CydandClaude Opus 4.8
833fe85d63
The RIO drops comms if the analog-reply ACK is late by more than a few ms. Empirically a slower CPU failed sooner, so the latency was dominated by how fast the game processes the RIO packet and ACKs -- core=dynamic (recompiler) + cycles=max keeps the board in sync (user-confirmed). game_rio.conf passes the real RIO through serial1=directserial realport:COM1. With real control input the sim advances and then faults (null-ptr deref at BTL4OPT.EXE CODE+0x123B) -- downstream of the minimal test.egg not setting the arena objectpath, so VIDEO/GEO/ARENA/thr_tshd.bgf isn't found. That is content config, separate from the now-working VPX protocol / renderer / RIO. Details in RIO-NOTES.md. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
33 lines
840 B
Plaintext
33 lines
840 B
Plaintext
[sdl]
|
|
output=opengl
|
|
[dosbox]
|
|
memsize=32
|
|
machine=svga_s3
|
|
[cpu]
|
|
# The RIO drops comms if an ACK is late by more than a few ms. Empirically
|
|
# (2026-07-03) a SLOWER CPU makes it fail SOONER -> the ACK latency is
|
|
# dominated by how fast the game processes the RIO packet and replies, so we
|
|
# run the CPU as fast as possible:
|
|
# - core=dynamic: recompiler, many x faster than the 'normal' interpreter.
|
|
# - cycles=max: full host speed.
|
|
core=dynamic
|
|
cputype=pentium
|
|
cycles=max
|
|
# RIO cockpit controls on COM1 (Phase 5 passthrough), plasma COM2 later.
|
|
[serial]
|
|
serial1=directserial realport:COM1
|
|
serial2=disabled
|
|
[autoexec]
|
|
mount c "C:\VWE\TeslaRel410\ALPHA_1"
|
|
c:
|
|
cd \REL410\BT
|
|
set VIDEOFORMAT=svga
|
|
set BLASTER=A220 I5 D1 H5 P330 T6
|
|
set TEMP=c:\
|
|
call setenv.bat r s n n
|
|
32rtm.exe -x
|
|
btl4opt.exe -egg test.egg
|
|
32rtm.exe -u
|
|
echo ALPHA1-RUN-DONE
|
|
pause
|