vwemirror/BT411
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

#172 corrected same day (user challenge upheld): the 6px-strip-was-invisible framing RETRACTED -- [lampblink] receipts prove the surround paints the flash at a metronomic 125ms (code-identical loop in 913), and Elengil's layout receipts show the stock surround with the lamp on screen. The real mechanism is EXPECTATION: a generator leak lights the button BY THE RADAR (authentically the only place -- no MFD carries a generator-leak indicator) while testers watch the MFDs. Ships: handout WHERE-TO-LOOK entry (#172) + the refined #136 voice-loop wording; the alert ring is GATED OFF by default (BT_ALERT_RING=1 keeps it available if Friday proves testers miss the lamp even knowing its location); [lampblink] shade-transition receipt added (BT_LAMP_LOG). Era question posted for Oracle: how prominent was the pod's physical generator button light?

Browse Source
This commit is contained in:
Joe DiPrima
2026-08-13 02:14:13 -05:00
parent 8b3d1c8113
commit 81b0887033
2 changed files with 78 additions and 20 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/OPEN_ISSUES_FOR_TESTERS.txt
+13 -3
View File
@@ -37,9 +37,19 @@ A. FIXED -- PLEASE VERIFY (headliners first: new fixes fielded THIS build)
TEST: take a leak (ram someone) -- the generator lamp for THAT loop
flashes, and stops when it should.
#136 Leaking SHUT loop responds to component toggles
EXPLAINED AS AUTHENTIC: the alarm SOUND is one-shot per new leak,
the lamp FLASH is continuous while leaking. Confirm the description
matches what you see and we close it.
EXPLAINED AS AUTHENTIC (refined): the leak voice STARTS at the leak,
LOOPS on its own while the leak persists, STOPS when it seals -- so
hearing it again does NOT mean a new leak. The lamp FLASH is
continuous while leaking. Confirm and we close it.
#172 "Coolant leak but no flashing indicator" -- WHERE TO LOOK
The lamp always flashed; it was in a place nobody watches. A
GENERATOR leak flashes the generator button BY THE RADAR (bottom
centre) -- NOT any MFD. That is authentic: no MFD carries a
generator-leak indicator. A WEAPON's leak lamp lives on its
engineering page and flashes when you page there (also authentic).
Loop leaks flash the loop button immediately. TEST Friday: take a
leak, look AT THE RADAR RAIL -- tell us if it still reads as missing
even knowing where to look.
#128 Ramming terrain = instant coolant leak
Ram damage eligibility fixed -- walls and rocks no longer puncture
loops. TEST: bump terrain hard; leaks come from combat, not curbs.
engine/MUNGA_L4/L4VB16.cpp
+65 -17
View File
@@ -1180,20 +1180,35 @@ void BTDrawCockpitPanels(LPDIRECT3DDEVICE9 device)
CkFill(device, 0, L.viewY, L.viewX, L.viewH, band); // left
CkFill(device, L.viewX + L.viewW, L.viewY, L.canvasW - (L.viewX + L.viewW), L.viewH, band); // right
// 2a) #172 ALERT BLOOM (glass-only deviation, #154-class, user-approved
// family): an ALERTING lamp in a surround strip bank is a ~6px sliver
// (minimumStrip) -- Elengil's fatal GeneratorB leak FLASHED lamp 0x1b
// for 28.4s and nobody saw it (night-16, ticket #172; the pod's
// physical backlit buttons don't have this problem). Draw a GROWN
// halo rect FIRST, under everything: neighbors repaint their own
// faces and the surfaces cover the inward part, so only the outward
// bloom survives -- which respects the "no on-top overlay" rule
// (twice field-broken, see the note below step 3). Fires ONLY while
// the lamp genuinely alternates (flash bits set AND the two levels
// differ), so steady lamps and same-level pulses are untouched.
// 2a) #172 ALERT RING (glass-only deviation, #154-class, user-approved
// 2026-08-13 -- the first cut, a +8px grown-face bloom, read as "the
// button got fat" on the bench and was rejected): an ALERTING lamp in
// a surround strip bank is a ~6px sliver (minimumStrip) -- Elengil's
// fatal GeneratorB leak FLASHED lamp 0x1b for 28.4s and nobody saw it
// (night-16; the pod's physical backlit buttons don't have this
// problem). Draw a thin PULSING OUTLINE 3px outside the lamp's
// border, UNDER everything: the lamp keeps its exact size, neighbors
// repaint their own faces, the surfaces cover the inward arc, and
// only the outward ring survives -- respecting the "no on-top
// overlay" rule (twice field-broken, see the note below step 3).
// The ring blinks WITH the lamp (same brightness sample) and only
// while it genuinely alternates (flash bits set AND levels differ);
// it vanishes on the lamp's own dark half-cycle, so it reads as
// "alarm light" rather than geometry.
// GATED OFF BY DEFAULT (BT_ALERT_RING=1 to enable), 2026-08-13: the
// bench + Elengil's layout receipts proved the lamp paints and blinks
// correctly on screen (see [lampblink]; code-identical loop in 913) --
// the field issue is EXPECTATION (generator leak lights the button by
// the RADAR, authentically; testers watch the MFDs), which the
// handout now states. The ring stays available in case Friday shows
// testers still miss it KNOWING where to look.
unsigned long tick = GetTickCount();
const int kAlertBloom = 8;
for (int i = 0; i < L.buttonCount; i++)
static int s_alertRing = -1;
if (s_alertRing < 0)
{ const char *ar = getenv("BT_ALERT_RING"); s_alertRing = (ar && *ar == '1') ? 1 : 0; }
const int kRingGap = 3; // lamp border -> ring inner edge
const int kRingWidth = 2;
for (int i = 0; s_alertRing && i < L.buttonCount; i++)
{
const BTCockpitBtn &b = L.buttons[i];
int state = CkLampState(b.address);
@@ -1201,10 +1216,18 @@ void BTDrawCockpitPanels(LPDIRECT3DDEVICE9 device)
&& (((state >> 2) & 0x3) != ((state >> 4) & 0x3)))
{
int shade = BTLampBrightnessOf(state, tick);
if (shade < 2) continue; // dark half-cycle: no ring
D3DCOLOR fill, border;
CkLampColors(b.colorClass, shade, b.inert, &fill, &border);
CkFill(device, b.x - kAlertBloom, b.y - kAlertBloom,
b.w + 2 * kAlertBloom, b.h + 2 * kAlertBloom, fill);
CkLampColors(b.colorClass, 3 /*bright*/, b.inert, &fill, &border);
int rx = b.x - kRingGap - kRingWidth;
int ry = b.y - kRingGap - kRingWidth;
int rw = b.w + 2 * (kRingGap + kRingWidth);
int rh = b.h + 2 * (kRingGap + kRingWidth);
// four bars = the ring; the gap between ring and lamp stays band-dark
CkFill(device, rx, ry, rw, kRingWidth, fill); // top
CkFill(device, rx, ry + rh - kRingWidth, rw, kRingWidth, fill); // bottom
CkFill(device, rx, ry, kRingWidth, rh, fill); // left
CkFill(device, rx + rw - kRingWidth, ry, kRingWidth, rh, fill); // right
}
}
@@ -1220,6 +1243,31 @@ void BTDrawCockpitPanels(LPDIRECT3DDEVICE9 device)
CkFill(device, b.x, b.y, b.w, b.h, border);
CkFill(device, b.x + 1, b.y + 1, b.w - 2, b.h - 2, fill);
// #172 SHADE-TRANSITION receipt (BT_LAMP_LOG only, first 24 blinks):
// proves the surround paint loop actually COMPUTES the alternation --
// the "did the pixels flash" question, answered without a screenshot
// (the PNG-capture rig cratered the frame rate to ~1fps and was
// scrapped). This base fill loop is code-identical to 913's, so its
// verdict applies to the field build.
if (getenv("BT_LAMP_LOG"))
{
static unsigned char s_lastShade[128];
static int s_blinkLog = 0;
int slot2 = b.address & 0x7F;
if ((state & 0x3) != 0
&& (((state >> 2) & 0x3) != ((state >> 4) & 0x3))
&& (unsigned char)shade != s_lastShade[slot2]
&& s_blinkLog < 24)
{
++s_blinkLog;
DEBUG_STREAM << "[lampblink] 0x" << std::hex << b.address
<< std::dec << " shade " << (int)s_lastShade[slot2]
<< "->" << shade << " tick=" << tick
<< "\n" << std::flush;
}
s_lastShade[slot2] = (unsigned char)shade;
}
// #172 paint receipt: one line per ALERT edge per lamp -- the night-16
// forensics could prove the lamp STATE but had zero visibility into
// pixels ("the seam-to-pixels observability hole"). Cheap: fires only
@@ -1235,7 +1283,7 @@ void BTDrawCockpitPanels(LPDIRECT3DDEVICE9 device)
DEBUG_STREAM << "[lamppx] 0x" << std::hex << b.address
<< (alerting ? " alert ON" : " alert OFF")
<< " cell=" << std::dec << b.w << "x" << b.h
<< "+bloom" << (alerting ? kAlertBloom : 0)
<< (alerting ? "+ring" : "")
<< " at " << b.x << "," << b.y
<< " state=0x" << std::hex << state << std::dec
<< "\n" << std::flush;