Files
TeslaRel410/restoration
CydandClaude Fable 5 d2be54a3b9 BT410 5.3.106: MISSION REVIEW LINKS -- the playback app, the winners circle, and the last Fail() in main
BTL4PB.CPP reconstructed against its AUTHENTIC surviving header (BTL4PB.HPP)
from the binary's 0x4d3b74-0x4d40d8 cluster, cross-checked against the
surviving RP411 twin -- and the review/camera launch in BTL4.CPP's main() is
real code now.  btl4.lib 11 -> 12 members, link clean, zero unresolved: the
linker pulled the entire playback chain for the first time.

The launch decode closed on two identifications:

  The `if (spool_file)` guard in the decomp means the manager getter is
  GetStoredSpoolFile -- the empty-spool getter cannot fail, so a review pod
  with no recorded battle simply starts nothing, which is the right field
  behaviour.

  The camera app's ctor (FUN_0047c3fc) sits in the ENGINE range with the
  (ResourceFile*, ApplicationID) shape: the camera pod's recorder IS the
  engine's own L4SpoolingApplication -- restored one commit ago -- started
  with ApplicationID 1 == BTL4.  BT shipped no camera app of its own.

What the playback app does, all binary-anchored: BTSpoolerTask pumps
recorded packets through DispatchPacket -> the client's own
ReceiveNetworkPacket; RunMission synchronises the clock (timeBias = Now -
the recorded run packet's stamp); the frame loop runs controls ONLY in
interactive review (mode 2), then update/interest/renderer/intercom; and
StopMission plays the WINNERS CIRCLE -- release the spool, park each
player's mech on the dropzone named "win1".."winN" by finishing rank,
freeze it (DoNothing performance), and tilt the camera up 45 degrees over
the tableau.

DOS-authentic ctor shape throughout: the binary's base call is (resource,
0x3eb, DefaultData) -- no windowing parameters ever existed in 4.10; the
RP411 twin's HINSTANCE/HWND threading is port drift, dropped.

Also fixed from 5.3.105's debt: L4PlaybackNetworkManager chains
NetworkManager::DefaultData -- the 1995 header declares no statics of its
own; the RP411 self-reference was drift too.

One [T3] in source: the idle frame's renderer service (the descendant's
ExecuteIdle) has no 1995 entry point; pre-egg the idle frame spins.

RUNTIME TEST PENDING, stated plainly: exercising review needs a
camera/review egg and a recorded spool on the rig (CAMERA-REVIEW-NOTES.md
has the roles).  The normal pod path regressed clean (mode gate untouched).

The Fail() census moves: BTL4.CPP and SEQCTL.CPP are now stub-free; 22
stubs remain across 15 files.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-03 15:54:31 -05:00
..

Scene Restorations

Modern re-renderings of the Division dVS scenes recovered from the Glaze developer drive (see HISTORY.md) and this repo's CONTENT tree. The originals ran on Division Ltd.'s pixel-pipeline hardware in 19941997; these tools parse the original data files and re-render them in WebGL, in the browser.

Viewers (open in any browser, fully self-contained)

  • vwe-archive.html — the Scene Archive: 26 curated scenes across all projects — Star Trek (Enterprise-D, Klingon flyby), Hull Pressure (sea demos, fish schools, sunken temple), BattleTech (polar map with mech lineup, Ravine and desert maps, stadium), Red Planet (Blade arena, the Mars canal milestone demo), and the Canyon/Maya-temple Division demos. Fly camera: drag to look, WASD to fly, R resets.
  • trek-scn.html — the original standalone Star Trek restoration (TREK.SCN and KLNGVID.SCN).

Toolkit

  • divformats.py — parsers for every Division format found on the drive: text VGF/VMF (DIV-VIZ2 geometry/materials, both the 1995 VERSION 2:07 and 1996 02:05 dialects, with header SCALE support), binary BGF/BMF (DIV-BIZ2) — the binary block-stream layout was recovered from Division's own reader source (DPL3/BIZREAD.C, 1994), including all 14 vertex layouts, pmesh/tristrip/polygon connectivity, LOD blocks (newer writers nest patches inside LODs; the largest LOD is taken), and embedded materials/textures/ramps — plus TGA and raw SVT textures, SPL splines, and SCN scene scripts.
  • audit.py — scans every .SCN on the drive and reports what fraction of its geometry survives, and in which format.
  • extract_all.py — extracts every scene with ≥60 % of its geometry recoverable (108 of 241 on the drive; the rest reference lost models) into allscenes.json, resolving models/materials/textures across the drive and this repo's CONTENT dirs.
  • gallery_build.py — curates scenes from allscenes.json into the self-contained vwe-archive.html (via gallery_template.html).
  • convert.py / viewer_template.html — the original single-scene Star Trek pipeline.

Rebuild

python extract_all.py     # requires the sda4/ drive dump beside this repo
python gallery_build.py

Fidelity notes

Faithful: vertices, triangles, material colors and shading ramps, textures, fog and light parameters, spline motion, camera start positions. Approximated: Division's rasterizer behavior, point sizes, timing (originals ran at 30 Hz). Not simulated: event-driven SPECIALFX particles. A few dev scenes had camera start positions inside geometry; those get overridden or auto-framed cameras (see START_OVERRIDE in gallery_build.py). Scenes missing one or more models from the drive say so in their header.