1442.9 %0.0 %17.134 m
BoomstickMid2339.1 %33.3 %00.09 m
Noscope 3000Skinny0——0—— m

Human shots only. Accuracy = hits the server confirmed ÷ shots. Headshot = share of hits. Kills / 100 = kills per hundred shots. Range = typical hit distance. Other kills: Explosions 4 · Melee 1.

Abilities& gadgets

Not counted yet. The server doesn't log ability or gadget use in this build; the next one does, and this section fills in.

The server

How the box held up

One cloud machine in US West (Oregon) running the lobby headless at 60 ticks a second. Ping is measured from every connected player every ten seconds.

Lobby B (launchers off) was up the whole time and nobody joined it.

  • 60 Hzserver tick, never below 58; slowest frame typically 25 ms
  • 45 msmedian ping · 9 in 10 under 77 ms · 99 in 100 under 89 ms
  • 12 KB/sper player each way during play
  • 0 %packet loss almost all session
  • 95 %of shots: client and server agreed on the hit (lag compensation rewind ≈ 65 ms)
  • 0 · 0crashes · kicks, across 17 joins

Notes

What we found. What shipped.

Found

  • Mantling into windows kept getting cancelled online — the server never agreed the climb had started.
  • Other players looked frozen or had their gun pointing at the floor.
  • Meme Launcher: zero kills with launchers on.
  • The Chauchat took 37 % of human kills — the first balance signal.

Shipped after

  • Mantle carried in the saved move so the server replays it (fixed by the evening build).
  • Remote bodies get a replicated movement base, so they animate again.
  • Title and tagline set: Pew Pew Pandemonium — The Actually Good Meme Shooter.

Matches

  1. SchoolTime's up
  2. MallTime's up · 13 min
  3. SlumScore limit · 6 min

Counted from the Lobby A server log, build 349ccb2. Weapon damage values come from the game's weapon table. No gameplay data about individual players is published anywhere. The numbers as JSON.