mirror of
https://github.com/jcreek/CosmicClash.git
synced 2026-09-10 16:04:04 +00:00
5bbb319161
New InputLeadController (scripts/input_lead_controller.gd, standalone and unit-tested like input_jitter_buffer.gd): fast attack (+3 immediately, debounced to once per 30 ticks) on any server-reported starve, slow release (-1 per 60 ticks, gated behind a one-time 2s clean-surplus bar) otherwise, clamped [1, 12]. Deliberately the only thing that adapts buffer depth - the server (InputJitterBuffer) stays a pure reporter, per §3.3's explicit warning that multiple control loops acting on one plant (buffer occupancy) oscillate and present as unattributable sticky controls. Wired into the client's per-tick input send: a lead change is realized as extra distance between the client's outgoing sequence numbers and what the server has consumed - an attack skips extra sequence numbers, a release duplicates the current one (sent again, unincremented). The server's ring buffer needs no special handling for either: a skipped seq is an ordinary drop, a duplicated one is a same-seq resend already discarded by the existing "already consumed" check. Verified with real two-process runs: on a clean LAN, one early attack (a momentary hiccup during connection setup) recovers via two releases within the test's own ~4s window, settling back near minimum. Under sustained 30% simulated loss, lead climbs to 7 via repeated attacks and never releases while genuine loss continues - confirming the debounce, attack, and release gates all fire on real conditions, not just in isolated unit tests. Full regression suite, including the net-sim-latency milestone gate, re-run clean.
81 lines
3.7 KiB
GDScript
81 lines
3.7 KiB
GDScript
class_name InputLeadController
|
|
extends RefCounted
|
|
|
|
# Client-owned input_lead control loop (multiplayer-todo.md §3.3, task 3.3).
|
|
# Standalone RefCounted, same reason as input_jitter_buffer.gd — scene-free
|
|
# so it's directly unit-testable against scripted depth traces.
|
|
#
|
|
# §3.3's own rationale for why this is the CLIENT's job alone, not shared
|
|
# with any server-side adaptation: three control loops acting on one plant
|
|
# (buffer occupancy) with different time constants is a textbook
|
|
# oscillation, and on a jittery link it presents to the player as
|
|
# intermittent sticky controls that are nearly impossible to attribute.
|
|
# The server (InputJitterBuffer, §3.2) only ever reports input_buffer_depth
|
|
# — it does nothing adaptive with it.
|
|
#
|
|
# "Lead" is realized concretely as extra distance between this client's own
|
|
# outgoing sequence numbers and what the server has actually consumed:
|
|
# skipping a sequence number (jumping the client's own seq counter by more
|
|
# than 1 for one tick) buys the server one more tick of buffered depth
|
|
# before it would starve; duplicating one (not incrementing the seq counter
|
|
# for one tick — the same seq gets sent again) narrows that margin by one
|
|
# tick of latency. The server's own ring buffer doesn't need to know this
|
|
# happened: a skipped seq just means "the redundant copies of it never
|
|
# existed, it's an ordinary drop" (already handled), and a duplicated seq
|
|
# is a same-seq resend, already discarded harmlessly once consumed
|
|
# (InputJitterBuffer.ingest()'s "seq <= last_applied_seq" check).
|
|
#
|
|
# Fast attack, slow release — a symmetric ±1-per-N-ticks slew would take
|
|
# two full seconds to absorb a single wifi spike, during which the player
|
|
# steers and the ship does not turn, "the most rage-inducing failure mode
|
|
# in any netcode" per §3.3's own words.
|
|
|
|
const LEAD_MIN := 1
|
|
const LEAD_MAX := 12
|
|
# "Never change it more than once per 30 ticks" (§3.3) — the floor that
|
|
# binds the fast-attack side; slow-release's own 60-tick cadence already
|
|
# exceeds it, so this one constant covers both.
|
|
const MIN_CHANGE_INTERVAL_TICKS := 30
|
|
const RELEASE_INTERVAL_TICKS := 60
|
|
const CLEAN_SURPLUS_TICKS := 120 # 2s at 60Hz
|
|
|
|
var lead := LEAD_MIN
|
|
|
|
var _ticks_since_change := 0
|
|
var _clean_surplus_ticks := 0
|
|
|
|
|
|
# Call once per client physics tick with the most recently known server-
|
|
# reported input_buffer_depth for THIS client's own slot (echoed in every
|
|
# snapshot, §3.2) — or -1 if no snapshot carrying that field has arrived
|
|
# yet. Returns the seq delta the caller should add for this tick's
|
|
# outgoing packet: ordinarily 1 (ship normally increments its send
|
|
# sequence by exactly one tick's worth), or 1+N / 0 on a tick where a lead
|
|
# change actually fires (skip N extra / duplicate the current one).
|
|
func update(input_buffer_depth: int) -> int:
|
|
_ticks_since_change += 1
|
|
if input_buffer_depth < 0:
|
|
return 1
|
|
|
|
if input_buffer_depth <= 0:
|
|
# A starve: the server's ring was empty for this player when it
|
|
# built that snapshot. React immediately, not after 2 seconds of
|
|
# evidence like release requires — but still debounced against
|
|
# MIN_CHANGE_INTERVAL_TICKS so a burst of consecutive starve
|
|
# reports doesn't compound into repeated, overlapping jumps.
|
|
_clean_surplus_ticks = 0
|
|
if _ticks_since_change >= MIN_CHANGE_INTERVAL_TICKS and lead < LEAD_MAX:
|
|
var new_lead := mini(lead + 3, LEAD_MAX)
|
|
var delta := new_lead - lead
|
|
lead = new_lead
|
|
_ticks_since_change = 0
|
|
return 1 + delta
|
|
return 1
|
|
|
|
_clean_surplus_ticks += 1
|
|
if _clean_surplus_ticks >= CLEAN_SURPLUS_TICKS and _ticks_since_change >= RELEASE_INTERVAL_TICKS and lead > LEAD_MIN:
|
|
lead -= 1
|
|
_ticks_since_change = 0
|
|
return 0 # duplicate this tick's seq — one tick of latency recovered
|
|
return 1
|