
  ####                                                      ###           ###
 ##  ##                                                      ##            ##
 ###     ##  ##  ## ###   ####   ## ###  ##  ##   ####       ##   ####     ##
  ###    ##  ##   ##  ## ##  ##   ### ## ####### ##  ##   #####  ##  ##    ##
    ###  ##  ##   ##  ## ######   ##  ## ####### ##  ##  ##  ##  ######    ##
 ##  ##  ##  ##   #####  ##       ##     ## # ## ##  ##  ##  ##  ##        ##
  ####    ### ##  ##      ####   ####    ##   ##  ####    ### ##  ####    ####
                 ####

                       A Sega Model 3 Arcade Emulator.

                   Copyright 2003-2025 The Supermodel Team

                                 CHANGE LOG

Commit 7835800 on 2025-09-23 by Bart Trzynadlowski

    Added required changelog.sh for package creation via pkg build target

Commit 15afe2c on 2025-09-23 by Bart Trzynadlowski

    std::strcpy -> strcpy to hopefully fix build on Ubuntu

Commit 2a2a158 on 2025-09-23 by Bart Trzynadlowski

    Fix SDL include path on macOS so that imgui builds

Commit 83a3416 on 2025-09-23 by Bart Trzynadlowski

    GitHub actions for building release packages on pushes to master

Commit 04e0391 on 2025-09-23 by Bart Trzynadlowski

    Makefiles: macOS no longer seems to require libAGL and various improvements
    for GitHub Actions support

Commit 9309f2a on 2025-09-23 by Bart Trzynadlowski

    Add pkg/ to .gitignore

Commit 089b875 on 2025-09-23 by Ian Curtis

    Merge branch 'master' of https://github.com/trzy/Supermodel

Commit d1790b4 on 2025-09-23 by Ian Curtis

    A GUI for supermodel. Not tested makefile but *should* work. To use it, just
    run supermodel with no command line paramaters. It's designed this way so
    people can still use old loaders / front ends if they want.

Commit 45c6157 on 2025-09-20 by dukeeeey

    Merge pull request #260 from laromicas/patch-1

    Update README.md
Commit fa11ad5 on 2025-09-17 by Ian Curtis

    Merge branch 'master' of https://github.com/trzy/Supermodel

Commit 5c9bcd1 on 2025-09-17 by Ian Curtis

    Increasing tolerence a bit for floating point compare in shaders. Fixes
    potential garbage being drawn.

Commit d030fb5 on 2025-09-15 by Bart Trzynadlowski

    Fix build on macOS, which was failing because ppcd still links against
    Frameworks

Commit 44280ae on 2025-08-19 by Ian Curtis

    Pixels with alpha always seem to use the clamped light model. This includes
    textures with 1 bit alpha, as they will acquire some alpha after going after
    bilinear filtering. For these textures we only discard them if they are less
    than a specific threshold, but that leaves many opaque pixels that are
    technically opaque but with some alpha. This fixes some corner cases with
    the lighting in daytona.

Commit 06aa4c9 on 2025-08-06 by laromicas

    Update README.md

    Added needed library for linux compiling.
Commit 487d642 on 2025-05-28 by Ian Curtis

    Merge branch 'master' of https://github.com/trzy/Supermodel

Commit df4f9de on 2025-05-28 by Ian Curtis

    Fix missing include

Commit 587da80 on 2025-04-24 by Bart Trzynadlowski

    Fix ppcd build on macOS

Commit 3af143d on 2025-04-24 by Bart Trzynadlowski

    Fix SDL2 framework fetch (fixes build on macOS)

Commit be8bf66 on 2025-04-24 by Bart Trzynadlowski

    .gitignore: Exclude Config/Supermodel.ini and .DS_Store

Commit 4c75b81 on 2025-04-24 by Ian Curtis

    Merge branch 'master' of https://github.com/trzy/Supermodel

Commit 69d03d6 on 2025-04-24 by Ian Curtis

    Fix crash bug upon emulator restart. Add the config registers and write
    pointers to the file saves. This will make save states incompatible with
    older versions.

Commit 14df6d9 on 2025-04-23 by Bart Trzynadlowski

    ppcd now built with Supermodel and fixed warnings in PPCDisasm.cpp and check
    for Reuslt::OKAY

Commit 002540c on 2025-04-20 by Ian Curtis

    These states could be handled better

Commit bb35a8a on 2025-04-19 by Ian Curtis

    Remove obsolete code

Commit 2272893 on 2025-04-17 by Ian Curtis

    Remove dead code

Commit af45604 on 2025-04-17 by Ian Curtis

    Sega bass fishing is flipping the block culling state multiple times in the
    same frame (in between memory writes). In multi-threaded mode this can cause
    the screen to flash like crazy, so we just sync this state at the start of
    the frame. The state can't cancel a frame that is currently rendering so
    it's probably fine to sync here.

Commit d8dfc35 on 2025-04-16 by Ian Curtis

    m_blockCulling must be initialised

Commit 5b8bb0b on 2025-04-16 by Ian Curtis

    These should be thread safe

Commit 3c34218 on 2025-04-16 by Ian Curtis

    The real3d gpu can accept data for a new frame whilst rendering the current.
    It copies this data to the start of the low culling ram. Previously we
    staggered the frame to get the writes to fall into the correct frame, but
    doing this meant the tilegen was out of sync with the 3d hardware. Fixing
    this allows us to get the tilegen to draw when it's meant to draw, which
    fixes the missing 2d animations in scud race.

Commit 210fa3d on 2025-04-10 by Ian Curtis

    Workarounds no longer required.

Commit 49e3671 on 2025-04-05 by Ian Curtis

    Merge branch 'master' of https://github.com/trzy/Supermodel

Commit 0e23b96 on 2025-04-05 by Ian Curtis

    Fix the emulator crashing the driver when going fullscreen. Also don't
    delete the 3d class when going fullscreen because then we lose the internal
    state and get stuck with a white screen.

Commit 5318f13 on 2025-04-04 by Bart Trzynadlowski

    Info log NVRAM and save states

Commit dcd8acc on 2025-04-03 by Bart Trzynadlowski

    Makefiles: DEBUG option now builds a debug build, EXTRA_DEBUG enables frame
    analyzer, and removed deprecated NEW_FRAME_TIMING flag

Commit 12e3b6a on 2025-03-31 by Ian Curtis

    Merge branch 'master' of https://github.com/trzy/Supermodel

Commit 0b508b6 on 2025-03-31 by Ian Curtis

    Fix a corner case with the quad shader where the attribute interpolation can
    be bad. (toxieainc)

Commit 90b5bfc on 2025-03-22 by Bart Trzynadlowski

    Merge pull request #236 from trzy/bart/consistent-decay

    Common digital sensitivity and decay speed settings for all digital buttons
    mapped to analog inputs
Commit d49656d on 2025-03-22 by Bart Trzynadlowski

    Fix broken copy command in build bot script

Commit 02fe4d6 on 2025-03-20 by Bart Trzynadlowski

    Introduced InputDigitalSensitivity and InputDigitalDecaySpeed settings for
    all non-keyboard digital inputs mapped to analog inputs; key settings remain
    and take precedence for keyboard keys

Commit 5b2c589 on 2025-03-12 by Bart Trzynadlowski

    Merge pull request #235 from trzy/bart/ini-file

    Write default Supermodel.ini if not present 
Commit 40aa3c7 on 2025-03-06 by Bart Trzynadlowski

    Write default Supermodel.ini if not present and remove it from distribution

Commit 16ceb67 on 2025-03-12 by Bart Trzynadlowski

    Merge pull request #233 from aderumier/sdlgamecontroller


Commit daf0fad on 2025-02-26 by Alexandre Derumier

    sdl: add support for gamecontroller api

    currently, supermodel only implement the sdl joystick api.

    That mean that if we use different controllers, they need to be correctly
    remapped && calibrate each time.

    This patch implement sdl gamecontroller api, through "InputSystem =
    sdlgamepad" config which normalize the mapping && calibration.

    gamecontroller is matching x-input on windows, so the limit is 6 axis (with
    2 separated Z triggers axis), 4 dpad/hats and 16 buttons. (x-input is
    limited to 10buttons).

Commit 112e7bb on 2025-02-27 by dukeeeey

    Merge pull request #234 from gm-matthew/tile-pairs

    Fixing tilegen bugs in Scud Race
Commit 14f885b on 2025-02-27 by gm-matthew

    Fixing tilegen bugs in Scud Race Fixed the long-standing "rolling start"
    glitch in all versions of Scud Race, and also a more recent sky background
    overdraw bug in credits of Super Beginner track in Scud Race Plus

    Removed "rolling start" patches from Scud Race ROMs since they are no longer
    needed

Commit e9f50fe on 2025-02-21 by Bart Trzynadlowski

    Merge pull request #231 from gm-matthew/white-flash

    Fixed JTAG config parameters when loading savestates
Commit 471a2c1 on 2025-02-21 by gm-matthew

    Fixed JTAG config parameters when loading savestates

Commit 6a8de59 on 2025-02-18 by Bart Trzynadlowski

    Merge pull request #229 from gm-matthew/white-flash

    Implement "Block Culling" modeword bit of Mercury ASIC
Commit 592b889 on 2025-02-19 by gm-matthew

    Added "white flash" support for legacy renderer

Commit 300c93b on 2025-02-19 by gm-matthew

    Implement "Block Culling" modeword bit of Mercury ASIC Disables 3D rendering
    and clears screen to white

    Added option to turn off if desired; set NoWhiteFlash to true in
    Supermodel.ini or add -no-white-flash to command line

Commit d43fd3e on 2025-02-11 by Bart Trzynadlowski

    Merge pull request #225 from gm-matthew/scsp

    Improved soundboard IRQ handling
Commit 58dd318 on 2025-02-11 by gm-matthew

    Improved soundboard IRQ handling

Commit d043dc0 on 2024-11-25 by dukeeeey

    Merge pull request #213 from toxieainc/quad_fix

    Improve edge test precision of quad test, and add missing '----'-area
Commit b291999 on 2024-11-25 by Ian Curtis

    Merge branch 'master' of https://github.com/trzy/Supermodel

Commit de108d4 on 2024-11-25 by Ian Curtis

    Improve viewport precision. (toxieainc)

Commit 4e7356a on 2024-11-06 by Bart Trzynadlowski

    Merge pull request #216 from toxieainc/ppc_rounding_fixes

    Properly enable fesetround/fegetround usage
Commit 5bc4a20 on 2024-11-06 by Bart Trzynadlowski

    Merge pull request #217 from jflatt/xdg

    Add xdg base dirs
Commit 4d331ee on 2024-11-06 by toxieainc

    avoid set_rounding_mode and check fpscr directly

Commit f0fdb26 on 2024-10-30 by Jason Flatt

    Revert "Add fallback in case neither .config or XDG environment variables
    exist"

    This reverts commit ad642ceeb91d21616b6688131013086ef1b944fb.

    Not needed

Commit ad642ce on 2024-10-30 by Jason Flatt

    Add fallback in case neither .config or XDG environment variables exist

Commit bb6832c on 2024-10-30 by Jason Flatt

    Add xdg base dirs

Commit 77a032f on 2024-10-24 by toxieainc

    make fselx more robust against NaNs in combination with compilers/options
    that may ignore NaNs/switch orders on comparisons

    follows fselx spec

Commit 312c230 on 2024-10-24 by toxieainc

    fix wrong clamping for fctiw(z)x if clamping values that are <-2^31

Commit 83ff033 on 2024-10-21 by toxieainc

    is_qnan_double() had 3 flaws: 0x0007fffffffffff -> 0x0007ffffffffffff, and
    0x000800000000000 -> 0x0008000000000000. In addition it checked if the
    remaining frac bits are 0, but due to the IEEE spec these can be used for a
    payload.

    On top, optimize the other fp64 special case testing functions

Commit a6a6cf8 on 2024-10-21 by toxieainc

    properly enable fesetround/fegetround usage according to the C++ spec

    unfortunately, e.g. MSVC ignores this pragma, so use what MSDN suggests in
    that case (GCC and clang may still require -frounding-math for compilation
    of at least this file)

    in addition, do not set fesetround mode once, but just each time before each
    math op that requires it otherwise other Supermodel code may use the wrong
    rounding modes (as it relies by default on it being FE_TONEAREST) note that
    changing the rounding mode is extremely rare, so this will not affect
    Supermodels performance by much

    and fix additional minor issue in ppc_fctiwx

    this fixes #203, maybe other emulation+math precision issues, too

Commit f9cb795 on 2024-10-20 by Ian Curtis

    Delete unused file

Commit e2538e2 on 2024-10-20 by Ian Curtis

    Merge branch 'master' of https://github.com/trzy/Supermodel

Commit c5494be on 2024-10-20 by Ian Curtis

    Update visual studio project files for latest SDL version

Commit 75d2e79 on 2024-10-14 by toxieainc

    sign IS needed for interpolation correctness :/

Commit a922488 on 2024-10-14 by toxieainc

    remove unnecessary sign

Commit 9a07f08 on 2024-10-13 by dukeeeey

    Merge pull request #211 from toxieainc/update_ext

    Update external libs
Commit eafdacf on 2024-10-13 by toxieainc

    improve edge test precision of quad test, and add missing '----'-area-case
    support for twisted/bow-tie quads

    ..and align triangle code more to the quad code for easier comparisons

    also generalize mipmap0-only optimization to all non-blended cases

    related to #201

Commit 916d912 on 2024-10-11 by Ian Curtis

    fix missing header

Commit a423d9f on 2024-10-10 by toxieainc

    update to latest from official repo

Commit b557b36 on 2024-10-10 by toxieainc

    update to latest via official zlib

Commit 9f862a4 on 2024-10-10 by toxieainc

    latest minimp3, incl. all patches collected via
    https://github.com/manxorist/minimp3/blob/openmpt/minimp3.h
    ..as original repo is dead

Commit c4d0269 on 2024-10-10 by dukeeeey

    Merge pull request #205 from toxieainc/more_analysis_fixes

    More static analysis fixes
Commit 18c6c95 on 2024-10-08 by dukeeeey

    Merge pull request #208 from toxieainc/dpi_awareness

    DPI awareness
Commit 21eda5f on 2024-10-08 by toxieainc

    whoops

Commit 08c9d16 on 2024-10-08 by toxieainc

    remove __restrict

Commit 0d6fdd2 on 2024-10-08 by toxieainc

    Flag SDL window as DPI-aware, otherwise the window content might be scaled
    by some windows graphics drivers

    addresses #206

Commit 2a974b3 on 2024-10-05 by toxieainc

    more fixes from recommendations

Commit 4dd393a on 2024-10-05 by toxieainc

    more static analysis, especially missing member initialization

Commit 64a7e35 on 2024-10-05 by Ian Curtis

    add missing header

Commit f573681 on 2024-10-04 by Bart Trzynadlowski

    Merge pull request #202 from toxieainc/analysis_fixes

    Static Analysis
Commit 3f6fe29 on 2024-10-03 by toxieainc

    some more from the review

Commit d9e1cc0 on 2024-10-03 by toxieainc

    address reviews

Commit 46453ed on 2024-10-02 by toxieainc

    address mostly harmless things found by static analysis

Commit 161f1e3 on 2024-09-30 by dukeeeey

    Merge pull request #198 from toxieainc/sleep_tweak

    improve SuperSleepUntil implementation
Commit 518e3df on 2024-09-30 by toxieainc

    only use SDL_CPUPauseInstruction if it exists

Commit ae293fe on 2024-09-30 by toxieainc

    use SDL_CPUPauseInstruction

Commit 14585f8 on 2024-09-29 by toxieainc

    improve SuperSleepUntil implementation: 1) as found in lots of experiments
    done for the VPX and PinMAME projects, Sleep() on windows can oversleep for
    over 1ms, especially when doing Sleep(>1) 2) thus loop Sleep(1) and end if
    its less than 2ms 3) in the spin to wait for the rest of the time, insert
    yield(=_mm_pause) or the determined by the Rust devs arm64 equivalent

    this actually gets rid of micro-stutter on my AMD based minipc e.g. in
    Daytona2

    then also use same Sleep() implementation in the network code (to avoid
    potential sideeffects between the 2 implementations)

Commit 6da55ab on 2024-09-27 by dukeeeey

    Merge pull request #197 from toxieainc/upscale_options

    Implement different upscaling filters for the 2D layers
Commit 1760bc5 on 2024-09-27 by toxieainc

    change default to bilinear

Commit 9af2601 on 2024-09-25 by toxieainc

    implement different upscaling filters for the 2D layers

Commit 9724c3e on 2024-09-23 by dukeeeey

    Merge pull request #196 from toxieainc/corruption_fixes

    Fix some warnings, including potential stack corruption by (v)sprintf
Commit d11eb3d on 2024-09-23 by toxieainc

    fix some warnings, including potential stack corruption by (v)sprintf

Commit 8fd480f on 2024-09-22 by Ian Curtis

    Merge branch 'master' of https://github.com/trzy/Supermodel

Commit ef54cbb on 2024-09-22 by Ian Curtis

    Fix some unitialised class variables

Commit 905bfa2 on 2024-09-22 by dukeeeey

    Merge pull request #195 from toxieainc/crtcolors

    Implement CRT color emulation
Commit 7ebda18 on 2024-09-22 by toxieainc

    address review and change to CRT color variants to enum and ifdef within
    shader

Commit 01300b3 on 2024-09-20 by toxieainc

    implement CRT color emulation

    so not scanlines or the other CRT aspects, just the differences in the
    region-specific TV color standards this means one can choose between 6
    different mappings now: 0=none (as before/default) 1=ARI/D93 (recommended
    for all JP developed games) 2=PVM_20M2U/D93 3=BT601_525/D93 4=BT601_525/D65
    (recommended for all US developed games) 5=BT601_625/D65 (recommended for
    all EUR/AUS developed games)

    wired into the supersampling summation pass, as this is the final output
    pass

Commit 8e7b850 on 2024-09-18 by dukeeeey

    Merge pull request #194 from toxieainc/compositing_optimization

    Increase precision and performance for Base/Alpha Layer compositing
Commit ecda60e on 2024-09-18 by dukeeeey

    Merge pull request #193 from toxieainc/aa_fix

    Fix texel offsets when sampling the AA buffer
Commit e2d28b1 on 2024-09-18 by toxieainc

    increase precision and performance for Base/Alpha Layer compositing

Commit c80bd04 on 2024-09-18 by toxieainc

    fix texel offsets when sampling the AA buffer

Commit c499584 on 2024-09-15 by Bart Trzynadlowski

    Merge pull request #192 from DirtBagXon/print-gl-info

    Fix -print-gl-info after fb785b36bd70c16001336e22a75950ba5ef3808f
Commit 4ad1478 on 2024-09-13 by DirtBag Xon

    Use ValueAsDefault

Commit bdf948e on 2024-09-13 by DirtBag Xon

    Fix -print-gl-info after fb785b36bd70c16001336e22a75950ba5ef3808f

Commit 9d09b07 on 2024-09-12 by dukeeeey

    Merge pull request #191 from gm-matthew/debugger

    Fix debugger build
Commit 536ea99 on 2024-09-12 by gm-matthew

    Fix debugger build

Commit 3d64a8d on 2024-09-10 by dukeeeey

    Merge pull request #190 from toxieainc/shader_fix

    Use correct GetUniformLocation
Commit b991d0d on 2024-09-10 by toxieainc

    use correct GetUniformLocation

Commit 706969a on 2024-09-09 by Ian Curtis

    Fix some uninitialised variables

Commit d34596a on 2024-09-08 by Ian Curtis

    Fix regression

Commit cab0131 on 2024-09-08 by Ian Curtis

    Must use the correct call type otherwise it's invalid

Commit c7e5417 on 2024-09-07 by Ian Curtis

    Fix build after merge

Commit 3adb898 on 2024-09-07 by Ian Curtis

    Merge

Commit 604b7d9 on 2024-09-07 by Ian Curtis

    Use strong type for result instead of bool

Commit 856e96b on 2024-09-06 by dukeeeey

    Merge pull request #188 from toxieainc/optimize_depthstencil

    slightly optimize depth & stencil usage
Commit fb785b3 on 2024-09-04 by toxieainc

    use only a single glReadPixels for LOS avoid creating a depth/stencil buffer
    in the New3D mode wire error handling for CreateFBO(DepthCopy) handle black
    FB init for triple buffering

Commit 6d1af82 on 2024-09-02 by Bart Trzynadlowski

    Merge pull request #186 from gm-matthew/poll-inputs

    Poll inputs before running frame
Commit a7656dc on 2024-09-01 by gm-matthew

    Poll inputs before running frame Should reduce latency somewhat. Thanks
    star69rem

Commit dc1479f on 2024-08-28 by Ian Curtis

    Merge branch 'master' of https://github.com/trzy/Supermodel

Commit e0cb663 on 2024-08-28 by Ian Curtis

    Port the GLSL tilegen code back to to software.
    - This is so we can draw the raster lines at the correct times. A few games
    (scud) are updating the tilegen mid frame. Currently these effects are
    missing from supermodel. All the efficiency of h/w rendering goes out the
    window when trying to render scan lines.

Commit 28474af on 2024-08-23 by Bart Trzynadlowski

    Merge pull request #182 from gm-matthew/JTAG

    New JTAG implementation
Commit 04f1ed7 on 2024-08-23 by gm-matthew

    Delete BitRegister source files

Commit 42ae86d on 2024-08-23 by Matthew Daniels

    Update Rules.inc
Commit 363baa8 on 2024-08-23 by gm-matthew

    Added comments to new JTAG class Removed old BitRegister class as it is no
    longer used

Commit 6638c84 on 2024-08-17 by Ian Curtis

    Fix build in vs. Upgrade to c++20?

Commit 1e98588 on 2024-08-16 by gm-matthew

    remove signed/unsigned mismatch warnings

Commit 34cd771 on 2024-08-15 by gm-matthew

    Change size_t to uint32_t otherwise savestates will be incompatible between
    32-bit and 64-bit builds

Commit 2871d45 on 2024-08-15 by gm-matthew

    Minor improvements

Commit 5d45f59 on 2024-08-14 by gm-matthew

    Removing patches from Sega Rally 2 and Star Wars Trilogy as they are no
    longer needed

Commit dd08f3b on 2024-08-14 by gm-matthew

    Fix compilation

Commit 8b97c65 on 2024-08-14 by gm-matthew

    New JTAG implementation Sega Rally 2 and Star Wars Trilogy no longer require
    patches to run Step 2.x games now display "daughter board was detected" as
    they do on real hardware Currently not compatible with old savestates; we
    may add partial support for old savestates before committing White flash to
    be implemented soon

Commit 721d4bc on 2024-08-13 by gm-matthew

    Unsigned fixed shading is toggled by specular enable, not via JTAG

Commit 309bf98 on 2024-08-10 by Bart Trzynadlowski

    Music.xml: Added Daytona 2 and Sega Rally 2 tracks

Commit 3383b26 on 2024-08-09 by Bart Trzynadlowski

    supermodel_build_bot.py: Include Config/Music.xml in package
Commit 6bae80b on 2024-08-09 by Bart Trzynadlowski

    Makefile.Win32: Fixed SDL2 flags when building in MSYS
Commit a0341cd on 2024-08-09 by Bart Trzynadlowski

    Rules.inc: Restored 'release' target for Supermodel3.com-hosted builds and
    added some comments

Commit cf8a61f on 2024-08-08 by Bart Trzynadlowski

    Update copyright date

Commit e636dd5 on 2024-07-22 by Bart Trzynadlowski

    Improved Makefiles (from h0tw1r3's PR): macOS Makefile will download SDL2
    Framework, hopefully fixed Musashi dependency issues that occasionally
    affect parallel builds

Commit 2df8511 on 2024-08-09 by Bart Trzynadlowski

    Update README.md: Copyright updated to 2024
Commit c705a89 on 2024-08-09 by Bart Trzynadlowski

    MpegAudio.cpp: Comment out debug logging of music start offsets

Commit dbfe2b1 on 2024-08-09 by Bart Trzynadlowski

    Added support for custom MPEG music in a new Music.xml config file

Commit dd90d0e on 2024-08-02 by dukeeeey

    Merge pull request #174 from gm-matthew/luminous15

    Step15Luminous() applies to all luminous polys, not just textured ones
Commit 6908e5f on 2024-08-02 by gm-matthew

    Step15Luminous() applies to all polys, not just textured ones

Commit ce9809c on 2024-08-02 by dukeeeey

    Merge pull request #173 from gm-matthew/textureNP

    New textureNP implementation
Commit 493c5aa on 2024-08-01 by gm-matthew

    New textureNP implementation This new implementation uses the previously
    unused viewport parameter "cota"

Commit 3d8557d on 2024-07-29 by Ian Curtis

    fix minor compile warning

Commit 49ca8c1 on 2024-07-29 by Ian Curtis

    Fix mipmaps Real3d mipmaps only go down to 2x2 pixels. The 1x1 pixels
    generated previously were actually the last part of the 2x2 mipmaps.

Commit 27a2336 on 2024-07-28 by dukeeeey

    Merge pull request #169 from gm-matthew/texture

    First 4MB of VROM maps to polygon RAM
Commit aeaa602 on 2024-07-28 by gm-matthew

    First 4MB of VROM maps to polygon RAM Daytona 2 PE replaces unused Dreamcast
    logo textures by filling in zeroed data from polygon RAM. Before we were
    reading from the first 4MB of VROM which is filled with 0xFF values,
    resulting in the replaced textures incorrectly becoming white and opaque.

Commit 32414ee on 2024-07-26 by dukeeeey

    Merge pull request #168 from gm-matthew/specular

    Specular lighting works differently on flat-shaded polys
Commit 6d5aeb1 on 2024-07-26 by gm-matthew

    Specular lighting works differently on flat-shaded polys Also lower
    multIndex values for smooth-shaded specular, more accurate to real hardware

Commit 5fa402f on 2024-07-19 by dukeeeey

    Merge pull request #165 from gm-matthew/widescreen

    Don't expand culling planes unless widescreen is enabled
Commit d7fdba0 on 2024-07-19 by gm-matthew

    Don't expend culling planes unless widescreen is enabled

Commit a2645c9 on 2024-07-14 by gm-matthew

    Give PowerPC more cycles to acknowledge sound IRQs 400 cycles was not enough
    and would cause VF3 to hang. 1000 cycles should be more than plenty

Commit 5a8cddd on 2024-07-13 by Ian Curtis

    remove dead code

Commit 79868e0 on 2024-07-08 by Ian Curtis

    fix compilation

Commit 40bb7ac on 2024-07-06 by dukeeeey

    Merge pull request #160 from gm-matthew/textureNP

    Texture mipmaps must be at least 2x2
Commit 2b186e6 on 2024-07-06 by gm-matthew

    Texture mipmaps must be at least 2x2 Fixes sky in Dirt Devils

Commit ee5d652 on 2024-07-05 by dukeeeey

    Merge pull request #159 from gm-matthew/textureNP

    Implementing texture NP values
Commit 3cf8a1f on 2024-07-05 by gm-matthew

    Implementing texture NP values For some reason Model 3 uses vertex
    coordinates rather than texel coordinates to calculate mipmap levels Revised
    microtexture implementation; results are very close to real hardware when
    running at native resolution with supersampling disabled

Commit 19612e1 on 2024-06-13 by Ian Curtis

    fix merge error

Commit 4d1484c on 2024-06-13 by Ian Curtis

    merge conflicts

Commit edb11dc on 2024-06-13 by Ian Curtis

    Performance improvements The old texture code was being bottle necked by the
    texture reads. We mirrored the real3d texture memory directly, including the
    mipmaps in a single large texture. I *think* most h/w has some sort of
    texture cache for a 2x2 or 4x4 block of pixels for a texture. What we were
    doing was reading the base texture, then reading the mipmap data from a
    totally separate part of the same texture which I can only assume flushed
    this cache. What I did was to create mipmap chains for the texture sheet,
    then copy the mipmap data there. Doing this basically doubles performance.

Commit 42e236b on 2024-06-09 by gm-matthew

    Add comment briefly explaining optimization

Commit f30a8ee on 2024-06-04 by gm-matthew

    Minor shader optimization If the mipmap level is 0.0, we only need to run
    texBiLinear() for the highest quality texture mipmap Adds around 10-20% more
    performance in Daytona 2

Commit a6466b3 on 2024-06-01 by Ian Curtis

    spelling

Commit 21e1485 on 2024-06-01 by Ian Curtis

    Fix some missing scroll fog background effects in ECA

Commit 7d32bde on 2024-05-22 by gm-matthew

    When disabling sound IRQs via MIDI control port, clear current IRQ Prevents
    Dirt Devils hanging during boot

Commit 976f020 on 2024-05-20 by dukeeeey

    Merge pull request #151 from gm-matthew/sound-irq

    Sound IRQs are acknowledged by writing to MIDI data port
Commit 8a0bf29 on 2024-05-20 by gm-matthew

    Sound IRQs are acknowledged by writing to MIDI data port

Commit 7b02f58 on 2024-05-04 by dukeeeey

    Merge pull request #149 from gm-matthew/scuddxo

    Only access SCSI device at 0xC0xxxxxx if it has been configured to do so
Commit 3e394d1 on 2024-05-04 by gm-matthew

    Only access SCSI device at 0xC0xxxxxx if it has been configured to do so 
    Fixes some step 1.5 games not working in last update

Commit 2245fca on 2024-05-02 by dukeeeey

    Merge pull request #147 from gm-matthew/scuddxo

    Adding newly dumped version of Scud Race
Commit 39dc16d on 2024-05-02 by gm-matthew

    Use integer value for game stepping

Commit 944e485 on 2024-05-02 by gm-matthew

    Adding newly dumped version of Scud Race Allow Step 1.5 games to access
    53C810 via 0xC00000xx if the netboard is disabled, because scuddxo requires
    this

Commit 250f84e on 2024-04-05 by Bart Trzynadlowski

    Updated .gitignore to exclude build directories and Supermodel release
    directories

Commit dec8503 on 2024-03-14 by dukeeeey

    Merge pull request #138 from h0tw1r3/fix/2d-shaders

    Fix 2D texture rendering on OSX
Commit d9f48db on 2024-03-09 by Bart Trzynadlowski

    CLogger: destructor needs to be virtual

Commit 93f622e on 2024-03-06 by Jeffrey Clark

    cleanup code format

Commit 047b15c on 2024-03-06 by Jeffrey Clark

    fix 2d shaders on OSX

Commit 1bb7de1 on 2024-03-05 by baraclese

    fix segfault in CMultiLogger

    vsprintf may change its va_list argument so repeatedly calling it with the
    same va_list arg is undefined behavior. Fix this by creating a copy of the
    va_list argument before each vsprintf call.

Commit e59ecea on 2024-01-28 by Bart Trzynadlowski

    PowerPC frequency determined by board stepping, config option is now just an
    override

Commit 620a581 on 2023-12-30 by Jacob Oxford

    No need to have changelog from 2011 in Readme.txt anymore.

Commit 75260f9 on 2023-12-29 by Ian Curtis

    Clamp LODscale. If the object is drawn at zero we produce an Inf value which
    will produce Nans with later calculations.

Commit 0e07f29 on 2023-12-27 by Ian Curtis

    Amend ambient fog logic Should stop the sky flashing in lemans24, and the
    background totally disappearing in sega rally after a game. The logic here
    is still not totally understood but works well enough for the games.

Commit c039d08 on 2023-12-26 by Ian Curtis

    Add supersampling anti-aliasing Late christmas present. Due to the way alpha
    works on the model3 adding regular anti-aliasing doesn't really work.
    Supersampling is very much a brute force solution, render the scene at a
    higher resolution and mipmap it.

    It's enabled via command line with the -ss option, for example -ss=4 for 4x
    supersampling or by adding Supersampling = 4 in the config file.

    Note non power of two values work as well, so 3 gives a very good balance
    between speed and quality. 8 will make your GPU bleed, since it is
    essentially rendering 64 pixels for every visible pixel on the screen.

Commit 33b84c8 on 2023-12-22 by dukeeeey

    Merge pull request #122 from gm-matthew/floatZ

    Floating point reversed z-buffer and new clipping code
Commit 6acb116 on 2023-12-22 by gm-matthew

    Added new uniform bool polyAlpha to shaders Added original matrix functions
    back into Mat4 class though they are currently unused

Commit 9f66fca on 2023-12-22 by gm-matthew

    Floating point reversed z-buffer and new clipping code Also always draw
    nodes with culling disabled even if they test as being outside the visible
    frustum

Commit 6f40953 on 2023-12-04 by dukeeeey

    Merge pull request #120 from gm-matthew/patch

    Fixed Scud Race Plus "ROLLING START" patch
Commit 5b9e970 on 2023-12-04 by gm-matthew

    Fixed Scud Race Plus "ROLLING START" patch Accidentally patched the wrong
    memory location in Scud Race Plus

Commit a19ba3c on 2023-12-03 by dukeeeey

    Merge pull request #118 from firewave/uninit

    Model3.cpp: avoid usage of uninitialized memory
Commit 11e9b14 on 2023-12-03 by dukeeeey

    Merge pull request #117 from firewave/mismatch

    Inputs/MultiInputSource.cpp: fixed mismatching allocation and deallocation
Commit 71e5baa on 2023-12-02 by firewave

    Model3.cpp: avoid usage of uninitialized memory

    ```
    ==213531== Thread 20 SoundBoardNoSyn:
    ==213531== Conditional jump or move depends on uninitialised value(s)
    ==213531==    at 0x1A1D1E: CModel3::RunSoundBoardThread() (Model3.cpp:2607)
    ==213531==    by 0x49BECEA: UnknownInlinedFun (SDL_thread.c:292)
    ==213531==    by 0x49BECEA: RunThread.lto_priv.0 (SDL_systhread.c:76)
    ==213531==    by 0x4F719EA: start_thread (pthread_create.c:444)
    ==213531==    by 0x4FF5653: clone (clone.S:100)
    ```

Commit 1cc94b7 on 2023-12-02 by firewave

    Inputs/MultiInputSource.cpp: fixed mismatching allocation and deallocation

    Detected by Cppcheck:
    ``` Src/Inputs/MultiInputSource.cpp:87:3: error: Mismatching allocation and
    deallocation: CMultiInputSource::m_srcArray [mismatchAllocDealloc]
     delete m_srcArray;
     ^
    ```

Commit 5332d8f on 2023-12-02 by firewave

    fixed some `-Wunused-*` compiler warnings

Commit 391e1b9 on 2023-12-01 by dukeeeey

    Merge pull request #113 from gm-matthew/handbrake

    Dirt Devils does not use the handbrake
Commit da62fa8 on 2023-12-01 by gm-matthew

    Dirt Devils does not use the handbrake

Commit df0034a on 2023-11-18 by Ian Curtis

    fix minor error

Commit bd63400 on 2023-11-17 by Ian Curtis

    Optimise shader code (gm_matthew)

Commit 44fb32f on 2023-11-17 by Ian Curtis

    fix compilation on linux

Commit cdf5e4b on 2023-11-17 by Ian Curtis

    Fix transparency depth testing The two transparency layers, might not be
    separate layers at all. We believe the hardware is writing each layer to
    every other pixel (stipple alpha), then using an anti-aliasing filter to
    effectively blend the pixels. Because the pixels don't overlap they don't
    depth test against each other. We are using separate layers to emulate this,
    so the depth buffer must be saved and restored between the layers.

Commit de61835 on 2023-11-17 by dukeeeey

    Merge pull request #107 from gm-matthew/lod-blending

    Implement LOD blending
Commit 8778f98 on 2023-11-17 by Ian Curtis

    Add missing edge on transluency function

Commit ac53101 on 2023-11-16 by gm-matthew

    Implement LOD blending If two translucent polygons with opposing patterns
    overlap the result is always opaque Also the LOD scale calculation depends
    on Euclidean distance of x, y and z, not just z

Commit d726356 on 2023-11-13 by Ian Curtis

    Remove debug code

Commit 33d65e0 on 2023-11-13 by Ian Curtis

    Merge branch 'master' of https://github.com/trzy/Supermodel

Commit efe3077 on 2023-11-13 by Ian Curtis

    Fix some stencil issues The stencil mask must be set to all 1s otherwise we
    can't clear all the bits in the stencil buffer. Also we no longer need to
    save/restore the depth buffer into between priority layers.

Commit 3022e41 on 2023-11-11 by Bart Trzynadlowski

    Added some TODO notes to Main.cpp

Commit 47253d8 on 2023-11-11 by Bart Trzynadlowski

    Fixed graphics analysis -gfx-state option

Commit a57b379 on 2023-11-11 by Bart Trzynadlowski

    DSB.h: when generating a debug build in gcc w/ -g, k_framePeriod was causing
    a linker error. Not sure why k_timerPeriod wasn't but made them both
    constexpr.

Commit ccca9af on 2023-11-11 by Bart Trzynadlowski

    byte_layout attribute for region tag, intended to support Sega Racing
    Classic 2 ROM extracted from Yakuza

Commit c40d6ac on 2023-11-10 by Ian Curtis

    Rewrite the stencil buffer usage slightly, so both the LOS and layered polys
    work. LOS uses the top bit of the stencil buffer. Fixes some minor issues
    with draw order in sega rally.

Commit f9a0901 on 2023-11-09 by Ian Curtis

    Fix regression Scroll attenuation seems to disable ambient fog. If scroll
    attenuation is enabled it's telling the h/w to draw the real3d spot light on
    fog for the scroll fog layer. This should fix a regression where the
    background disappears on sega rally.

Commit a8976b5 on 2023-11-06 by Ian Curtis

    typo

Commit a065df2 on 2023-11-05 by Ian Curtis

    Correct scroll fog attenuation logic (gm_matthew) The attenuation value
    effects the spot light on fog for the scroll fog layer. The old implemention
    was nearly correct but the paramaters for the mix function were swapped.

Commit aecf617 on 2023-11-05 by Ian Curtis

    Various fixes
    - Scroll fog needed a break statement otherwise it could draw potentially
    multiples times.
    - The fog atttenuation paramater doesn't appear to effect scroll fog, other
    than to potentially enable it. This fixes some issues with emergency call
    ambulance.
    - Added support for the node discard attributes which sega rally2 uses to
    throw away specific geometry. This fixes junk that starts to show up in the
    levels after multiple runs.

Commit 628eeeb on 2023-11-03 by dukeeeey

    Merge pull request #101 from gm-matthew/lod-and-blending

    Add LOD switching and blending out culling nodes
Commit 8dda8e9 on 2023-11-03 by Ian Curtis

    Fix the line of sight function. Polygons have a line of sight value that's
    either a 0 or 1. We write this value into the stencil buffer, then read the
    app can read it back to determine the polygon attribute is visible or not.
    The returned value is 1/depth value in world coordinates. The first bit of
    the float is actually a control bit, 1 indicates no geometry hit, 0
    indicates geometry hit.

Commit 483dbdd on 2023-11-02 by gm-matthew

    Add LOD switching and blending out culling nodes We're not blending between
    LODs yet as we're not sure of the best way to implement it Also implementing
    control bits for disable culling and enable model scale (needed for ECA and
    Harley)

Commit 3a85bd9 on 2023-11-01 by Ian Curtis

    remove debug code

Commit d16ba90 on 2023-11-01 by Ian Curtis

    Update comments for the culling node function. gm_matthew worked out that
    the last 2 bits of the model scale float appear to be control bits for
    disable culling and valid model scale. Strange logic but the line of sight
    (LOS) function does something similar by re-using float bits for control
    words.

Commit 37065ce on 2023-10-31 by Ian Curtis

    Merge branch 'master' of https://github.com/trzy/Supermodel

Commit 8fea5bd on 2023-10-31 by Ian Curtis

    gm_mathew - the line scroll values, the top 16bits are the even lines, the
    bottom 16bits are the odd lines. We had it swapped. This fixes the number
    below the word results in daytona, when you finish a game.

Commit d42317d on 2023-10-30 by Bart Trzynadlowski

    Update README.md to define GIF widths (for Safari)
Commit acab274 on 2023-10-30 by Ian Curtis

    Set language standard to c++ 17

Commit f89da17 on 2023-10-27 by Ian Curtis

    Fix white gfx on linux / intel gpus. These optimisations originally came
    from toxieainc

Commit c5f9a3a on 2023-10-24 by Ian Curtis

    Amend the ambient fog logic to be disabled if fog density/start is set to
    zero. This fixes a regression where the sky stops drawing in lost world.

    Rewrite the scroll fog shader slightly. Scroll fog is a 2d layer, it has no
    depth, or can be considered to be drawn at infinity, therefore is not
    effected by the viewport light. The scroll attenatuion value also I think
    attenatutes the scroll fog alpha value. This fixes various effects in
    emergency call ambulance.

Commit a214c6d on 2023-10-22 by Ian Curtis

    Prior scroll fog logic was nearly correct. For scroll fogging to draw it
    needs either a start value or a fog density value, but these can come from a
    different viewport if they have the same colour fog set. This fixes the
    credits in vf3 which sets scroll fog, but it never draws on the original
    hardware.

    For a long time we've had bug reports that in vf3 the background in the
    Dural levels was the wrong colour, it should have been much darker. We
    thought it was again missing scroll fog, but gm_mathew worked out it was
    actually coming from the fogAmbient paramater. Not only does it darken the
    fog, but it also has an effect on the 2d background from the tilegen,
    similar to scroll fog. This also fixes the sky in lemans24.

Commit 2af0787 on 2023-10-17 by Ian Curtis

    Remove debug code

Commit 79ddc8c on 2023-10-17 by Ian Curtis

    Rewrite the logic for the scoll fog (render buffer clear colour). Each
    viewport can potentially have a scroll fog value so the logic of which value
    to pick wasn't immediately clear. TLDR I think it picks the highest value
    starting from the lowest priority layers. This fixes the sky in daytona
    battle on the edge.

Commit 86d4772 on 2023-10-15 by Ian Curtis

    Scroll fog is like a clear colour for the render target, but can be semi
    transparent as the tilegen layer can be shown underneath. I assumed that
    transprent polys would be drawn to the render target for transparent polys
    but unfortunately the logic does not work. So where else could it be drawn?
    Well with limited memory it must be draw to the buffer for opaque polys,
    which means this render target is also blended.

Commit 392900f on 2023-10-15 by gm-matthew

    Implement correct "round to nearest" mode On PowerPC round to nearest ties
    to even, not away from zero Also implement correct behavior for ppc_fresx 
    Fixes "tips to win" sequence in Daytona 2 BOTE

Commit d326410 on 2023-10-14 by Ian Curtis

    Merge branch 'master' of https://github.com/trzy/Supermodel

Commit 6595b93 on 2023-10-14 by Ian Curtis

    Rewrite the renderer a bit to spit out the finished graphics from the 3d
    chip on separate buffers. One buffer is for opaque pixels, and 2 more for
    translucent pixels. Before the frame was composited on the back buffer,
    which meant the tilegen had to have been drawn first. This way the images
    are now totally independant of the tilegen chip so can be drawn as soon as
    the register write 0xC is written to the tilegen.

    Some games update the tilegen after the ping_ping bit has flipped at 66% of
    the frame, so we need to split the tilegen drawing up into two stages to get
    some effects to work. So having the tilegen draw independantly of the 3d
    chip can make this happen.

Commit 0e494a0 on 2023-10-03 by Bart Trzynadlowski

    Updated build script to use SFTP (Supermodel3.com server upgrade no longer
    supports ftp)

Commit d656643 on 2023-09-27 by Ian Curtis

    gm_mathew worked out that the camera flashes in sega rally 2 have the
    'reset-matrix' attribute set inside the culling nodes, which was unsupported
    until now because we never found any games to have used this bit. The reset
    matrix is a bit strange, I would assume it would just reset the matrix back
    to identity, instead it seems to just reset the rotation back to identity
    whilst preserving the scale/position.

Commit e6b5468 on 2023-09-25 by Bart Trzynadlowski

    Update README.md: copyright dates.
Commit 7867c01 on 2023-09-25 by Bart Trzynadlowski

    Model3.cpp: Comment describing scan line timing.
Commit 801945d on 2023-09-03 by Matthew Daniels

    Whitespace
Commit 24d24db on 2023-09-03 by gm-matthew

    Ping-pong flip timing depends of the value of tilegen register 0x08 Also
    make the debugger display 16 bytes per line when using the "listmemory"
    command. Mirrored system registers can now be watched

Commit a00e8de on 2023-09-23 by Ian Curtis

    Update rules file. Not tested this but hopefully it should work

Commit 6b0d5c4 on 2023-09-23 by Ian Curtis

    Merge branch 'master' of https://github.com/trzy/Supermodel

Commit c6ea81d on 2023-09-23 by Ian Curtis

    Emulate the entire tilegen chip in a GLSL shader. (This is now possible with
    opengl 3+). The tilegen drawing was emulated on the CPU, but was one of the
    most expensive functions in the emulator according to a profiler. On a
    modern GPU it's pretty much free, because a GPU is a massive SIMD monster.

    Tilegen shaders are mapped to uniforms, and the vram and palette are mapped
    to two textures.

    TODO rip out the redundant code in the tilegen class. We don't need to
    pre-calculate palettes anymore. etc

    The tilegen code supports has a start/end line so we can emulate as many
    lines as we want in a chunk, which will come in later as some games update
    the tilegen immediately after the ping_pong bit has flipped ~ 66% of the
    frame.

    The scud rolling start tilegen bug is probably actually a bug in the
    original h/w implementation, that ends up looking correct on original h/w
    but not for us. Need hardware testing to confirm what it's actually doing.

Commit 38a9508 on 2023-08-28 by gm-matthew

    Prevent "ROLLING START" scrolling glitch in Scud Race There is a quirk of
    the tilegen chip that causes this glitch not to occur on real hardware, but
    we can't be sure what it is without testing Much easier to just patch all
    the Scud Race ROM sets

Commit c9b718e on 2023-09-01 by gm-matthew

    Fix Z80Debug.cpp compile with C++20 String literals should not be used to
    initialize char* but C++17 and earlier were letting it slide

Commit ebff9a8 on 2023-09-02 by Bart Trzynadlowski

    SCSP: added more detail to comment explaining 68K clock speed as recommended
    by Brian Troha

Commit b2fee42 on 2023-08-19 by gm-matthew

    DMA device register always returns Step 1.x PCI ID Step 2.x games by AM3
    request PCI ID this way and expect to see 0x16c311db

Commit 7924fed on 2023-07-05 by gm-matthew

    Set soundboard CPU to correct clock speed The 68K on the soundboard is rated
    at 12 MHz but runs at 11.2896 MHz, which is 256 cycles/sample with a 44100
    Hz sample rate. Removed SoundClock and Freq as they are not needed

Commit 015e8e9 on 2023-06-26 by gm-matthew

    Drive board uses the value 0xFF (not 0x00) on ports 42 and 45 to stop all
    effects when port 46 is set to 0xFF; fixes FFB effects continuing after game
    ends

Commit 84aa972 on 2023-05-16 by Matthew Daniels

    Add description of swtrilgy patch from MAME

    Star Wars Trilogy will sometimes write to a JTAG register and then read it
    back to verify it, but this doesn't work correctly with Supermodel's current
    JTAG implementation; it will try again and never succeed, getting stuck in
    an infinite loop. This patch stops the code from branching up and trying the
    write/read cycle again, allowing the subroutine to continue normally.
     We need to implement shift register widths correctly for each JTAG device
    (Mercury, Venus, Earth, etc.) for the unpatched code to work properly...
    right now Supermodel implements one JTAG device with a huge 197-bit data
    shift register
Commit ac0e640 on 2023-05-07 by Bart Trzynadlowski

    Games.xml: Removed seemingly unnecessary swtrilgy and swtrilgya patches,
    which should re-enable JTAG configuration and fix Endor Death Star tunnel
    shading

Commit 583d237 on 2023-05-06 by ToBul

    Fix GCC 13 build.

    MSYS and probably a few Linux distros are now shipping GCC 13. The new
    standards and issues are listed here,

    https://gcc.gnu.org/gcc-13/porting_to.html

Commit 18f3c23 on 2023-03-31 by CapitaineSheridan

    Work around to prevent I/O error after a while on fishing games with tension

Commit 813edb8 on 2023-04-05 by CapitaineSheridan

    Crosshair optimization:
    -create crosshairs only once in Init() instead of created them at every
    cycle
    -use matrix scale to apply the aspect ratio

Commit 6993dfc on 2023-03-28 by Bart Trzynadlowski

    Makefile.Win32: fixed detection of Windows Command Prompt for rmdir command

Commit 87de86f on 2023-03-17 by Bart Trzynadlowski

    Crosshair: use actual adjusted viewport resolution (configuration resolution
    is not the same thing). Fixes crosshair alignment for resolutions with
    non-Model 3 aspect.

Commit 95fc08e on 2023-03-15 by Bart Trzynadlowski

    Windows app manifest added to enable PerMonitorV2 DPI awareness

Commit f497de5 on 2023-03-15 by ToBul

    Update README.txt

    Fix old typos, leaving alone the American English. Remove the "Virtua
    Striker 2 '99" unlock code. I don't think it works,  others have also tried, 
    https://www.supermodel3.com/Forum/viewtopic.php?f=3&t=2096&p=19784#p19784

Commit 5fdb50a on 2023-03-14 by Bart Trzynadlowski

    README.txt: Updated Section 5 (video settings), squeezed everything back
    into 80 columns, changed 'Virtua Fighter 2 '98' to 'Virtua Striker 2 '98' in
    Section 10

Commit b5ca365 on 2023-03-14 by Bart Trzynadlowski

    NetBoard: guard against freeing null pointers and INetBoard needed a virtual
    destructor (was causing crashes on exit on some systems)

Commit 3848d27 on 2023-03-14 by CapitaineSheridan

    Autobuild script update following crosshair update

Commit abc2900 on 2023-03-13 by CapitaineSheridan

    m_crosshair is not needed and the config should be checked every frame
    because otherwise Alt-I command was broken (the number of crosshairs to
    render is mapped to a UI key) [Bart]

Commit a319bbe on 2023-03-12 by CapitaineSheridan

    Better optimisations. Moved all UI rendering from Main to CCrosshair. Take
    care of syntax about crosshair style.

Commit b19fdb6 on 2023-03-12 by CapitaineSheridan

    Change Media folder to Assets folder. Naming variable convention. Command
    line change -crosshair-style = vector|bmp. Config ini variable change
    CrosshairStyle = vector|bmp. CCrosshair::DrawCrosshair optimisation. 
    Modifying UpdateCrosshairs() to eliminate global variable.

Commit 1b066fe on 2022-12-23 by CapitaineSheridan

    Crosshair refactor
    -add a new separate class for crosshair
    -crosshair coordinates are calculated by matrix instead of recreating every
    object at correct coordinates
    -add ability to scale crosshair by dpi
    -add ability to use bitmap crosshair (located in ./Media/). 32bits bmp
    format + alpha
    -cmd line "-bitmapcrosshair" or "-vectorcrosshair" and/or
    BitmapCrosshair=0|1 in config file
    -these changes are only for lost world game with Crosshairs=1|2|3

    don't forget to copy the 2 crosshair images in Supermodel/Media folder

Commit 0cc1b37 on 2023-03-13 by Bart Trzynadlowski

    README.txt: Fixed a typo, end-of-sentence formatting in section 10, and
    added ECA cheat codes as per Brian Troha's suggestion

Commit 05c57ab on 2023-03-11 by Bart Trzynadlowski

    README.txt: fixed typo

Commit 188bb5c on 2023-03-10 by Bart Trzynadlowski

    README.txt: Added region codes to section 10 (c/o Brian Troha)

Commit 7e7686c on 2023-03-10 by Bart Trzynadlowski

    Games.xml: Corrected name of Virtual On 2 to Cyber Troopers Virtual-On
    Oratorio Tangram

Commit ae5af4c on 2023-03-04 by Bart Trzynadlowski

    Fixed window position config storage and command line parsing

Commit 83144f8 on 2023-03-04 by Bart Trzynadlowski

    Config nodes: added the ability to clear out node values or create empty
    leaf nodes

Commit 043f901 on 2023-03-04 by joachim

    comply to @trzy requested changes #

Commit 4a64a0d on 2023-03-04 by joachim

    comply to @trzy requested changes

Commit c301a57 on 2023-03-02 by joachim

    -pos=<x>,<y> -> correct error when not present

Commit c97e3ac on 2023-03-02 by joachim

    Add two video settings:
    -pos=<x>,<y>            Position [Default: centered]
    -borderless             Windowed mode with no border

    These 2 settings are usefull when setting up a multiplayer game on one
    machine.

    Example for 4 windows in fullHD:

    start "Master" /D"Master_P4" Supermodel.exe -input-system=sdl
    ..\ROMS\dayto2pe.zip -res=960,540 -borderless -pos=0,0 start "Slave1"
    /D"Slave1_P4" Supermodel.exe -input-system=sdl ..\ROMS\dayto2pe.zip
    -res=960,540 -borderless -pos=960,0 start "Slave2" /D"Slave2_P4"
    Supermodel.exe -input-system=sdl ..\ROMS\dayto2pe.zip -res=960,540
    -borderless -pos=0,540 start "Slave3" /D"Slave3_P4" Supermodel.exe
    -input-system=sdl ..\ROMS\dayto2pe.zip -res=960,540 -borderless -pos=960,540

Commit 5fa1904 on 2023-03-02 by joachim

    Allow joystick to be fetch if windows has no focus in SDL input mode. This
    is usefull for multiple instance of supermodel networked on the same
    machine.

Commit f7dcc1c on 2023-03-06 by Bart Trzynadlowski

    Games.xml and README.txt updated by Brian Troha (new Ocean Hunter ROM sets
    and several updates to version descriptions, etc.)

Commit ffc3209 on 2023-02-25 by The Brink of Tomorrow

    Update README.md with macOS build details

    Update README.md with macOS build and run instructions
Commit 78f5056 on 2022-12-28 by Aaron Paden

    Add missing filename to ErrorLog call.

Commit 7e025e6 on 2022-12-23 by CapitaineSheridan

    Memory allocated but not released.

Commit 69a458e on 2022-12-25 by Ian Curtis

    Make sure to invalidate texture memory when going fullscreen otherwise it
    might not be syned correctly.

Commit 11c01a3 on 2022-12-24 by Ian Curtis

    SDL_SetWindowFullscreen isn't needed as we call ResizeGLScreen which sets
    the fullscreen mode afterwards.

Commit 6dcf144 on 2022-12-24 by Ian Curtis

    We must explicitly call SDL_SetWindowFullscreen after setting the display
    mode otherwise the refresh rate doesn't change.

Commit ad0aed4 on 2022-12-19 by Ian Curtis

    be a bit more flexible with the display frequencies we'll match

Commit 2560451 on 2022-12-19 by Ian Curtis

    The model 3 (actually model 1,2 and 3) all used some crazy refresh rate of
    57.524hz. If your screen refresh rate is 60hz, supermodel will work fine,
    but really run too fast. Anyway if you create a custom refresh rate,
    supermodel will automatically pick this, and set it when the emulator is in
    fullscreen mode. Apparently this works with most monitors, even cheapo low
    end ones.

    To active set your command line to:
    -res=1920,1080 -vsync -fullscreen -true-hz

    Replace the resolution with whatever your monitor's native resolution is.

Commit 984f78a on 2022-12-17 by Ian Curtis

    Merge branch 'master' of https://github.com/trzy/Supermodel

Commit 57ead60 on 2022-12-17 by Ian Curtis

    Better opengl resource allocation. Fixes going fullscreen and back.

Commit 50d947d on 2022-12-14 by gm-matthew

    Add FileSystemPath.cpp to Visual Studio project

Commit d83e475 on 2022-12-06 by Fernando Casas Schössow

    Fixed build on macOS

Commit 07429b9 on 2022-12-04 by Fernando Casas Schössow

    Switched to Util::Format() to generate screenshot filename

Commit 317f8ba on 2022-11-21 by Fernando Casas Schössow

    First review code changes (identation and cosmetic changes)

Commit d11efb8 on 2022-11-15 by Fernando Casas Schössow

    Changed pathType from string var to enum

Commit df7787f on 2022-11-14 by Fernando Casas Schössow

    Keep Supermodel files (config, nvram, saves, etc.) in a predictable path
    when running on Linux

Commit db455ba on 2022-11-08 by toxieainc

    add undefd codepath for correct use of the 18bit DAC path (which seems to be
    triggered for all games)

    the volume correction to bring the data back into a valid range is not
    really needed in practice though, only Daytona2 seems to need it, and also
    only extremely rarely, so lets just live with a tiny bit of clamping for
    that game then

    while add it, make some formatting similar to MAME, and add one comment
    regarding a most likely wrong recent MAME change

Commit 153c510 on 2022-11-26 by Ian Curtis

    Fix uniform location.

Commit 7513cc7 on 2022-11-25 by Matthew Daniels

    Update Model3.cpp

    whitespace
Commit 6a392f2 on 2022-11-25 by gm-matthew

    Add comment

Commit 916d3d3 on 2022-11-22 by gm-matthew

    whitespace

Commit 2063303 on 2022-11-22 by gm-matthew

    Wait until IRQ2 is acknowledged before starting IRQ40 sequence. Fixes Ski
    Champ hanging at FBI screen when region set to USA

Commit 79a2414 on 2022-11-20 by gm-matthew

    Incomplete texture headers uploaded via FIFO are now discarded. Ski Champ
    sends one more 32-bit word via DMA than it should which was causing
    Supermodel to erroneously intepret the extra word as another texture

Commit acc7161 on 2022-11-13 by ToBul

    [Makefile.Win32] $MSYSTEM check never fails.

    'make clean' from Win command shell doesn't fully clean, condition always
    true.

Commit 5f30614 on 2022-11-09 by Ian Curtis

    FIx shaders failing to build on apple h/w

Commit fe36b07 on 2022-11-08 by Ian Curtis

    Update makefile

Commit e47258c on 2022-11-07 by Ian Curtis

    Remove some left over depreciated opengl functions. They were basically
    no-op anyway with shader path.

Commit 9348fd8 on 2022-11-07 by Ian Curtis

    Merge branch 'master' of https://github.com/trzy/Supermodel

Commit 40c8259 on 2022-11-07 by Ian Curtis

    Rewrite the whole project for GL4+. I figured if we removed the limitation
    of a legacy rendering API we could improve things a bit. With GL4+ we can do
    unsigned integer math in the shaders. This allows us to upload a direct copy
    of the real3d texture sheet, and texture directly from this memory given the
    x/y pos and type. This massively simplifies the binding and invalidation
    code. Also the crazy corner cases will work because it essentially works the
    same way as the original hardware.

    The standard triangle render requires gl 4.1 core, so should work on mac.
    The quad renderer runs on 4.5 core. The legacy renderer should still work,
    and when enabled a regular opengl context will be created, which allows
    functions marked depreciated in the core profiles to still work. This will
    only work in windows/linux I think. Apple doesn't support this.

    A GL 4.1 GPU is now the min required spec. Sorry if you have an OLDER gpu.
    GL 4.1 is over 12 years old now.

    This is a big update so I apologise in advance if I accidently broke
    something :]

Commit 4c727ab on 2022-11-03 by toxieainc

    change all INT16 mixing/intermediate clamping to floats

    also fixes 3 bugs: 1) mpeg right channel volume was always using the left
    channel volume, too 2) too high MusicVolume setting was not clamped to
    0..200 3) too high SoundVolume setting was not clamped to 0..200

Commit 121f81c on 2022-11-02 by Bart Trzynadlowski

    68K now uses run-time hooks for IRQ and instruction hook callbacks

Commit 33c04ce on 2022-10-29 by Bart Trzynadlowski

    Games.xml: Added re-dumped magtruck and retained old ROM set with patch
    (mgtrkbad) for users who fail to obtain newest MAME ROMs

Commit b2e5a20 on 2022-10-17 by Ian Curtis

    Fix build for linux (just missing header)

Commit f0872cc on 2022-10-16 by Ian Curtis

    Upgrade the glsl shader class a bit. Add a const char* comparitor so we can
    use a map with a simple "string" key. Fine to use when performance is not
    critical.

Commit 6fbf938 on 2022-10-16 by Ian Curtis

    Add a proper git ignore file .. And fix the legacy renderer to build with
    new visual studio (2019)

Commit e644751 on 2022-10-07 by toxieainc

    after clarifications, minor cleanup

Commit 75771b3 on 2022-10-05 by toxieainc

    remove 2 dupe function calls, where ProcessLos should actually be rather
    costly

Commit 412e923 on 2022-10-04 by Fernando Casas Schössow

    Fix new 3D engine texture2DLod function missing error on Linux/MESA. MESA
    requires needed extensions to be explicitly declared in code otherwise it
    will not enable them.

Commit aae58ed on 2022-10-03 by toxieainc

    change one more place with potential aliasing issue

Commit 5f97e5a on 2022-10-03 by toxieainc

    address review

Commit 519d695 on 2022-10-01 by toxieainc

    address review and move bit casts to new header

Commit 1b0e3be on 2022-08-25 by toxieainc

    use correct slash

Commit 22ffb5c on 2022-08-19 by toxieainc

    Harmless math warning fixes. map->unordered_map in CNew3D::CacheModel. Avoid
    general aliasing problems (using bit_cast or the recommended/optimized-away
    memcpy).

Commit 0eef09b on 2022-08-08 by toxieainc

    Revert "port some changes from MAME/MESS"

    This reverts commit b5f3d8b68e1ae62a08b00a730b6b8150758c0af3.

Commit d7041a8 on 2022-07-23 by toxieainc

    fix missing version GLSL compiler warnings on startup

    happens (at least) on Intel drivers

Commit 7052e83 on 2022-07-18 by toxieainc

    port some changes from MAME/MESS

Commit a60f998 on 2022-08-08 by toxieainc

    Revert "port some changes from MAME/MESS"

    This reverts commit b5f3d8b68e1ae62a08b00a730b6b8150758c0af3.

Commit 6c24bf7 on 2022-07-23 by toxieainc

    Optimize quad rendering by replacing the double computations for the area
    computation with the equivalent shoelace algorithm along with kahans
    algorithm to compute the cross products. In addition pull out the /w to the
    end, which increases precision even more.

    Thus making the result similar to the fp64 version, tested e.g. with harleys
    first person mode (all artifacts still gone as before).

    Improves perf on low end boards (e.g. Intel) dramatically (10fps with harley
    before, now 70 on my laptop), and even on higher end (e.g. NVIDIA laptop,
    quads are now faster than tri pipeline on my laptop).

Commit d919f38 on 2022-07-18 by toxieainc

    port some changes from MAME/MESS

Commit dda6d4c on 2022-08-16 by Bart Trzynadlowski

    Added some options to default Supermodel.ini that otherwise are not
    documented anywhere because README.txt is out of date

Commit e06788e on 2022-08-14 by ToBul

    Fix SUPERMODEL_DEBUGGER build. - Ian

    As requested by Bart; Ian's last SourceForge commit. Also, purely to appease
    my OCD, a change I forgot on my last 'Games.xml' PR that really shouldn't
    bother me as much as it does.

Commit d4633a8 on 2022-08-08 by toxieainc

    Revert "port some changes from MAME/MESS"

    This reverts commit b5f3d8b68e1ae62a08b00a730b6b8150758c0af3.

Commit 03002d3 on 2022-07-18 by toxieainc

    port some changes from MAME/MESS

Commit 2df681f on 2022-07-18 by toxieainc

    equivalent optimizations, NFC

Commit cb9d73c on 2022-07-18 by toxieainc

    port some (simple) changes from MAME/MESS

Commit d60feab on 2022-07-18 by trzy

    Merge pull request #1 from toxieainc/master

    Fix some errors, fix some performance warnings, and some simple (but
    effective) optimizations
Commit 7b87908 on 2022-07-18 by toxie

    Merge branch 'trzy:master' into master

Commit 27e071d on 2022-07-18 by toxieainc

    add comment

Commit dda48d7 on 2022-07-17 by toxieainc

    address review

Commit 24173a2 on 2022-07-17 by toxieainc

    revert "" -> std::string()

Commit c42a2d8 on 2022-07-17 by toxieainc

    check for __GNUC__ instead of just assuming it

Commit bd4399b on 2022-07-16 by trzy

    Merge pull request #2 from ToBul/master

    Games.xml: Align ROM names with MAME.
Commit e1e61c4 on 2022-07-12 by Tony Bulmer

    Games.xml: Align ROM names with MAME.

    Minor and inconsequential change. Makes manual ROM building 'slightly'
    easier when using latest MAME ROM set (currently 0.245).

Commit abb1273 on 2022-07-11 by toxieainc

    slightly optimize DrawTileLine (less branches/logic)

Commit b98be0c on 2022-07-11 by toxieainc

    optimize ReadRegister, as localtime (at least on MSVC/Windows) is
    significantly showing up in profiling

    e.g. in SW Trilogy

    caches localtime until time changes (i.e. limits calls to 1/sec)

Commit ba7145d on 2022-07-11 by toxieainc

    use intrinsics for endian swap

Commit b2b8396 on 2022-07-11 by toxieainc

    missed 2 changed files with last commit

Commit 4961951 on 2022-07-11 by toxieainc

    fix a real error (m68kdasm) and some harmless performance warnings and use
    modern headers

Commit e0053b3 on 2022-07-11 by toxieainc

    fix some real errors (Model3,InputSystem), minor errors (SCSP,SDL/Main) and
    some performance warnings

Commit 46eff8c on 2022-06-23 by Bart Trzynadlowski

    supermodel_build_bot.py: New build bot script for GitHub

Commit 35a7e8a on 2022-06-25 by trzy

    Update README.md
Commit 237f275 on 2022-06-23 by Bart Trzynadlowski

    Rules.inc: whitespace error

Commit d42f7d6 on 2022-06-23 by Bart Trzynadlowski

    Rules.inc: Fixed git revision parsing

Commit 8c8cc3d on 2022-06-23 by Bart Trzynadlowski

    Preparing for Git snapshot build bot and updated copyright header to
    reference full lifespan of Supermodel repo

Commit 3803d6b on 2022-06-20 by trzy

    Updated Supermodel.ini comment to reference Supermodel.log instead of
    error.log
Commit 015b960 on 2022-06-20 by Bart Trzynadlowski

    Corrections to README.md and Supermodel.ini

Commit 6460a06 on 2022-06-20 by Bart Trzynadlowski

    Removed disassembly of Scud Race Plus

Commit 849f2e8 on 2022-06-20 by trzy

    Update README.md

    Trying again to make the images fit side-by-side.
Commit ff7f6d5 on 2022-06-20 by trzy

    Update README.md

    Fixed image layout and added some links.
Commit b58b5ce on 2022-06-20 by Bart Trzynadlowski

    Smaller Real3D logo

Commit 81a43e0 on 2022-06-20 by Bart Trzynadlowski

    Added initial REAMDE.md

Commit a742a73 on 2022-06-19 by Ian Curtis

    njz3: Fix sound volume with new quadrophonic audio code

Commit e3bc56c on 2022-06-14 by Ian Curtis

    values must be signed, to handle negative numbers, my bad 

Commit 78eab65 on 2022-06-14 by Ian Curtis

    minor fix

Commit 7be381c on 2022-06-12 by Bart Trzynadlowski

    Added configurable refresh rate. Default is 60 but -true-hz on the command
    line enables 57.524 Hz mode, which some players report making Spikeout and
    the timing of combos in VF3 more authentic. The config key is RefreshRate
    and is specified as a floating point value. Should the desire arise,
    arbitrary refresh rates can be set in Supermodel.ini using this key but
    almost no validation of the value is performed. Thanks to forum user trap15
    for the original code and PonMi for advocating for this patch to be
    included.

Commit 83e7b5f on 2022-06-09 by Ian Curtis

    njz3: Proper quadrophonic audio support. Need 4+ speakers to take advantage
    otherwise down mixed to stereo.

Commit 5e61e2f on 2022-03-31 by Bart Trzynadlowski

    Added default values for PortIn, PortOut, and AddressOut

Commit 84ae9df on 2022-02-06 by Ian Curtis

    fix: explicit specialization requires 'template <>'

Commit fe7baa1 on 2022-02-05 by Bart Trzynadlowski

    Bitmap size field in the BITMAPV4HEADER was not being set correctly (the
    image viewers I've been using are unaffected by this but it ought to be
    corrected)

Commit 3f6937e on 2022-02-05 by Bart Trzynadlowski

    When dumping textures, two T1RGB5 texture maps are written: with and without
    contour processing. Contour processing can be disabled per-polygon and some
    textures intended to be used without contour processing contain pixels with
    T=1 that are clearly not supposed to be transparent. We now output
    textures_t1rgb5_contour.bmp and textures_t1rgb5_opaque.bmp.

Commit 32933ef on 2022-02-01 by Bart Trzynadlowski

    Added a -dump-textures option (config key DumpTextures) that writes texture
    BMP files, one for each known format (12 in all currently)

Commit 646b46f on 2022-01-23 by Matthew Daniels

    Adjust the number of PowerPC cycles executed per frame to assume 57.524 Hz
    refresh rate

    Fighting Vipers 2 attract mode no longer requires 100+ MHz to run smoothly

    Supermodel itself still runs at 60 Hz

Commit 0ef9089 on 2022-01-02 by Ian Curtis

    Double buffer the line of sight values. Should allow the correct values to
    be returned if GPU threading is enabled. 

Commit 1a7e319 on 2021-12-29 by Matthew Daniels

    Fixing Magical Truck Adventure bad ROM dump which was causing attract mode
    to stop rendering

Commit bc774ca on 2021-12-19 by Ian Curtis

    methods should be marked as noexcept because they cannot throw

Commit b8aac0a on 2021-12-17 by Ian Curtis

    white space.

Commit 6f7130c on 2021-12-17 by Ian Curtis

    Better document real3d registers

Commit 6ff099b on 2021-12-17 by Matthew Daniels

    Clear all Real3D DMA registers on reset; uninitialized values were causing
    issues with MSVC debug builds

Commit 3e9c887 on 2021-12-15 by Matthew Daniels

    Another header include fix

Commit 94772ae on 2021-12-15 by Matthew Daniels

    Include required header files in Model3GraphicsState.h - fixes potential
    build errors in GCC

Commit 0c47ac8 on 2021-11-30 by Matthew Daniels

    Increase MIDI FIFO buffer size for SCSP; should prevent FIFO overflows

    Also use similar macro for DSB FIFO buffer

Commit e93c5d7 on 2021-11-28 by Matthew Daniels

    PowerPC IRQ line deassertion; DMA only fires interrupts when required

    PowerPC no longer clears its own IRQ line; it is now cleared by the IRQ
    controller when there are no more pending interrupts. Not all games clear
    DMA interrupts so it was necessary to tweak the 53C810 SCSI controller and
    the Real3D DMA interface to only fire interrupts if a certain register is
    correctly set. 53C810 has the documented DIEN (DMA Interrupt Enable)
    register; Real3D DMA seems to use the low bit of the dmaConfig register.

    Also I removed the net IRQ as no games seem to actually use it.

Commit 9ffce8b on 2021-11-22 by Matthew Daniels

    Getting rid of most of the includes from Supermodel.h; each file now
    explicitly includes the header files it needs.

    Making changes to a header file should no longer force the entire project to
    recompile.

Commit 50465f9 on 2021-11-05 by Matthew Daniels

    Shared RAM on simulated netboard is now double-buffered; Spikeout Final
    Edition no longer requires a hack to work. Preparing to enable simulated
    netboard to run in its own thread

Commit b621106 on 2021-10-30 by Matthew Daniels

    New JTAG patch for Sega Rally 2; linked mode now fully working. Few other
    minor changes

Commit 94d8a1b on 2021-10-26 by Matthew Daniels

    Increase size of DSB2 FIFO buffer; fixes Sega Rally 2 music occasionally not
    switching or fading out

Commit b18f6d1 on 2021-05-09 by Matthew Daniels

    Non-linked games no longer falsely list the net board under extra hardware

Commit b7755c9 on 2021-05-09 by Matthew Daniels

    Fix 16-bit variables not displaying correctly in Z80 debugger (at least in
    MSVC builds)


Commit a20366d on 2021-04-19 by Matthew Daniels

    Fix netboard resets and detection of simulated netboard linking to itself

Commit 5a570ce on 2021-04-14 by Matthew Daniels

    Adding simulated netboard; all linked games except Sega Rally 2 working at
    full speed.

Commit 98d4f1c on 2021-03-22 by Matthew Daniels

    Corrected alignment issues with netboard RAM; send/receive parameters no
    longer need to be manipulated to work correctly. Fixes bugs including Scud
    Race linked car orientation.

Commit 4b9221e on 2021-03-15 by Ian Curtis

    Add some async network methods. Async send copies to an internal buffer and
    sends in another thread so doesn't block at all. Added CheckDataAvailable
    method to the TCPReceive class. Can check or wait for data without blocking.

Commit 020e2d1 on 2021-03-13 by SpinDizzy

    DSB2 clock speed from 4MHz to 11MHz.

Commit eb61bd9 on 2021-03-07 by SpinDizzy

    New dumps added: Virtua Striker 2 '99.1 (Step 1.5, Export, USA, Revision B) 
    Virtua Striker 2 '99.1 (Step 1.5, Japan, Revision B) Thanks to Brian Troha
    and The Dumping Union

Commit de470aa on 2021-03-07 by SpinDizzy

    Fixed cycle counting overflow that was causing DSB2 to halt execution.

Commit 6cf9974 on 2021-03-03 by Ian Curtis

    missing header

Commit 46b1de2 on 2021-03-03 by SpinDizzy

    DSB2 IRQ 2 now fires at 1KHz rather than once per frame, improving  music
    timing in Daytona USA 2 and Sega Rally 2. DSB1 CPU timing  increased from
    1MHz to 4MHz, improving music fade timing in Scud Race.  Thanks to
    gm_matthew for these discoveries.

Commit e337425 on 2021-02-20 by Ian Curtis

    Just a few compile warnings

Commit 0bfc2ae on 2021-02-19 by Ian Curtis

    Fix project file

Commit f23b2f9 on 2021-02-18 by SpinDizzy

    Build bot script.

Commit e4998f5 on 2021-02-18 by SpinDizzy

    Skichamp Driveboard error fix.

Commit 08d4735 on 2021-02-18 by SpinDizzy

    Huge refactor of the Driveboard:
    -Separate each possible boards (wheel, joystick, skipad, billboard).
    -Defined a Driveboard type in Games.xml for each games.
    -Due to the refactoring, Driveboard Savestates have changed (a common base
    data + a specific board data are saved).
    -Backwards compatibility with previous save states is maintained.
    -Driveboard rom section is no longer required anymore. This disables
    Driveboard emulation in case the rom is not found.
    -Added Billboard emulation (vf3, vs2, fvipers2, von2). 7 segments and lamps
    Outputs are redirected to Supermodel outputs.
    -Changes project to C++ 17 standard.

Commit ab36777 on 2021-01-30 by SpinDizzy

    Introduce new config keywords.

    In xinput mode, lets the choice to have left and right gamepad motors
    vibrate together. XInputStereoVibration = 1 (both motors) [default] 
    XInputStereoVibration = 0 (separate motors)

    In sdl input mode, new control option to set minimum strength above which a
    Model 3 constant force command will be simulated on an sdl gamepad device. 
    SDLConstForceThreshold = 30 [default] note : the vibration strength can be
    mod with SDLConstForceMax = [val]


Commit 6627b1e on 2021-01-05 by SpinDizzy

    The controls of bass/getbass games were no longer good, following recent rom
    set name changes in Games.xml.

Commit 2fa9c8d on 2021-01-05 by SpinDizzy

    Corrected game set names & game titles (Brian Troha) New clones
    -Sega Bass Fishing Deluxe (USA) (Brian Troha)
    -Dirt Devils (USA, Revision A) (Brian Troha)

    ------------ Note/Warning: This will match future Mame 0.228 changes. Please
    Check/Edit carefully all your rom set because there are many changes (e.g.
    set name). Thinking of delete or rename your nvram, or restart from the
    start.

Commit 6626afc on 2020-12-30 by SpinDizzy

    Added rumble skipad in Ski Champ game. This is only a simulation (cmd are
    interpreted), not an emulation because we lack informations on the game
    itself and its driveboard (does it have dedicated cpu/rom ?).

Commit f3a7ec5 on 2020-12-30 by SpinDizzy

    Service menu activated in swtrilgyp. Note : The "feedback stick test" always
    failed (change cab type DX to SD if needed).

Commit c8f47cc on 2020-12-30 by SpinDizzy

    - Add -O3 optimization option for OSX and Linux.
    - Revert netboard to disable by default.

Commit 34c24c0 on 2020-12-28 by SpinDizzy

    - Suppressed net board console output (using debug logging now)
    - Added GPL headers to net board source files

Commit 8df56dd on 2020-12-27 by SpinDizzy

    - Star Wars Trilogy untextured Death Star surface bug in gcc builds was
    caused by -Ofast (thanks to rokfpoewrkcpoqwkcp for discovering this). Using
    -O3 now.
    - Net board enabled by default.


Commit cce9914 on 2020-12-25 by SpinDizzy

    Updated rules file.

Commit 53185c3 on 2020-12-24 by SpinDizzy

    Introducing a special release build target with SVN version stamping.

Commit 1f50519 on 2020-12-21 by SpinDizzy

    Magical Truck Adventure rom patch to unlock region. (enter service menu,
    then Start P1, Start P1, Service, Start P1, Service, Test)

Commit 72c0c60 on 2020-12-20 by Ian Curtis

    -wide-bg option to stretch the background tile layer when wide-screen mode
    is enabled - Daro Land

Commit 7d2b21a on 2020-12-01 by SpinDizzy

    Change the duration of constant force effect in SDL2 to correct daytona
    games


Commit d70a79a on 2020-11-06 by SpinDizzy

    Makefiles update

Commit 5143669 on 2020-10-24 by SpinDizzy

    Star Wars Trilogy lever feedback activated

Commit 402baf8 on 2020-10-21 by SpinDizzy

    Force Feeback for Linux

Commit db01349 on 2020-09-19 by Ian Curtis

    hopefully fix building on linux

Commit 981cdc3 on 2020-09-19 by SpinDizzy

    Added Screenshot feature (ALT+S).

Commit ca57659 on 2020-09-16 by SpinDizzy

    Remove using namespace from headers.

Commit c0729c1 on 2020-09-15 by Ian Curtis

    fix compiling with visual studio

Commit 75f8ee6 on 2020-09-14 by Ian Curtis

    fix missing polys with quad renderer

Commit 309537d on 2020-09-14 by SpinDizzy

    Balance initial value set to "0" rather than "false". "False" is interpreted
    as 0 when decoded as a float, so there will be no effect here, but this
    expresses the intended logic better.

Commit 72b2c0e on 2020-09-14 by SpinDizzy

    Fixed a longstanding bug that caused stereo channels to be flipped
    incorrectly. This was due to the initial audio buffer write position being
    aligned to the middle of a 4 byte (2 byte left, 2 byte right) audio sample. 
    In multi-threaded mode, some sort of race condition caused this alignment to
    be fixed until audio playback was temporarily paused (via pausing, loading a
    state, etc.) Audio playback should now be fixed and work consistently in all
    cases.


Commit 38c3ff6 on 2020-09-08 by Ian Curtis

    Clip scroll fog to viewport, instead of filling entire screen. Fixes fog bug
    in Daytona credits.

Commit aaa26f6 on 2020-09-07 by Ian Curtis

    Update scroll fog logic. Should fix the credits in virtua fighter and
    daytona that have fogging applied to the background layer and shouldn't.

Commit 1d37358 on 2020-09-04 by SpinDizzy

    - Add missing library from previous windows makefile update. 
    - NEW_FRAME_TIMING build option was removed because it was no longer used.

Commit 95488ac on 2020-09-03 by SpinDizzy

    Windows makefile update

Commit 60c923f on 2020-08-31 by SpinDizzy

    - Reworked logging system to support output to multiple outputs: files,
    stdout, stderr, and syslog (OutputDebugString on Windows, syslog on other
    systems).
    - Added ordinal logging level: all, debug, info, error. Compiling with DEBUG
    defined is no longer necessary for debug logging. Be careful when logging
    debug level because it produces copious output.
    - Fixed -enter-debugger options when compiled with SUPERMODEL_DEBUGGER.
    - Command line parse errors no longer ignored; Supermodel will now exit.


Commit 6491ed4 on 2020-08-26 by SpinDizzy

    Rename command line SCSP option to -legacy-scsp and -new-scsp


Commit d395506 on 2020-08-26 by SpinDizzy

    Correct the returned value from previous getbass patch There was a strange
    side effect on Ocean Hunter game

Commit 9e8e993 on 2020-08-25 by SpinDizzy

    Getbass related Game now boots in game Add missing analog input and buttons

Commit 97d61a9 on 2020-08-24 by SpinDizzy

    Small oversight in the new SCSP code

Commit 816d541 on 2020-08-22 by Ian Curtis

    Various SCSP improvements and code cleanup:
    - Ported MAME's implementation
    - Corrected FM sound for songs in VF3 that use it 
    - Music tempo now closer to real hardware thanks to emulating two SCSP
    chips.
    - Add LegacySoundDSP config option for games with SCSP DSP glitches (ex.
    engine noises in Sega Rally 2, and loud garbage on Bahn's stage in Fighting
    Vipers 2)
    - Renamed SysFPS to "SoundClock" (since raising this appears to adjust the
    sound frequency).

    (Submitted by Paul Prosser)

Commit cf64375 on 2020-08-22 by SpinDizzy

    New ecap crc rom + rom patch

Commit 27cda80 on 2020-08-22 by SpinDizzy

    Low level z80 core bug Discovered when looking at lemans ffb board (real
    Model3 rom, not Model2 stcc one) This could affect other ffb board and/or
    scud race DSB1 sound board However I didn't notice any bad things

Commit d77433b on 2020-08-01 by Ian Curtis

    Remove some debug code

Commit ba917ac on 2020-07-31 by Ian Curtis

    Update glew version

Commit 5e434e2 on 2020-07-27 by Ian Curtis

    build fixes for mac

Commit 858b84f on 2020-07-27 by SpinDizzy

    skichamp rom patch : prevent driveboard error at boot

Commit 6af8825 on 2020-07-03 by SpinDizzy

    Linux build compatibility changes

Commit 12081a9 on 2020-07-03 by SpinDizzy

    Update makefiles

Commit f7d75b7 on 2020-07-01 by Ian Curtis

    silence a few casting warnings

Commit a73a06d on 2020-07-01 by SpinDizzy

    Netboard related : Define a new hardware line in games.xml

Commit 345310f on 2020-06-17 by SpinDizzy

    _rotl() replacement

Commit 9a651bd on 2020-06-17 by Ian Curtis

    Only try and make a connection if we have enabled EmulateNet

Commit 1a3407b on 2020-06-17 by Ian Curtis

    Fix corrupted memory in network mode

Commit 597ea93 on 2020-06-16 by Ian Curtis

    remove old files

Commit daef27b on 2020-06-16 by Ian Curtis

    Replace network code with new tcp implementation

Commit 8071efb on 2020-06-13 by Ian Curtis

    Add SDLnetworking + some basic tcp classes

Commit d688fdd on 2020-06-12 by Ian Curtis

    cache variable, without it is unusably slow in debug mode

Commit 8b4a4ec on 2020-06-10 by Ian Curtis

    add missing file

Commit 90df4af on 2020-06-08 by Ian Curtis

    update vs project files

Commit f69d7c7 on 2020-06-08 by Ian Curtis

    fix output values

Commit 6f6c98c on 2020-05-11 by Ian Curtis

    The translator map seems to work with paletted colour values too, and the
    result is not clamped. Colours are passed to the GPU as unsigned bytes to
    multiplying by 16 will overflow, so we do the logic in the shader. If we
    passed floats we could skip the shader logic.

Commit e2ad593 on 2020-05-07 by Ian Curtis

    When the translator map is enabled the colour values seem to use 4 bits,
    0-16. Why not 0-31 I don't know. Anyway virtua on is overflowing these
    values for some of the smoke effects. For us the numbers were wrapping
    around. Clamping is enough to fix this. (Thanks model123)

Commit 5b40605 on 2020-04-24 by SpinDizzy

    custom resolution bug fix

Commit 4f347ff on 2020-04-19 by SpinDizzy

    SDL1 to SDL2 internal changes

Commit 14e2a50 on 2020-01-22 by Ian Curtis

    remove files

Commit 247921d on 2020-01-22 by SpinDizzy

    

Commit 5ace960 on 2020-01-04 by Ian Curtis

    tweak value so the map indicator still works at the start of spikeout

Commit 6b8a291 on 2020-01-04 by Ian Curtis

    Sutherland-Hodgeman clipping only works for planar primitives. Scud is
    rendering non planar quads and this breaks our algorithm a little when using
    the quad renderer. Stretching the near plane a few % is enough to fix it.

Commit d711afa on 2020-01-04 by Ian Curtis

    Only need to clip against 4 planes for poly culling

Commit 605dad2 on 2019-12-27 by Ian Curtis

    Fix the sky in harley in first person mode where pixels in the sky are
    culled when they shouldn't be with the quad renderer. Basically we were
    running out of precision in a rare corner case. Doing the maths with double
    precision and casting back to float was enough to fix the issue.

Commit a5858e6 on 2019-12-10 by Ian Curtis

    To calculate the mipmap value the opengl needs to calculate the difference
    in the texture coordinates between adjacent pixels. If there are no adjacent
    pixels the hw may need to spawn invisible fragments outside of the visible
    polygon. Our algorithm was culling these out so the mipamp calculation
    failed for edge pixels. Anyway this fixes quad rendering looking broken with
    the latest nvidia drivers.

Commit 06688de on 2019-11-15 by Ian Curtis

    Really should cull against 5 frustum planes.

Commit e1827cf on 2019-11-08 by Ian Curtis

    Stretch the time the ping_pong bit flips a bit to bump writes into correct
    frame. Otherwise they end up getting written at like 99.8% of the frame and
    overlapping with the next.

Commit ed2816c on 2019-11-08 by Ian Curtis

    Optimise Games.xml (Krom)

Commit 2e6968f on 2019-11-07 by Ian Curtis

    Thanks to our anonymous contributor who managed to patch a model 3 game and
    run it on real h/w, we were able to obtain some timing values we had been
    long been missing for correct emulation. The h/w polls something called the
    ping_pong bit at start-up to sync the GPU with the CPU. Unfortunately we
    didn't have the correct values and just manually used some per game hacks to
    get games to run. This mostly worked, but some games were writing more or
    less frames than they should have been for a given time period. When the
    ping_pong bit flips at 66% of the frame time, games were writing data for a
    new frame, which meant writes were often straddling 2 separate frames. We
    aren't 100% sure if IRQ2 or the ping_pong bit is vblank. 

Commit 06b43c4 on 2019-11-02 by Ian Curtis

    Ocean hunter in the middle of the game is passing a few matrices with
    FLT_MAX inside them, which blows apart our near/far calculation. The
    hardware must have some method to sanitize the near/far for instance if you
    render something extremely close to the origin you will also blow apart any
    near/far calc.

Commit 714f6a1 on 2019-07-28 by Ian Curtis

    skip invalid culling nodes

Commit 17f4103 on 2019-07-26 by Ian Curtis

    A culling node must follow immediately after a viewport.

Commit cc83000 on 2019-02-22 by Ian Curtis

    Fix possible negative pointer arithmetic which was causing the music to wrap
    around.

Commit a76e46e on 2019-02-21 by Ian Curtis

    Update makefiles 

Commit 8aa476a on 2019-02-21 by Ian Curtis

    Previous mpeg audio code worked for me in debug mode, but the release build
    had some serious corruption issues when being built in visual studio.
    Replaced the old spaghetti code with a more modern implementation.

Commit c86dab0 on 2019-02-19 by SpinDizzy

    - Disabled Alt-O (dump timings) and Alt-U (dump input state) when 
    SUPERMODEL_DEBUGGER not defined
    - Option '-print-inputs' works again if no ROM set specified


Commit 6fe878b on 2019-02-07 by SpinDizzy

    Restore Sega Rally 2 music in some practice stages 


Commit 383e6ac on 2019-02-05 by SpinDizzy

    Fixed MPEG music looping glitch.

Commit bb89bc8 on 2019-02-05 by SpinDizzy

    Refactor code from previous svn768 (Note : break old savestate
    compatibility)

Commit 90faeaf on 2019-02-01 by Ian Curtis

    Fix the audio in sega rally2. The game has packed separate tracks into the
    left and right audio channels. Selecting the correct track fixes the garbled
    mess. (Spindizzi)

Commit f634d19 on 2019-01-24 by Ian Curtis

    Fix shader. Why this worked on my nvidia card .. I'm not sure.

Commit 8094c2e on 2019-01-21 by Ian Curtis

    Composite the alpha layers at the end of rendering. To do this we need to
    mask the alpha pixels with the opaque pixels from the next priority layer.
    Fixes some overlapping shadows in vf3tb that have different priority layers.
    I assume that was a game bug, but it works on the real h/w.

Commit c171356 on 2019-01-20 by SpinDizzy

    -Added support for specifying multiple sections simultaneously in INI 
    files, e.g., [ daytona2, dayto2pe ]
    -Forgetfulness in 763

Commit 864bb36 on 2019-01-16 by Ian Curtis

    Technically backface culling should cull polys when the dot product = 0,
    since when it equals 0 the polys is perpendicular to the camera and thus
    invisible. When you complete sega rally2 (if you can get that far lol) the
    champagne is invisible. The reason is because all the face normals are 0. So
    when multiplied by the model matrix they are still 0 and get culled.
    Tweaking the condition to only cull polys when greater than 0 fixes this,
    and allows these 'bad' polys to render as they do on the model3.

Commit 98106df on 2019-01-13 by SpinDizzy

    Inputs can now be configured on a per-game basis

Commit b0813ef on 2019-01-13 by SpinDizzy

    - PCI bridge can be specified in Games.xml for games that use a different
    version than expected based on stepping. No more hard-coded exceptions.
    - Real3D PCI ID can be specified in Games.xml for exceptions that require
    it.
    - Real3D status bit timing specified in Games.xml for exceptions that
    require it.

Commit d8f736e on 2019-01-06 by Ian Curtis

    timing fix for srally2x (Spindizzi)

Commit 24cbc40 on 2019-01-06 by Ian Curtis

    Updates games.xml for new prototype games (Spindizzi)

Commit 0f49e0b on 2018-12-08 by Ian Curtis

    remove commands that do nothing

Commit 08eeb20 on 2018-12-08 by Ian Curtis

    clear scissored out areas

Commit 6075352 on 2018-11-27 by Ian Curtis

    Optimise depth calculation for quad shader.

Commit 6eaf204 on 2018-11-11 by Ian Curtis

    Update Games.xml for games that were recently dumped

Commit 78039e9 on 2018-10-28 by Ian Curtis

    

Commit 538a714 on 2018-10-23 by Ian Curtis

    Fade out microtextures with higher textures LODs. Fixes incorrectly applied
    microtextures in LA machineguns.

Commit b8da2d6 on 2018-10-22 by Ian Curtis

    debug code should be disabled

Commit bf4d725 on 2018-10-19 by Ian Curtis

    Smooth texture repeat makes no sense for alpha/contour textures with pixel
    dilate. HW seems to treat them as non smooth anyway. Also implement some
    line of sight stuff, used by Scud. Really need to check the
    threading/synchronisation of the register reads but it basically works. 

Commit df02d6a on 2018-10-18 by Ian Curtis

    return correct PCI ID

Commit 3dc69f8 on 2018-10-17 by Ian Curtis

    fix debugging code

Commit 42eaa7f on 2018-10-14 by Ian Curtis

    fix copy paste error :]

Commit 3eeec35 on 2018-10-13 by Ian Curtis

    Correctly emulate the real3d pro-1000 texture modes in our shader.

Commit a5ad1c4 on 2018-10-07 by Ian Curtis

    Revert some patches (Spindizzi)

Commit 7bcc70a on 2018-10-01 by Ian Curtis

    Update makefile 

Commit 3dbfa73 on 2018-09-19 by Ian Curtis

    Both lemans24 and dirt devils are rendering some totally transparent objects
    as the first model in each viewport. The HW seems to skip these entirely,
    and including them breaks our near/far plane calculations leading to Z
    fighting. 

Commit e3255e4 on 2018-09-18 by Ian Curtis

    Tweak the ratio for the near plane. Fixes missing text on the map indicator
    at the start of spikeout. The vertices are 0.003 units apart so a tonne of
    precision is required at the near plane to not depth test out.

Commit c5b156d on 2018-09-18 by Ian Curtis

    remove debug code

Commit 5dab991 on 2018-09-16 by Ian Curtis

    avoid double matrix load

Commit a7e851d on 2018-09-14 by Ian Curtis

    updated command line help

Commit b5f9ad9 on 2018-09-13 by Ian Curtis

    Quad rendering engine. Set QuadRendering = 1 in the ini file to use, or
    -quad-rendering at the end of the command line to use.

Commit e4f5f0b on 2018-09-02 by Ian Curtis

    Sega bass fishing is starting some off the meshes with a shared vertex.
    Without having a previous vertex to share with, the values were just
    uninitialised and junk. The hardware itself seems to hold the value from the
    previous mesh, and sega bass fishing relies on this undefined behaviour to
    function correctly. Storing the cached vertices as member variables instead
    of locally to the function means it holds the values from the last mesh, and
    uses those values instead of random garbage.

Commit c1c2ded on 2018-08-09 by Ian Curtis

    remove dead variable

Commit 9f5a19a on 2018-07-09 by Ian Curtis

    Use correct PCI IDs (Harry Tuttle)

Commit 80cf1f9 on 2018-07-04 by Ian Curtis

    add crosshair effect (Harry Tuttle)

Commit 73ff32a on 2018-06-18 by Ian Curtis

    fix cast, and remove extra ; that had snuck into the shaders

Commit f9be125 on 2018-06-17 by Ian Curtis

    Opaque polys definitely depth test with less than or equal. Virtua striker
    is drawing co-planar polys for the bar gfx

Commit c7ffd0a on 2018-06-16 by Ian Curtis

    Draw transparent polys to separate layers and composite at the end. This
    solves a tonne of transparency errors we had been battling with for a long
    time. The model3 is strange in the fact it only supports a max of two
    translucent overlapped polys. They are not blended into the frame normally.
    Doing this means the topmost translucent polys only are visible in the
    scene, the equivalent of doing a depth pass first, but without the added
    cost.

Commit 7497e3f on 2018-05-29 by Ian Curtis

    forgot parentheses

Commit f9f43ca on 2018-05-28 by Ian Curtis

    

Commit c1e0e37 on 2018-05-28 by Ian Curtis

    optimise hash function and parse out transparency layer select bit

Commit 2af58f4 on 2018-05-24 by Ian Curtis

    select correct bit ..

Commit 8452d4c on 2018-05-24 by Ian Curtis

    Parse out TranslucencyPatternSelect

Commit 4d5c9e7 on 2018-05-06 by Bart Trzynadlowski

    Fixed Makefiles for non-MSYS builds. Unfortunately had to revert a lot of
    the automated configuration that Harry had inserted. Users will again need
    to hand-edit Makefile.Win32 for their configuration. Verbose warnings
    (-Wall) have been restored.

Commit 4d275ba on 2018-05-03 by Ian Curtis

    Parse siblings after children. It's the only way to get Daytona water fall
    to look correct without h/w poly sorting.

Commit 3f9b1b7 on 2018-05-03 by Bart Trzynadlowski

    Added a stretch mode (-stretch) and modified new renderer's viewport code
    (hopefully correctly) to handle aspect ratio correction and widening the way
    the legacy one does.

Commit c6b86c0 on 2018-05-02 by Ian Curtis

    Render the opaque part of texture transparency in the first pass. Fixes
    numerous transparency related issues in lemans24 and some in ocean hunter. I
    don't really know if this is the 'correct' way of solving this because the
    real3d pro-1000 is a complete black box. There is still a lot we don't
    understand about how it handles transparency. 

Commit 11a3cdb on 2018-04-24 by Ian Curtis

    remove unnecessary copies 

Commit b39de6e on 2018-03-30 by Ian Curtis

    Fix controller inputs (Spindizzi)

Commit 9349414 on 2018-03-26 by Bart Trzynadlowski

    Inconsequential typo

Commit b5f1b7a on 2018-03-26 by Bart Trzynadlowski

    Reverted version to 0.3a-WIP. Release version will be updated to 0.4a.

Commit f0ea6b4 on 2018-03-26 by Bart Trzynadlowski

    Added ability to load save state at startup using -load-state or
    InitStateFile (in config file)

Commit d21e71d on 2018-03-21 by Ian Curtis

    work around for non working jtag code in star wars

Commit b184ca3 on 2018-03-10 by Ian Curtis

    Clamping must be applied before the spotlight is applied on step 1.0 h/w. 

Commit 609d1ba on 2018-03-08 by Ian Curtis

    Fix step 1.0 games with NET_BOARD defined (Spindizzi)

Commit d8572a9 on 2018-02-24 by Ian Curtis

    Update network code (Spindizzi)

Commit 995cf6a on 2018-01-29 by Ian Curtis

    white space

Commit 10cefb7 on 2018-01-29 by Ian Curtis

    Support both types of lightgun in lost world. 2nd type fixes some missing
    fx. Thanks to Any for pointing this out. To use the old type change
    analog_gun1 to gun1 in the xml file.

Commit 6bc4224 on 2018-01-25 by Ian Curtis

    fix a few compile warnings

Commit b210636 on 2018-01-25 by Ian Curtis

    Add the netboard stuff to the ini file. Stops crashes when they are missing.

Commit 820f27c on 2018-01-25 by Ian Curtis

    A hack to get spikeout to work (uncomment to enable) (Spindizzi)

Commit 964f08c on 2018-01-25 by Ian Curtis

    Disable the netboard if EmulateNet=0 (Spindizzi)

Commit 85a638b on 2018-01-22 by Ian Curtis

    remove extra qualifier 

Commit 04758e3 on 2018-01-22 by Ian Curtis

    Update make files and fix build for the various options (Harry Tuttle)

Commit b5bbda6 on 2018-01-22 by Ian Curtis

    fix header files

Commit 571d106 on 2018-01-21 by Ian Curtis

    Remove using namespace std from the header files ..

Commit 8266eec on 2018-01-15 by Ian Curtis

    Parse out line of sight position. Only known to be used by scud.

Commit 99b5b9b on 2018-01-15 by Ian Curtis

    make the logic a bit clearer

Commit 22a6166 on 2018-01-15 by Ian Curtis

    fix Ian's bad maths :)

Commit a1350d9 on 2018-01-14 by Ian Curtis

    Update microtexture coordinate logic (Harry Tuttle)

Commit 303c77f on 2018-01-13 by Ian Curtis

    fix illegal xml character sequence

Commit 863ea95 on 2018-01-10 by Ian Curtis

    simplify maths

Commit adb5196 on 2018-01-10 by Ian Curtis

    comment out unneeded code

Commit d708ce9 on 2018-01-10 by Ian Curtis

    top/bottom were swapped for off axis projection

Commit 6efa031 on 2018-01-08 by Ian Curtis

    Rewrite projection maths based upon previously unknown viewport values. The
    previous values used roughly worked as the normals for frustum planes.
    Perhaps they were only used for culling and not actually rendering, as
    sometimes the values don't work correctly.

Commit cf78506 on 2018-01-08 by Ian Curtis

    fix build

Commit 0b6da91 on 2018-01-07 by Ian Curtis

    add the rest of the network board code (Spindizzi)

Commit dc3048d on 2018-01-06 by Ian Curtis

    use the correct c++ deleters 

Commit 19e6aab on 2018-01-06 by Ian Curtis

    use a few macros to make debugging less painful

Commit 314279e on 2018-01-04 by Ian Curtis

    Add network board emulation. Currently not linked with the rest of the
    project.

Commit f7678d9 on 2018-01-03 by Ian Curtis

    remove using namespace from headers ..

Commit 53c9828 on 2018-01-03 by Ian Curtis

    rename to avoid clash with a #define

Commit 3ab93cf on 2018-01-02 by Ian Curtis

    add asynchronous sending functions 

Commit 58dc18c on 2017-12-29 by Ian Curtis

    remove dead variable 

Commit 78ae00f on 2017-12-29 by Ian Curtis

    Add basic networking files to supermodel. Currently these are win32 specific
    but can port to SDL or whatever later.

Commit 080acd0 on 2017-12-21 by Ian Curtis

    Fix the smallest texture lods  (Harry Tuttle)

Commit a772dd3 on 2017-12-20 by Ian Curtis

    fix array out of bounds 

Commit 4702fef on 2017-12-19 by Ian Curtis

    remove debug code

Commit 429d6cb on 2017-12-17 by Ian Curtis

    mark as static

Commit cc28d5f on 2017-12-17 by Ian Curtis

    Implement sub 8x8 tile encoding. Fixes incomplete mipmap chains which
    previously only went down to 8x8 pixels before. It wasn't known these
    textures existed before. (Harry Tuttle)

Commit cede674 on 2017-12-13 by Ian Curtis

    Fix some bad/missing textures in fighting vipers. The game is referencing
    totally illegal texture sizes. The real h/w was just handling it somehow,
    whilst we were returning null for the textures. This might produce junk
    textures for the lower mipmaps, but this can be clamped in a later update.

Commit 62d9584 on 2017-11-23 by Ian Curtis

    Shift fog maths to fragment shader to fix bug in virtua on.

Commit bddc86a on 2017-11-20 by Ian Curtis

    The vertex shader is run before clipping is done. If fogging values are
    calculated and clamped in the vertex shader it can interpolate bad values,
    since the range has been truncated. This happens if the polys overlap the
    near plane. The solution to clamp the values in the fragment shader. This
    fixes a bunch of fogging errors I long thought were transparency related
    errors in the ocean hunter.

Commit bc7f1e9 on 2017-11-07 by Ian Curtis

    Fixed shading also works with untextured polys. Fixes some shading issues in
    La machineguns. (Harry Tuttle)

Commit df44ad4 on 2017-11-05 by Ian Curtis

    Update step 1.5 luminous logic with Harry's findings. Fixes various shading
    issues.

Commit 8e13c3b on 2017-10-19 by Ian Curtis

    Finish the hash function (Harry Tuttle)

Commit 7f6df0c on 2017-10-07 by Ian Curtis

    Remove fixed shading check from specular.

Commit 4d685e0 on 2017-10-06 by Bart Trzynadlowski

    Makefile fixes: Moved Windows-specific source files out of Makefile.inc,
    using 'bin' and 'obj' when bits are not explicitly set, and changed output
    file to Supermodel rather than Supermodel.exe, which works on all platforms.

Commit eb798ed on 2017-10-05 by Ian Curtis

    Star wars is the only game to pass unsigned fixed shaded values (per vertex
    brightness) to the renderer. Originally we thought that the specular flag
    would turn on unsigned values since it's the only game to set specular with
    these polys, but this logic turned out to be incorrect. The JTAG interface
    seems to config the GPU to turn on this functionality. (Harry Tuttle)

Commit 7c84bc6 on 2017-10-05 by Ian Curtis

    update visual studio project

Commit 43cf6b3 on 2017-10-05 by Ian Curtis

    update spotlight code (Harry Tuttle)

Commit cee468b on 2017-10-05 by Bart Trzynadlowski

    Harry's patch: add dirtdvlsj and change "Star Wars Trilogy" to "Star Wars
    Trilogy Arcade"

Commit 2397b82 on 2017-10-05 by Bart Trzynadlowski

    New, saner Makefiles. They still require some customization but share a core
    Makefile include with auto-generation of dependencies. Multi-threaded
    building (e.g., make -j6) is supported now.

Commit 9bac783 on 2017-09-30 by Ian Curtis

    add missing header

Commit 3ca3708 on 2017-09-27 by Bart Trzynadlowski

    Get rid of unused variable warning

Commit 0afd902 on 2017-09-26 by Ian Curtis

    Fix the sky in one of the levels in virtua on. The problem was caused by the
    fact we didn't split the mesh by the fog % burn through, so all the polys
    had the same value. We could use a vertex attrib, but % burn through fog is
    so rarely used and when it is, normally every poly in the mesh has the same
    value.

Commit 1eac755 on 2017-09-25 by Bart Trzynadlowski

    Fix tiny memory leak

Commit 18f460d on 2017-09-24 by Ian Curtis

    

Commit da1342b on 2017-09-24 by Ian Curtis

    update project files

Commit 0376f74 on 2017-09-24 by Ian Curtis

    work around for visual studio bug 

Commit adecec7 on 2017-09-24 by Ian Curtis

    fix missing header

Commit 7590f50 on 2017-09-24 by Ian Curtis

    cannot specify explicit initializer for arrays - fix for visual studio

Commit 55bb02d on 2017-09-24 by Bart Trzynadlowski

    - New work-in-progress frame timing code (disabled by default, compile with
    NEW_FRAME_TIMING defined to activate it)
    - New JTAG emulation, moved into its own class, CJTAG
    - Removed game-specific sun clamp hacks from CNew3D (JTAG and Real3D
    emulation will call the appropriate method to configure this at run-time)
    - Removed JTAG from Real3D save state data and reused some of that space for
    new state variables having to do with the internal JTAG-based config as well
    as new frame timing state variables

Commit c4f8471 on 2017-09-24 by Bart Trzynadlowski

    Added functions to safely read and write bools

Commit 1db2f6b on 2017-09-24 by Bart Trzynadlowski

    Added functions to extract bits as integer values and updated comment about
    bit vector layout

Commit 3041ca1 on 2017-09-23 by Bart Trzynadlowski

    Added SetOnes() and SetZeros()

Commit c3eef70 on 2017-09-23 by Bart Trzynadlowski

    Safe guard against shifting empty register.

Commit 98b0354 on 2017-09-23 by Bart Trzynadlowski

    Hex formatting support extended to 64 bits.

Commit 61a0517 on 2017-09-23 by Bart Trzynadlowski

    Added ShiftOutLeft() and ShiftOutRight()

Commit 7a2e6b3 on 2017-09-23 by Bart Trzynadlowski

    Renamed shift in/out functions to add/remove to be less ambiguous about
    their functionality

Commit 8663169 on 2017-09-19 by Bart Trzynadlowski

    Added ToHexString() and made ostream serialization use this

Commit e12a945 on 2017-09-18 by Bart Trzynadlowski

    Fix for VS2015 and newer by MetalliC

Commit f2030ac on 2017-09-07 by Bart Trzynadlowski

    Added BitRegister class (for use in upcoming JTAG refactor): an inefficient
    but flexible container for manipulating a dynamic bit stream.

Commit a3e2fb7 on 2017-09-06 by Ian Curtis

    Calculating the length of the vertex results in quite broken values for
    fogging when the vertices are traversing the view frustum. The hardware also
    wouldn't have been doing a square root per vertex for fogging. Simply using
    the z values is enough. Todo check clamping of values for the near/far
    planes.

Commit 2786d95 on 2017-09-02 by Ian Curtis

    add interface for sun clamp

Commit 266c911 on 2017-08-30 by Ian Curtis

    Optimise vertex data to cut down on unnecessary copying 

Commit 165926a on 2017-08-29 by Ian Curtis

    Modern hardware does backface culling in window space by calculating the
    face normal for the polygon, then doing a dot product against the view
    vector. The real3d pro-1000 on the other hand passes a pre-calculated face
    normal for each polygon which is used for culling. We were using this face
    normal to rewind the polygons so that regular backface culling would work.
    This worked 99.9% of the time. However this was failing on some models in
    Virtua Striker. The reason was because the pre-calculated face normals being
    passed were actually completely different to the actual face normals for the
    poly (not just inverted like you would expect). This broke our code. The
    solution was to emulate face culling directly in the vertex shader using the
    pre-calculated face normals directly. Only minimally tested this but
    hopefully there are no obvious regressions.

Commit f04a285 on 2017-08-29 by Bart Trzynadlowski

    DMA and register space access for Real3D have been unified and
    CReal3D::ReadRegister() returns little endian data now that the PowerPC
    handler byte-swaps back, to be consistent with how DMA registers work.

Commit 50d1f95 on 2017-08-25 by Ian Curtis

    Changed fixed shaded logic for step 1.5 hw based upon Harry's findings

Commit b21ed49 on 2017-08-21 by Ian Curtis

    cosmetic

Commit cd59787 on 2017-08-19 by Ian Curtis

    Harry made some important discoveries with regards to fixed shading on the
    model 3 (per vertex poly brightness values). Firstly values are allowed to
    be negative, and they are used as a drop in replacement in the standard
    lighting equation for the normal dot light vector. This quite radically
    changes the brightness in LA Machine guns, but now correctly matches the
    arcade.

Commit 94e6f9a on 2017-08-19 by Bart Trzynadlowski

    Patched Sega Rally 2 DX to boot up. Enjoy Sega Rall 2 without MPEG music at
    50% speed ;)

Commit 507990a on 2017-08-19 by Bart Trzynadlowski

    Fixed up memory region dumping code to use the new Util::FlipEndian
    functions

Commit 21f813a on 2017-08-16 by Bart Trzynadlowski

    Fixed a bug in MCRFS instruction related to the VXSOFT, VXSQRT, and VXCVI
    bits. Thanks to Spindizzi.

Commit 0efd4da on 2017-08-15 by Ian Curtis

    Finish fixed shading for 2.0 hardware. I'm pretty sure it actually works
    identically to step 1.5 hardware. The oddball is LA machine guns where the
    viewport ambient doesn't seem to effect the brightness. But the ambient
    works differently in this game because it uses the unclamped light model.
    Still need to investigate if the diffuse factor effects fixed shading.

Commit b1cc9a6 on 2017-08-15 by Ian Curtis

    With fixed shading, when lighting is disabled, fixed shading appears to be a
    flat shaded version based upon the poly colour, instead of per vertex
    attributes.

Commit 2422caa on 2017-08-15 by Ian Curtis

    Pad struct and align on 4 byte boundary.

Commit eac76f0 on 2017-08-14 by Ian Curtis

    simplify

Commit 81188be on 2017-08-14 by Ian Curtis

    fixed shading doesn't effect the alpha channel

Commit c12e4c3 on 2017-08-14 by Ian Curtis

    Fixed shading (per vertex poly colours) on step 1.5 hardware have the
    viewport ambient value added to them. This fixes various shading on scud. To
    do this had to switch the maths to the vertex shader.

Commit 8f62271 on 2017-08-11 by Ian Curtis

    Fix attribute locations. glBindAttribLocation must be called linking

Commit a315627 on 2017-08-11 by Bart Trzynadlowski

    Added support for loading shaders from files for new engine and tile
    renderer. Changed help text to reflect new engine being default.

Commit 7dc1eac on 2017-08-10 by Ian Curtis

    make the new3d engine default :] Legacy engine can still be selected using
    -legacy3d in the command line

Commit d7403af on 2017-08-10 by Ian Curtis

    use generic vertex attributes

Commit 57d0a51 on 2017-08-09 by Ian Curtis

    Technically matrix casts are only allowed in glsl 1.2 and later

Commit 063b7d3 on 2017-08-09 by Ian Curtis

    Remove debug code

Commit 5709ee2 on 2017-08-09 by Ian Curtis

    Sometime ago I managed to work out that specular on the model3 is not real
    specular, and really is just an extension of diffuse lighting. But attempts
    were derailed by corner cases and the fact we were not handling the normals
    correctly. Anyway Harry managed to successfully come up with an algorithm,
    and coefficients that give an almost perfect match to specular on the
    model3, based soley on observations from video footage! He also worked out
    that the lighting on hardware 1.5 onwards appears to be unclamped (ie
    greater than 1). This quite radically changes the brightness of some of the
    games, but much better matches the original hardware.

Commit 1d33887 on 2017-08-03 by Ian Curtis

    cosmetic

Commit 3fc2815 on 2017-08-03 by Ian Curtis

    Add the unclamped light model we know exists to the shaders.

Commit b1bd877 on 2017-08-01 by Ian Curtis

    Ski champ for some reason is passing denormalised numbers for the
    modelscale. This is causing a NaN in our shader killing the lighting for
    these models, so we simply skip these, since they are essentially zero
    anyway. Thanks to Harry for finding this bug :)

Commit 9aa3f13 on 2017-07-29 by Ian Curtis

    We were force normalising the result of matrix * normal, which looked
    correct in most cases. But this didn't preserve the scaling of the matrix,
    or the scaling of the model normals which resulted in many over bright
    areas. On it's own this generally worked, but games like Star Wars looked
    quite broken. Harry correctly figured out if you scale these normals by the
    scaling value that is sometimes present in the culling nodes the lighting
    looks correct. Still more work to do to correctly figure out the model3's
    lighting model.

Commit a9b49c1 on 2017-07-22 by Ian Curtis

    Implement unclamped light model based upon Harry's findings. Unknown how
    this is turned on/off.

Commit 2335f31 on 2017-07-08 by Ian Curtis

    Use c++11 raw string literals

Commit 3efb055 on 2017-06-29 by Ian Curtis

    Clamp the ambient light to a max of 0.75. LA Machine guns seems to use an
    ambient value of 1, which is full bright for everything. But for some reason
    the hardware seems to treat this as ~ 0.75. This fixes various missing
    shading in the game. This doesn't seem to have any negative effects on other
    games I have tested. Clamping in the shader as we might need the full range
    of values for fixed shading.

Commit 9e9a549 on 2017-06-25 by Ian Curtis

    remove WIP code

Commit ec3872d on 2017-06-25 by Ian Curtis

    convert sun angle coordinate system outside of the shader

Commit 9778110 on 2017-06-21 by Ian Curtis

    A second go at fixing the fixed lighting in star wars .. Fixed shading is
    per vertex poly colours. For some reason in star wars they are treated as
    unsigned values instead of signed, like in every other game. These polys are
    all marked with specular enabled, where as in the rest of the games they are
    missing this flag. That's the only difference I can find.

Commit 1532168 on 2017-06-20 by Ian Curtis

    fix double offset

Commit b00315b on 2017-06-20 by Ian Curtis

    Revert previous fixed shading changes. Implement new logic for fixed shading
    on step 1.5 hardware based upon Harry's findings.

Commit bb3c538 on 2017-06-20 by Ian Curtis

    remove debug code

Commit 260daa1 on 2017-06-17 by Ian Curtis

    Fix the fix shading in star wars. Fixed shading still needs some work for
    mag truck and dirt devils to match the arcade.

Commit 73a7d27 on 2017-06-07 by Ian Curtis

    cosmetic

Commit 1193069 on 2017-06-07 by Ian Curtis

    Tweak scroll fog logic again .. should fix missing 2d fogging in Spikeout.

Commit 45741b2 on 2017-05-06 by Ian Curtis

    rewrite scroll fog logic

Commit 80004b0 on 2017-04-25 by Ian Curtis

    Remove debug code

Commit 7694e4e on 2017-04-25 by Ian Curtis

    minor culling node optimisations

Commit ff936c5 on 2017-04-20 by Ian Curtis

    don't need cast, conversion is implicit

Commit 912bb0e on 2017-04-20 by Ian Curtis

    better state handling

Commit 9b226ed on 2017-04-17 by Ian Curtis

    parse out some culling node values

Commit 3c78571 on 2017-04-15 by Ian Curtis

    There are 2 bits in the polyheader labelled discard. When both are set ie in
    sega rally, the polygons are totally discarded. Scud seems to set bit 1
    which seems halve the opacity.

Commit 21836c0 on 2017-04-14 by Ian Curtis

    The real3d has a flag that allows the hw to skip polygons so that they
    aren't rendered. Sega rally2 uses this for the dust effects. Triangles or
    quads that have this flag are junk, and look like random corruption if you
    try to draw them. We assumed polys that shared vertices with these bad
    polygons were also bad. It turns out the last few shared vertices in these
    polys were in fact valid, and thus the entire quad/triangle strip should be
    drawn and not discarded. 

Commit a2fcbf8 on 2017-04-14 by Bart Trzynadlowski

    Linux Makefile fixed by Sergio Benjamin

Commit fc50bde on 2017-04-14 by Ian Curtis

    light parameters must be clamped (HarryTuttle)

Commit 391aafd on 2017-04-12 by Ian Curtis

    Virtua fighter has a few massive polys that shouldn't be visible. It looks
    like these polys have a bad matrix, but the node distances are all zero.
    This means the bounding box that encapsulates them is of size zero. Skipping
    over these entirely seems to fix the issue.

Commit 4a70d5e on 2017-04-12 by Bart Trzynadlowski

    Updated Util::Config::Node unit tests with multiple leaf node test

Commit e462fbe on 2017-04-11 by Ian Curtis

    remove unused variable

Commit 7681779 on 2017-04-11 by Bart Trzynadlowski

    Extended -gfx-state to produce culling node analysis

Commit 9dd2b24 on 2017-04-11 by Bart Trzynadlowski

    Fixed behavior of Util::Config::Node when adding nested nodes (only leaf
    level can be duplicated; e.g., foo/bar and foo/bar will create one foo with
    two bar children, rather than two foo each with one bar).

Commit 901b8c3 on 2017-04-09 by Bart Trzynadlowski

    Spindizzi's SCSP DSP fix for the VF3TB cave stage

Commit 9989eae on 2017-04-09 by Bart Trzynadlowski

    Hooked up the command line help options (thanks, Spindizzi!) and updated
    copyright info

Commit 53fcb57 on 2017-04-09 by Bart Trzynadlowski

    Removed unnecessary patches from lemans24

Commit 06cb570 on 2017-04-08 by Ian Curtis

    add missing include

Commit a649a0d on 2017-04-08 by Bart Trzynadlowski

    Removed patches from source code and moved them into game XML file. Created
    ROMSet.cpp. Print a more descriptive error when game XML file fails to load.

Commit 86ffda6 on 2017-04-08 by Bart Trzynadlowski

    Updated help text to include -game-xml-file option

Commit c8053e3 on 2017-04-08 by Bart Trzynadlowski

    Moved ROM patches into Games.xml and added patches to swtrilgy to make test
    emu work and to eca/ecau to pass the network check

Commit 788a480 on 2017-04-07 by Ian Curtis

    add missing include

Commit 006260b on 2017-04-07 by Bart Trzynadlowski

    Harry Tuttle's fixes to fishing controls: invert fishing rod y and fishing
    stick y axis values so setting a negative axis is not necessary. Also
    changed reel to an analog input (similar to a pedal), whose neutral position
    is 0.

Commit 4ca7d5e on 2017-04-07 by Bart Trzynadlowski

    Harry Tuttle's change to Harley gear shift to use a single bit instead of
    two (no detectable difference in behavior, though)

Commit 205282a on 2017-04-06 by Bart Trzynadlowski

    Removed unused file: Game.cpp 

Commit 25f2272 on 2017-04-05 by Bart Trzynadlowski

    Fixed lamachin title -- thanks Jiterdomer!

Commit 101e432 on 2017-04-05 by Ian Curtis

    cosmetic 

Commit edfb521 on 2017-04-05 by Ian Curtis

    Daytona seems to use this completely undocumented feature of the real3d
    pro-1000, the ability to invert texture colours. This patch fixes the
    colours on the cars, and the signs which should flash by alternating their
    colours.

Commit cc32e0b on 2017-04-05 by Bart Trzynadlowski

    Removed last vestiges of CINIFile from input system and removed Games.cpp,
    INIFile.cpp, and ROMLoad.cpp

Commit df283d9 on 2017-04-05 by Bart Trzynadlowski

    Using Game::INPUT_* instead of obsolete GAME_INPUT_*

Commit c013eab on 2017-04-05 by Bart Trzynadlowski

    Changes by Harry Tuttle: game list more accurately reflects MAME's, added a
    root level XML "games" node, print ROM version along with title

Commit f6707ec on 2017-04-04 by Bart Trzynadlowski

    Fixed graphics state analyzer 

Commit c4bf565 on 2017-04-04 by Bart Trzynadlowski

    Added cstring

Commit ad6b4b0 on 2017-04-04 by Bart Trzynadlowski

    Added remaining games largely thanks to the efforts of the mysterious '阳阳'

Commit 4ebcf58 on 2017-04-03 by Ian Curtis

    more whitespace ..

Commit a141635 on 2017-04-03 by Ian Curtis

    white space ..

Commit c617a45 on 2017-04-03 by Ian Curtis

    Cache dereferenced values. Cuts loading time from 15 to 5 seconds in debug
    mode :)

Commit 5aafbbf on 2017-04-03 by Bart Trzynadlowski

    Included cstring for memcpy

Commit 2aa9dc7 on 2017-04-03 by Bart Trzynadlowski

    Rewrote large parts of ROM loading code to more gracefully handle merged
    sets where overwritten parent ROMs are not present. Better handling of
    multiple zip files. Optimized ROM loading when stride and chunk size are
    equal and got rid of safety checks that I don't believe are necessary any
    longer.

Commit f0e00c5 on 2017-04-02 by Ian Curtis

    Rewrite the spot light code, and implement the missing fog logic.
    (HarryTuttle)

Commit 07660be on 2017-04-01 by Bart Trzynadlowski

    Fixed missing file detection (set_difference requires inputs sorted by
    desired key) and improved error reporting

Commit e04c368 on 2017-03-31 by Bart Trzynadlowski

    Made scuda a child set of scud

Commit 68ebf35 on 2017-03-31 by Bart Trzynadlowski

    Removed TODO comment

Commit b0957e6 on 2017-03-31 by Bart Trzynadlowski

    Smarter game selection when a zip file contains multiple games (prefer child
    when parent is also present in zip)

Commit e40d268 on 2017-03-31 by Bart Trzynadlowski

    Warn about parent ROM sets having parent defined (not allowed)

Commit 47ce132 on 2017-03-31 by Bart Trzynadlowski

    Print missing files from games

Commit fd0f7ec on 2017-03-30 by Bart Trzynadlowski

    Support loading of parent ROM sets (e.g., scudp will load scud)

Commit cd1a551 on 2017-03-28 by Ian Curtis

    compile fixes

Commit 1a76684 on 2017-03-28 by Ian Curtis

    MSYS2 compile fix (HarryTuttle)

Commit 69eac95 on 2017-03-28 by Ian Curtis

    Replace glsl 3 function. Apparently this doesn't work on OS X.

Commit 70259ee on 2017-03-28 by Ian Curtis

    fix compile warnings

Commit 318f4cf on 2017-03-28 by Bart Trzynadlowski

    Sort games found in zip file by descending order of matching files so that
    game with the *most* matching files is picked

Commit aa4331c on 2017-03-28 by Bart Trzynadlowski

    Added: bass, daytona2, dirtdvls, harley, lemans24, scuda, skichamp,
    spikeofe, spikeout, von2, vs2, vs298, vs2v991 Fixed: dayto2pe title

Commit 40486fc on 2017-03-28 by Bart Trzynadlowski

    Fixed scud mpeg roms


Commit ddbf08a on 2017-03-27 by Ian Curtis

    don't ask

Commit f320af2 on 2017-03-27 by Ian Curtis

    fix project files

Commit 3fefff3 on 2017-03-27 by Ian Curtis

    fix multiple default constructors 

Commit 5c15dac on 2017-03-27 by Ian Curtis

    add missing include

Commit f34e25d on 2017-03-27 by Bart Trzynadlowski

    Massive internal change: removed CConfig object and replaced it with a
    hierarchical config object system (Util::Config::Node). Games are now
    defined in an XML file. Hopefully I didn't break too many things :/

Commit 26a4417 on 2017-03-27 by Bart Trzynadlowski

    Cosmetic: renamed step to stepping

Commit 10befe4 on 2017-03-27 by Bart Trzynadlowski

    No need to enclose values in quotes when writing INI file. Print that file
    was saved to console and log.

Commit 35d9842 on 2017-03-27 by Bart Trzynadlowski

    Added a TODO list

Commit 2e6a58b on 2017-03-27 by Bart Trzynadlowski

    Added srally2

Commit c4a6382 on 2017-03-27 by Bart Trzynadlowski

    Encapsulated zip file handling and added a "parent" member to the Game
    struct

Commit ba3a0b5 on 2017-03-26 by Bart Trzynadlowski

    Checking in GameLoader.cpp, GameLoader.h, and ROMSet.h (not currently used
    in build) because I need to make a substantial modification in my local tree

Commit 9e19b46 on 2017-03-25 by Bart Trzynadlowski

    Fixed dayto2pe sample ROMs

Commit ff624af on 2017-03-25 by Ian Curtis

    The mipmap data in some games seems to being sent behind the base textures,
    and when the base textures are bound and created the mipmap chain is not yet
    valid in memory. To fix this, we invalidate the base textures, when the
    mipmap chain is updated.

Commit 06c1cdc on 2017-03-24 by Ian Curtis

    Just use 1 hash map for the textures, instead of 1 for each format. Should
    greatly speed up texture invalidation. TODO invalidate texture mipmaps ..
    the h/w actually seems to stream the mipmap data separate from the base
    textures in some cases.

Commit 98d56ff on 2017-03-24 by Ian Curtis

    cosmetic 

Commit 82ac030 on 2017-03-24 by Ian Curtis

    calculate the correct number of mipmap textures

Commit 17f8784 on 2017-03-24 by Ian Curtis

    Originally to do mipmapping, I just used the base texture and then let the
    h/w automatically generate the mipmap chain. I wasn't sure the real3d api
    even exposed to the programmer the ability to directly set the mipmap data.
    Anyway, as pointed out by HarryTuttle scud actually uses darker mipmaps for
    an effect on the water. This patch uses the uses the actual mipmap data from
    the real3d memory. Only tested scud, so hopefully I didn't cause any
    cataclysmic errors :)

Commit df3805e on 2017-03-24 by Bart Trzynadlowski

    Added fvipers2

Commit 17f7c2d on 2017-03-23 by Ian Curtis

    use 32bit types

Commit 3c2cf17 on 2017-03-23 by Bart Trzynadlowski

    Added more games to Games.xml: dayto2pe, eca, lamachin, lostwsga, magtruck,
    oceanhun, vf3. Fixed swtrilgy (MPEG music).

Commit 61159c0 on 2017-03-23 by Bart Trzynadlowski

    Added 8- and 16-bit tilegen VRAM access handlers to fix missing columns in
    Star Wars Trilogy tilemaps. Thanks to Spindizzi for initial confirmation
    that this was the problem.

Commit 9fb6acc on 2017-03-22 by Ian Curtis

    fix scissor code for native resolution (thanks HarryTuttle)

Commit 519cce3 on 2017-03-19 by Ian Curtis

    remove debug code

Commit 9780f33 on 2017-03-19 by Ian Curtis

    The model3 has this weird issue where the 2d and 3d planes are misaligned by
    2 pixels. Originally I made a quick hack that shifted subviewports 2 pixels.
    But apparently that wasn't enough, as the issue effects everything. Thanks
    to HarryTuttle for pointing this out, and making a patch :)

Commit b19ceb0 on 2017-03-13 by Ian Curtis

    Texture coordinates seem to be scaled relative to the fixed texture size of
    256. Thanks to HarryTurtle for figuring this crazy logic out. The road in
    scud matches the arcade now almost perfectly.

Commit dcd018a on 2017-03-12 by Ian Curtis

    Change microtexture scaling values. Seem to much better match expected
    output. (thanks HarryTuttle)

Commit 732e3ef on 2017-03-11 by Ian Curtis

    More fixes to microtexture coordinates. They go vertically up and down the
    texture sheet, and not horizontally as I had guessed. (Thanks HarryTurtle)

Commit 6636776 on 2017-03-11 by Ian Curtis

    Fix microtexture texture coordinates (thanks HarryTuttle)
        

Commit 503c88d on 2017-03-07 by Ian Curtis

    fix the bad projection matrix in dirt devils and lost world

Commit f173c3c on 2017-02-21 by Ian Curtis

    The hardware actually seems to treat the quad primitive as a triangle strip.
    The winding order can actually change between the first and second
    'triangle' inside the quad, leading our code to incorrectly backface cull
    the 2nd triangle. So we must re-calculate the winding order for the 2nd
    triangle. This fixes a massive missing poly in the road in la machine guns,
    probably other games too. The hardware is unaffected by the winding order,
    as it doesn't calculate the poly normal from the vertices directly, instead
    it uses the provided normal in the poly header.

Commit 5b9741b on 2017-02-20 by Ian Curtis

    Support high priority polygons. They are treated as a separate overlay plane
    by the h/w. Fixes some missing fx in harley.

Commit 2ae2010 on 2017-02-18 by Ian Curtis

    Add high priority poly function. Used by harley.

Commit 4a11619 on 2017-02-16 by Ian Curtis

    When the translator map?? bit is enabled the colour format seems to be
    4-5bit. Max observed value is 16. Scaling these values up seems to produce
    the correct colours in harley and dirt devils.

Commit 7407d03 on 2017-02-08 by Ian Curtis

    Put a check in there, in case near values come out really close to zero.
    Fixes a slight corner case in lemans24. A check probably should have been
    there originally :)

Commit 24cbeed on 2017-02-07 by Ian Curtis

    Dynamically calculate near/far frustum planes for each viewport priority
    layer by clipping the overlapping meshes with the 4 frustum planes. A bit
    brute forcy, but the results are great.

Commit 144125a on 2017-02-05 by Ian Curtis

    mark the nodes without valid distances set

Commit c82dc55 on 2016-12-22 by Ian Curtis

    cosmetics 

Commit 72dd0db on 2016-12-22 by Ian Curtis

    only need to cull against 4 clipping planes

Commit 20a9514 on 2016-12-15 by Ian Curtis

    For scroll fog set the viewport to the entire area, allows fog to work in
    widescreen mode

Commit ca70467 on 2016-12-15 by Ian Curtis

    properly cache un-normalised texture coordinates, fixes a texturing
    regression in Daytona 

Commit f355e2d on 2016-12-15 by Ian Curtis

    fix microtexture colours

Commit 89e75a9 on 2016-12-14 by Ian Curtis

    Only check scroll fog value from the first viewport from each priority
    layer. Seems to fix problems with background fogging.

Commit 707e945 on 2016-12-09 by Ian Curtis

    delete unused files

Commit 2c23268 on 2016-12-09 by Ian Curtis

    Support microtexture relative scale. The exact scaling values are not known,
    but a 2x difference between each level seems highly likely. We know from
    visual inspection from scud that the first level is exactly 4. The SDK has a
    function that accepts a float input for scale and does this. lod = (1/scale)
    + 0.5. If lod>3 lod = 3. So this is our best guess.

Commit da8b63b on 2016-12-06 by Ian Curtis

    The model3 shares vertices between polygons. But it also seems to do this
    for polygons with different textures. This means the texture coordinate for
    potentially different sizes textures is shared. Currently this was broken
    leading to some corrupted looking textures. The fix is to simply cache the
    old texture coordinates, and recalculate the normalised coordinates in the
    new texture coordinate space.

Commit 10a5d2c on 2016-12-06 by Ian Curtis

    refactor

Commit eece2d7 on 2016-12-04 by Ian Curtis

    tweak scroll fog logic

Commit adbebb0 on 2016-12-04 by Ian Curtis

    clean up

Commit a310e8e on 2016-11-28 by Ian Curtis

    minor optimisations 

Commit abc8c24 on 2016-11-27 by Ian Curtis

    Only check the geometry for culling with valid node distances. We will need
    this later.

Commit fe60d11 on 2016-11-26 by Ian Curtis

    Revert changes .. if we clip against the real3d clipping planes, our wide
    screen hack won't work anymore.

Commit 665fa90 on 2016-11-26 by Ian Curtis

    Get clipping planes directly from the viewport data. Also no need to clip
    against a far plane

Commit 54c4057 on 2016-11-22 by Ian Curtis

    Need to walk back in memory and check the previous texture tiles to see if
    they overlap with the current textures being invalidated. If they overlap
    they must be cleared too. Algorithm is a bit brute forcy and more expensive
    than I'd like but it fixes the bad textures in Spikeout.

Commit 84ae8f0 on 2016-11-21 by Ian Curtis

    Skip processing the mipmap section of memory. The real3d doesn't create
    mipmaps down to 1x1 that is needed to do mipmapping properly on modern
    hardware anyway.

Commit 2ddc8a3 on 2016-11-20 by Ian Curtis

    Clear queued texture uploads after processing them. Saves re-uploading the
    same textures whilst paused

Commit 10dd530 on 2016-11-15 by Ian Curtis

    More accurate 2D colours. White should map to 255 not 248

Commit 483ba2f on 2016-11-11 by Ian Curtis

    Fix the shading in scud. When lighting is disabled the fixed shading
    intensities appear to be offset values added to the colour value.

Commit 0a7c0d0 on 2016-10-21 by Ian Curtis

    change throttle logic for smoother playback, we could get closer to 60fps if
    SDL timers weren't so lame

Commit 2f49012 on 2016-10-09 by Ian Curtis

    change scroll fog logic to parse the lower priority viewports first

Commit f0a15fe on 2016-10-09 by Ian Curtis

    implement scroll fog, fixes a tonne of missing fx

Commit ef11990 on 2016-10-08 by Ian Curtis

    Parse out scrollfog value used for blending over the 2d background. 

Commit 33e795d on 2016-10-08 by Ian Curtis

    Fix scenes with missing fog in ocean hunter. The game was passing negative
    fog densities. The sign seems to be ignored by the model 3.

Commit 9eeb81a on 2016-10-08 by Ian Curtis

    same fix for legacy engine

Commit 33cf6c8 on 2016-10-08 by Ian Curtis

    our scratch mem must match correct max texture size

Commit c472d90 on 2016-10-06 by Ian Curtis

    Fix the sky in advanced stage in Dirt Devils. It uses a massive (for 1998
    :p) 1024x256 texture.

Commit c3581c9 on 2016-10-06 by Ian Curtis

    Support multiple microTextures, fixes crazy texturing in sega ski champ. I'm
    assuming microtextures are always located on the other memory bank to the
    base texture. This logic seems to work for all our games anyway.

Commit e21ea93 on 2016-09-07 by Bart Trzynadlowski

    Added RGBA4 format to BMP file writer

Commit 3df3610 on 2016-08-25 by Bart Trzynadlowski

    Made parsing of hex values more robust

Commit 23435ae on 2016-08-25 by Bart Trzynadlowski

    Initial check in of Games.xml (only two games defined so far)

Commit 458684e on 2016-08-24 by Bart Trzynadlowski

    Fixed new 3D engine's fragment shader to work on some older GPUs by reducing
    the number of discard statements to one

Commit 4b184b0 on 2016-08-24 by Bart Trzynadlowski

    Updated game loader to make use of latest changes to config tree: generic
    value types and eliminated pointers

Commit c2f80ce on 2016-08-24 by Bart Trzynadlowski

    Support for hex numbers in string to integer conversions (with '0x' prefix)
    and special keywords in string to bool conversions
    (true/false/on/off/yes/no)

Commit f08ad15 on 2016-08-24 by Bart Trzynadlowski

    Added Util::Stricmp() and small tweak to serialization of Util::Format

Commit 2efe18b on 2016-08-21 by Bart Trzynadlowski

    Added support for multiple value types in config nodes. Removed Ptr_t and
    ConstPtr_t from public interface, moved config tree builders into their own
    file, and made them pass by reference and value. Exceptions can be thrown
    now on lookup failures. Removed s_empty_node -- failed lookups in []
    operator create a permanent "hidden" child. Updated comment in
    NewConfig.cpp.

Commit 01a6ad1 on 2016-08-19 by Bart Trzynadlowski

    Added copy constructor and assignment operator using C++11 move semantics

Commit a981646 on 2016-08-18 by Bart Trzynadlowski

    Defined inputs

Commit fbeeb3c on 2016-08-18 by Bart Trzynadlowski

    Support for deep copies via assignment operator. This will pave the way for
    elimination of Ptr_t and ConstPtr_t.

Commit 5652aab on 2016-08-18 by Bart Trzynadlowski

    Added Write() method to write buffer to an ostream, and added an ostream
    operator overload

Commit 1dc360a on 2016-08-11 by Bart Trzynadlowski

    Changes to game loader interface, added 32-bit word swapping utility
    function

Commit 41105c1 on 2016-08-10 by Bart Trzynadlowski

    Added Logger.o to OSX and UNIX Makefiles. Should fix compilation on these
    platforms.

Commit 821243b on 2016-08-10 by Bart Trzynadlowski

    Removing MSVC Makefile 

Commit a2423a8 on 2016-08-10 by Bart Trzynadlowski

    Initial check-in of game loader (not operational yet)

Commit 037bfc6 on 2016-08-10 by Bart Trzynadlowski

    Byte swapping utility function

Commit 718c237 on 2016-08-10 by Bart Trzynadlowski

    Added ValueAsBool(), ValueAsBoolWithDefault(), and
    ValueAsUnsignedWithDefault()

Commit f4311a0 on 2016-07-10 by Ian Curtis

    update visual studio project

Commit 52d6eac on 2016-07-10 by Bart Trzynadlowski

    Added Util::Config::Node::ValueAsUnsigned() with support for base 10 and 16
    ('0x' prefix)

Commit 331d704 on 2016-07-10 by Bart Trzynadlowski

    Added Util::ToLower()

Commit a84aacf on 2016-07-09 by Bart Trzynadlowski

    Config::Node::Add() now supports nested keys (e.g., Add("foo/bar/baz", "0")
    will create three nested config nodes)

Commit 24c3411 on 2016-07-09 by Bart Trzynadlowski

    Changed Config::Node::Create() to Add() and added a Set() method (for INI
    semantics). Turned Test_Config.cpp into a proper unit test reporting
    pass/fail for each test.

Commit 2e069af on 2016-07-07 by Bart Trzynadlowski

    Adding tinyxml2 lib

Commit 724dc93 on 2016-07-07 by Bart Trzynadlowski

    XML parsing support for new config system

Commit bcc663d on 2016-07-05 by Bart Trzynadlowski

    Moved logging functions into OSD/Logger.cpp (eventually logging system will
    be reworked entirely) and added new config tree structure (not yet used). 

Commit 0650ffc on 2016-07-04 by Bart Trzynadlowski

    Updated Win32 gcc, UNIX, and OSX Makefiles

Commit 0ab4866 on 2016-07-04 by Bart Trzynadlowski

    Updated Fragment_NoSpotlight.glsl (fragment shader for legacy engine without
    spot light)

Commit 732bff2 on 2016-06-29 by Ian Curtis

    disable spot light if the extent is zero

Commit 1c39956 on 2016-06-29 by Bart Trzynadlowski

    Saving progress on scudp disassembly

Commit efa7776 on 2016-06-27 by Ian Curtis

    parse some unimplemented spot light values from viewport

Commit 012253e on 2016-06-27 by Bart Trzynadlowski

    Added disassembly of Scud Race Plus

Commit de04335 on 2016-06-24 by Bart Trzynadlowski

    Marked offset of code in CROM for scudp

Commit d058a29 on 2016-06-19 by Ian Curtis

    stencil translucent non textured polys

Commit 85c4d85 on 2016-06-19 by Ian Curtis

    refactor

Commit c2b1db1 on 2016-06-16 by Ian Curtis

    Cull geometry based on the culling distances in the nodes. Should give some
    sort of speedup on lower end pcs where the CPU is predominantly the bottle
    neck.

Commit f31f06b on 2016-06-13 by Bart Trzynadlowski

    Fix compilation when SUPERMODEL_DEBUGGER enabled

Commit d799c63 on 2016-06-10 by Ian Curtis

    cache bound texture

Commit 48a79a8 on 2016-06-10 by Ian Curtis

    cleaning

Commit b0911af on 2016-06-07 by Bart Trzynadlowski

    Added drive board and lamp raw outputs (submitted by SailorSat)

Commit 8440bc7 on 2016-06-03 by Ian Curtis

    support negative exponent 

Commit 5081ee9 on 2016-06-03 by Ian Curtis

    Add code to convert the 16bit float format used by the hardware back into
    float 754 format. 16bit floats are used for the culling radiuses in the
    nodes.

Commit 6dc5845 on 2016-06-02 by Bart Trzynadlowski

    Added -vsync and -no-vsync option descriptions to command line help

Commit 9bc1fbf on 2016-06-02 by Ian Curtis

    remove dead code

Commit 968652b on 2016-06-02 by Bart Trzynadlowski

    Added a Split() function to Util::Format.

Commit d1330d3 on 2016-06-02 by Bart Trzynadlowski

    XInput fix: search for all three possible XInput DLLs (xinput1_4.dll,
    xinput1_3.dll, and xinput9_1_0.dll). I have not tested 32-bit builds.

Commit ff8e9d6 on 2016-05-30 by Ian Curtis

    Check for overflow. Fixes some fade effects

Commit 9a49fef on 2016-05-30 by Bart Trzynadlowski

    Legacy engine: fixed texture offsets (VON2 was broken by last commit).
    Texture coordinates are wrapped within the bank first and then the bank can
    be swapped. This also allows the full range of texture coordinate bits to be
    used as specified in the Pro-1000 SDK.

Commit 36074d9 on 2016-05-30 by Ian Curtis

    Y texture offsets are just 5 bits. Texture coordinate wrap around happens in
    the same texture sheet, not into the next one. Fixes various texturing bugs
    in sega rally.

Commit d1f4967 on 2016-05-29 by Bart Trzynadlowski

    Legacy engine: Culling node texture Y offset is only 5 bits (fixes Sega
    Rally 2 initials decal)

Commit 2ee0d90 on 2016-05-28 by Bart Trzynadlowski

    New engine: modf -> std::modf (fixes compilation on gcc)

Commit 1a908dd on 2016-05-28 by Bart Trzynadlowski

    Legacy engine: expanded color index to 12 bits from 11. Does this affect any
    games?

Commit 5c1a240 on 2016-05-28 by Bart Trzynadlowski

    New engine: expanded color indices to 12 bits from 11. Added
    PolygonHeader::ColorIndex and SensorColorIndex methods.

Commit 2ef5203 on 2016-05-28 by Bart Trzynadlowski

    Forgot to describe one polygon header field in Model.cpp

Commit e4a5c34 on 2016-05-28 by Bart Trzynadlowski

    Updated polygon header descriptions

Commit 0ebb8d5 on 2016-05-28 by Bart Trzynadlowski

    Legacy engine: Added stencil buffering for layered models and shadows. Not
    as accurate as new engine because stencil buffering is applied per-model
    rather than per-polygon, so the entire model must consist of stencil-tested
    polygons. In practice, though, this seems to work fine.

Commit ce03c13 on 2016-05-28 by Ian Curtis

    Optimise non smooth texture repeat to only effect texture boundaries, rather
    than poly boundaries. Fixes a few ugly artefacts

Commit 661f19b on 2016-05-27 by Ian Curtis

    fix transparency values, values are from 0-32, not 0-16 where the top bit
    disables

Commit 2086b1c on 2016-05-27 by Ian Curtis

    Stencil layered polys, fixes shadow z fighting in some of the games. Other
    games maybe using stipple alpha (yuck), not sure yet.

Commit 8979246 on 2016-05-26 by Ian Curtis

    remove bad rom patches

Commit 3f9936d on 2016-05-26 by Ian Curtis

    don't clamp after spot light code, looks much closer to h/w

Commit 8ba8d72 on 2016-05-26 by Bart Trzynadlowski

    Hooked up Sega Bass Fishing / Get Bass controls

Commit 89c987f on 2016-05-26 by Ian Curtis

    basic spotlight code

Commit a0910f9 on 2016-05-25 by Bart Trzynadlowski

    Fixed fog in legacy 3D engine. Now computed same as new engine, with 4-bit
    light modifier and corrected fog mixing in shaders.

Commit 6dc35f9 on 2016-05-25 by Bart Trzynadlowski

    Updated legacy engine vertex shader to use specular shininess and
    coefficient along with old equation. Still not correct (too strong compared
    to the old method).

Commit c5ce2b9 on 2016-05-22 by Ian Curtis

    fix alpha testing for texture format 7

Commit 65d6342 on 2016-05-21 by Ian Curtis

    remove dead code

Commit fd3eb49 on 2016-05-21 by Ian Curtis

    support 4 bit textures

Commit e108b56 on 2016-05-19 by Ian Curtis

    correctly mask pointers

Commit a79fd27 on 2016-05-19 by Ian Curtis

    optimise hash table access for better performance

Commit ff8d311 on 2016-05-18 by Ian Curtis

    support colour table pointer

Commit 7e3c097 on 2016-05-16 by Ian Curtis

    remove debug code

Commit 828fd8e on 2016-05-16 by Ian Curtis

    Fog start is apparently the fog value at the viewport (0,0,0) I guess like
    an ambient value. Default value is negative

Commit 9e279a0 on 2016-05-16 by Ian Curtis

    Non luminous polygons seem to be effected by the lightModifier also. API
    seems to have no restriction in this regard, neither does the SDK. The ocean
    hunter requires this for some of the 2d stuff. Light modifier values only
    seem to exist between 0-16, which is strange for 5 bits of precision.
    Possibly top bit disables the effect and lower 4 bits are the value. This
    clearly differs from the SDK, but without this the values come out clearly
    wrong.

Commit 62eafac on 2016-05-16 by Ian Curtis

    adjust (hopefully correct) fog equation

Commit bc08276 on 2016-05-15 by Ian Curtis

    preliminary specular

Commit c9147bc on 2016-05-14 by Bart Trzynadlowski

    - Removed a stale comment

Commit 897b1ac on 2016-05-14 by Bart Trzynadlowski

    Legacy renderer changes:
    - Updated shading model. Unfortunately, it is far from perfect but it seems
    to be the best I can do for now. Not really much of a difference from before
    except that the Scud Race castle is fixed at the expense of the Yosemite
    level in LA Machineguns being too bright.
    - Added lots of notes in the shading code.
    - Passing both specular coefficient and shininess to shader. This will
    probably break specular lighting for now until the shader is updated to use
    the shininess correctly.
    - Color table address in polygon RAM is now obtained from culling nodes as
    they are traversed (found this in the Pro-1000 SDK).

Commit 147faf3 on 2016-05-14 by Bart Trzynadlowski

    Added palette bit to list of bits to dump for analysis

Commit c4b5b18 on 2016-05-13 by Ian Curtis

    Disable light modifier on fog, seems not to work .. values don't make sense.
    Maybe the model3 doesn't use these.

Commit c8f8cd1 on 2016-05-13 by Ian Curtis

    remove dead code

Commit e295ad6 on 2016-05-13 by Ian Curtis

    double up polys for two sided polys, fixes some diffuse lighting problems

Commit ddf3d52 on 2016-05-10 by Bart Trzynadlowski

    FLT_MAX -> std::numeric_limits<float>::max()

Commit 8c082ab on 2016-05-10 by Bart Trzynadlowski

    Added some IRQ-related code (commented out for now) from MAME findings. May
    incorporate in the future when investigating IRQs and timing again.

Commit 34c81ea on 2016-05-09 by Ian Curtis

    update header documentation

Commit 43add90 on 2016-05-09 by Ian Curtis

    Fix bad texture seems

Commit 846c9d0 on 2016-05-08 by Bart Trzynadlowski

    Reverted stencil mask change. This reintroduces the Scud Race 'ROLLING
    START' bug. Previous 'fix' did not work for Magical Truck Adventure anyway
    and also broke some games (Daytona 2 transmission and course select screen,
    ECA title screen). The stencil mask is not shifted by 16 pixels and I
    suspect that either there is some unknown trick to horizontal scrolling or
    that the behavior is affected by hidden raster timing registers hidden in
    the JTAG interface.

Commit a9095e6 on 2016-05-08 by Bart Trzynadlowski

    Added a register read handler for tilegen (although nothing seems to benefit
    from it)

Commit 9249eaa on 2016-05-08 by Bart Trzynadlowski

    Fixed logic for clearing bottom layer. Added new functions to 2D renderer
    interface. Avoid drawing top and bottom surfaces when no layers are present
    there. Added some comments to tile

Commit 5d04895 on 2016-05-07 by Bart Trzynadlowski

    Rewrote 2D renderer and fixed 2D layer priorities, stencil mask clipping
    (Scud Race 'ROLLING START'), and cleaned up code. Unfortunately, the new
    code is about 50% slower, mostly due to the need to perform stencil clipping
    after scrolling.

Commit 69a44a5 on 2016-05-07 by Ian Curtis

    add texture helper functions

Commit 242cdda on 2016-05-07 by Ian Curtis

    remove pointless loop

Commit ded3168 on 2016-05-07 by Ian Curtis

    Split face colour from per vertex poly colour. Fixes colour bleeding between
    connected polys in harley.

Commit 595e24a on 2016-05-07 by Bart Trzynadlowski

    Changed XInput DLL failure message

Commit 05e1e4d on 2016-05-06 by Ian Curtis

    update poly header class

Commit fee1cfb on 2016-05-05 by Bart Trzynadlowski

    Fixed compilation on gcc (gotos cannot cross variable declarations)

Commit da4bab5 on 2016-05-05 by Ian Curtis

    Flat shaded polys can't share vertex normals with adjacent connected polys.
    Fixes some bad lighting in Harley on pavements

Commit b2da849 on 2016-05-05 by Ian Curtis

    support flat shading

Commit 3693a51 on 2016-05-04 by Ian Curtis

    Make sure texture is bound to correct texture unit, stop flashing micro
    textures in daytona

Commit b056900 on 2016-05-04 by Ian Curtis

    set correct microtexture wrap mode

Commit c545b9f on 2016-05-04 by Ian Curtis

    preliminary microtexture

Commit 396fba7 on 2016-05-03 by Ian Curtis

    skip viewports marked as disabled, might stop parsing some bad data

Commit 6d1aa4d on 2016-05-03 by Ian Curtis

    fix compilation

Commit fa65542 on 2016-05-03 by Bart Trzynadlowski

    Added support for A1RGB5 to BMP file writer and added a (commented-out)
    example of how to use it to dump texture memory in ~CReal3D()

Commit c459772 on 2016-05-01 by Ian Curtis

    wrap invalid texture sizes, seems to fix bad textures in srally2 (skid
    marks)

Commit 7ec7906 on 2016-04-30 by Ian Curtis

    Turn off lighting if fixed shading (or no normals) flag is present. Fixes
    some of the lighting weirdness in scud

Commit 48767b2 on 2016-04-30 by Bart Trzynadlowski

    Legacy engine: fixed per-vertex fixed shading to incorporate Ian's findings
    that the intensity value is signed and that fixed shading applies only when
    lighting is enabled

Commit 23a5839 on 2016-04-30 by Ian Curtis

    Daytona seems to use 3 bits for texture transparency, will figure it out
    later but for now this works

Commit 30ca179 on 2016-04-30 by Ian Curtis

    Fix fixed shading maths

Commit e0513ec on 2016-04-29 by Bart Trzynadlowski

    Updated to-do comments

Commit c499e66 on 2016-04-29 by Ian Curtis

    fixed shading code

Commit 57d5bc8 on 2016-04-29 by Bart Trzynadlowski

    Legacy engine: preliminary per-vertex fixed shading. Only limited testing
    with LA Machineguns (appears to work), Sega Rally 2 (appears okay), and Scud
    Race (fixes expert course bridge but not other mis-colored polygons).

Commit 2785ecd on 2016-04-28 by Ian Curtis

    read front buffer

Commit 7456aa7 on 2016-04-28 by Bart Trzynadlowski

    Luminous bit is now also dumped in polygon header analysis because shading
    modes may depend on whether or not lighting is enabled

Commit 34e8f00 on 2016-04-28 by Bart Trzynadlowski

    Legacy engine: extracting fixed shading component (but still can't use it
    because unsure of how to activate)

Commit 0ae9e80 on 2016-04-28 by Bart Trzynadlowski

    Fixed polygon header bit analysis code: was forgetting to swap GL buffer
    back to retrieve pixels from last rendered frame

Commit 09e27e3 on 2016-04-27 by Ian Curtis

    color flag

Commit 74e7ff3 on 2016-04-27 by Ian Curtis

    update project

Commit a246fc1 on 2016-04-27 by Ian Curtis

    make compilable with visual studio

Commit 8f6ab2f on 2016-04-27 by Ian Curtis

    simplify

Commit 3c5fcbd on 2016-04-27 by Bart Trzynadlowski

    Forgot a newline

Commit c0ec610 on 2016-04-27 by Bart Trzynadlowski

    Polygon header bit analysis code (need to #define DEBUG and use -gfx-state
    option)

Commit 04fba14 on 2016-04-27 by Bart Trzynadlowski

    Added utility function to write BMP files

Commit 3682ab7 on 2016-04-26 by Bart Trzynadlowski

    Updated polygon header description with double-sided bit (forgot it in last
    update!)

Commit 85a77bb on 2016-04-26 by Bart Trzynadlowski

    Added an updated description of the polygon header to the top of
    Legacy3D/Models.cpp

Commit 3e8439c on 2016-04-26 by Ian Curtis

    fix model normals, add code for fixed shading, as of yet flag to turn it on
    is unknown

Commit b730fc4 on 2016-04-26 by Bart Trzynadlowski

    Legacy engine: vertex normals are standalone and do not need to be added to
    polygon normal

Commit 1127773 on 2016-04-24 by Bart Trzynadlowski

    Added a debug option to load and render a single frame of graphics state
    from a save state file. Only available if compiled with DEBUG defined.
    Invoked using: -gfx-state=<file> on command line.

Commit 1265c8d on 2016-04-24 by Bart Trzynadlowski

    Using IEmulator interface instead of CModel3 directly

Commit 9baa048 on 2016-04-24 by Bart Trzynadlowski

    Changed save state file version (r340 changed Real3D save format, breaking
    compatibility with older states, and states were probably already
    incompatible with 0.2a)

Commit 3bd8e20 on 2016-04-24 by Bart Trzynadlowski

    Fixed white space and warnings. Fixed second argument to -dis option in
    cases where other numerical parameters are specified on command line.

Commit 908d979 on 2016-04-22 by Bart Trzynadlowski

    Flattened out nested if-statements in new engine's fragment shader to
    improve compatibility with older GPUs (e.g., NVS 300)

Commit 58d187e on 2016-04-19 by Ian Curtis

    parse viewports in normal order

Commit df9fefd on 2016-04-18 by Ian Curtis

    walk siblings before children, fixes some of the transparency drawing issues
    in ocean hunter

Commit 057ee94 on 2016-04-18 by Ian Curtis

    simplify

Commit a5301ef on 2016-04-17 by Ian Curtis

    avoid cast warning

Commit c4dc96e on 2016-04-17 by Ian Curtis

    simplify list code

Commit ad96676 on 2016-04-15 by Bart Trzynadlowski

    Updated Mac OS X Makefile

Commit 34d4d89 on 2016-04-15 by Bart Trzynadlowski

    Fixed 8-bit texture decoding in legacy 3D engine. Contour mode/alpha still
    needs to be made consistent (as much as possible) with new engine.

Commit d6ce41f on 2016-04-14 by Bart Trzynadlowski

    Updated Supermodel.ini with default Magical Truck Adventure controls

Commit 9a2bb78 on 2016-04-13 by Bart Trzynadlowski

    Hooked up Magical Truck Adventure controls

Commit d2563c0 on 2016-04-11 by Ian Curtis

    you were right Bart :) (Fixes broken sound in last build)

Commit 499ac0e on 2016-04-11 by Ian Curtis

    Don't use variable sized variables across platforms, unless needed

Commit 6d313fc on 2016-04-11 by Ian Curtis

    Fix - cannot specify explicit initializer for arrays

Commit 94c66ee on 2016-04-11 by Bart Trzynadlowski

    Figured out Step 1.0 VROM texture uploads. Fixes missing textures in VF3 and
    VF3TB.

Commit ef87093 on 2016-04-11 by Bart Trzynadlowski

    Debugger compilation fixed: CBus -> IBus

Commit 2524f2e on 2016-04-11 by Bart Trzynadlowski

    Fixed Magical Truck Adventure by emulating VROM texture ports as a FIFO.
    Now, VF3 crashes on Akira's stage (and probably the other stages with
    texture glitches). Other games may be affected, too (not thoroughly tested
    yet).

Commit 7752561 on 2016-04-10 by Bart Trzynadlowski

    Fixed Magical Truck Adventure country check (accidentally removed a line in
    previous commit).

Commit 7ade3a6 on 2016-04-10 by Bart Trzynadlowski

    Magical Truck Adventure expects 0x03 to be returned from MIDI control port
    (thanks, Ville). Game now boots but crashes because VROM texture uploads
    clobber memory for some reason.

Commit 198a469 on 2016-04-10 by Bart Trzynadlowski

    Created IEmulator interface, renamed CBus to IBus and CPCIDevice to
    IPCIDevice. Frames now rendered when emulator paused.

Commit f52cb3c on 2016-04-08 by Ian Curtis

    tweak frustum planes

Commit e08b754 on 2016-04-05 by Bart Trzynadlowski

    Changed help text to explain that -sound-volume and -music-volume only take
    effect when DSB is present

Commit 22907b0 on 2016-04-05 by Bart Trzynadlowski

    Fixes for non-Windows builds: Makefile updated and fixed header files in
    Vec.cpp

Commit 5f68544 on 2016-04-04 by Ian Curtis

    revert viewport code

Commit ecea4ca on 2016-04-04 by Ian Curtis

    change tex format 4

Commit 38d5d14 on 2016-04-04 by Bart Trzynadlowski

    Clamping diffuse+ambient light intensity to 1.0 (legacy engine)

Commit faf5f9c on 2016-04-04 by Bart Trzynadlowski

    Added a comment about viewports

Commit 6d94267 on 2016-04-04 by Bart Trzynadlowski

    Fixed texture enable bit (fixes Star Wars light sabers and Sega Rally 2
    cones) and added a hack for (and more comments about) color modulation. Scud
    Race, Sega Rally 2, LA Machineguns, etc. look correct now, but Fighting
    Vipers 2 shadows are no longer dark enough. Other bugs may exist. 

Commit 1d4f5c8 on 2016-04-04 by Bart Trzynadlowski

    Fixed Types.h include

Commit a202d8a on 2016-04-04 by Bart Trzynadlowski

    Do not print extra hardware message when there is none

Commit d8aefa4 on 2016-04-04 by Bart Trzynadlowski

    8-bit texture upload byte selection (thanks to Ville; but not tested yet);
    code cleanup

Commit d155fbd on 2016-04-03 by Ian Curtis

    rgb values were swapped for paletted colours

Commit 3bc2794 on 2016-04-03 by Ian Curtis

    Use the correct texture enable bit

Commit 3c8352f on 2016-04-03 by Bart Trzynadlowski

    Added a comment: observation of Star Wars and Scud Race color modulation
    issues.

Commit f09dd45 on 2016-04-02 by Ian Curtis

    Update vs project for encryption code

Commit 688af26 on 2016-04-02 by Ian Curtis

    Possible work around for slightly incorrect viewport positions in ECA and
    harley

Commit a99f022 on 2016-04-02 by Ian Curtis

    add missing include

Commit 6e5c301 on 2016-04-02 by Bart Trzynadlowski

    Encryption device emulation (thanks to MAME), fixed warnings in Model3.cpp,
    added a string formatter helper, and updated Win32 GCC Makefile.

Commit c0f6794 on 2016-04-02 by Bart Trzynadlowski

    Added virtual destructor to CDSB

Commit 0d98983 on 2016-04-02 by Bart Trzynadlowski

    Added encryption keys, converted to cstdint types, fixed whitespace

Commit c7ad13f on 2016-04-02 by Bart Trzynadlowski

    Whitespace, cstdint types

Commit 9c2aff7 on 2016-04-02 by Ian Curtis

    add missing file

Commit c741282 on 2016-04-02 by Ian Curtis

    Use proper texture wrapping mode for 2d. Stops garbage down the top and left
    sides. Should really use NPOT texture.

Commit 736f5ab on 2016-04-01 by Ian Curtis

    add missing project files

Commit 7e7847b on 2016-03-31 by Ian Curtis

    Allow asymmetric projection matrices. Fixes a bunch of missing effects and
    bad camera angles in harley. 

Commit c398cab on 2016-03-29 by Ian Curtis

    update project

Commit aee8939 on 2016-03-28 by Ian Curtis

    add LOD table data. Maybe we can get clipping planes out of these

Commit e79b241 on 2016-03-28 by Ian Curtis

    properly mark dynamic meshes as such

Commit e1278f4 on 2016-03-27 by Ian Curtis

    allow 8 bit  mono textures to be contour textures, fixes a bunch of missing
    transparency in harley la riders 

Commit ffaed8a on 2016-03-26 by Ian Curtis

    clear the rom mesh buffer 

Commit 4992e59 on 2016-03-26 by Ian Curtis

    Cache rom models, for better performance. 

Commit 8488f22 on 2016-03-24 by Ian Curtis

    

Commit 0327db2 on 2016-03-24 by Ian Curtis

    WIP

Commit 23172c9 on 2016-03-23 by Ian Curtis

    fix for bad matrix with NaN

Commit 5c51344 on 2016-03-23 by Bart Trzynadlowski

    Fixed detected axis printf() and compiler warnings (hopefully I did not
    parenthesize any conditionals incorrectly!)

Commit f031e5d on 2016-03-22 by Ian Curtis

    finish front/back face culling code based on matrix determinant. Optimised
    opengl to avoid redundant state changes.

Commit 44ea902 on 2016-03-22 by Ian Curtis

    add missing project file

Commit 8ea5978 on 2016-03-22 by Ian Curtis

    fix parenthesis error 

Commit e7eb912 on 2016-03-22 by Ian Curtis

    add maths functions

Commit 93f6b01 on 2016-03-22 by Ian Curtis

    Update project to vs 2013. Clean compile warnings

Commit ce99c47 on 2016-03-22 by Ian Curtis

    Remove double copy of shaders

Commit 5b3eea2 on 2016-03-22 by Ian Curtis

    fix basic compile errors (with vs)

Commit 09747bc on 2016-03-22 by Bart Trzynadlowski

    Tweaked Windows gcc Makefile

Commit 3627213 on 2016-03-22 by Bart Trzynadlowski

    Fixed compiler warnings, whitespace, and moved local variable declarations
    to site of assignment

Commit ec7588e on 2016-03-22 by Bart Trzynadlowski

    Include Version.h

Commit 028727b on 2016-03-22 by Bart Trzynadlowski

    Fixed typo that broke build

Commit 4c7186b on 2016-03-22 by Bart Trzynadlowski

    Fixed warnings and whitespace

Commit 7860968 on 2016-03-22 by Bart Trzynadlowski

    Fixed warnings and whitespace

Commit bd09b3b on 2016-03-22 by Bart Trzynadlowski

    Fixed warnings and whitespace

Commit f0488a3 on 2016-03-22 by Bart Trzynadlowski

    Moved lastCycles static int out of header file (fixes compiler warnings)

Commit 3afe3cf on 2016-03-21 by Bart Trzynadlowski

    Moved

Commit 47c54d1 on 2016-03-21 by Bart Trzynadlowski

    Updated to support new IRender3D interface and added FrameTimings object.

Commit 708455a on 2016-03-21 by Bart Trzynadlowski

    CRender3D -> CLegacy3D

Commit 60ce618 on 2016-03-21 by Bart Trzynadlowski

    Update Supermodel.h with graphics path and moved version string to separate
    header file

Commit d61b01a on 2016-03-21 by Bart Trzynadlowski

    Removing legacy 3D engine files that were already moved to Graphics/Legacy3D

Commit 34301e9 on 2016-03-21 by Bart Trzynadlowski

    Adding new 3D engine by Ian Curtis. Old engine moved to Graphics/Legacy3D/.

Commit 9802bbb on 2016-03-21 by Bart Trzynadlowski

    Fixed compilation errors involving min and max functions

Commit 0579e65 on 2016-03-21 by Bart Trzynadlowski

    Added registers to debugger: ctr, xer, srr0, srr1, sdr1, imiss, dmiss, hid0,
    hid1

Commit 27c20ff on 2016-03-21 by Bart Trzynadlowski

    Fixed compiler warnings

Commit 8f87bb1 on 2016-02-23 by Bart Trzynadlowski

    53c810: do not increment DSP when executing interrupt instruction. Fixes
    invalid polygons in Virtua Fighter 3. Thanks to Ian Curtis for discovering
    that the polygons were incorrectly-copied garbage.

Commit 8e6712d on 2015-12-07 by Bart Trzynadlowski

    Fixed display of PowerPC condition code bits in debugger (were backwards)
    and changed XER to print as hex.

Commit 3d00225 on 2015-12-07 by Bart Trzynadlowski

    Support for loading an arbitrary (non-game) CROM image.

Commit 1035b0c on 2015-06-13 by Bart Trzynadlowski

    Gracefully handle the case when a zip file contains no known Model 3 ROM
    set.

Commit 6bf4244 on 2015-06-13 by Bart Trzynadlowski

    Converted line endings from DOS to UNIX format.

Commit ee8afa1 on 2015-05-12 by Bart Trzynadlowski

    Contribution by Luciano Lopez:

    - Added vs215o ROM set.
    - Security board data for vs298 now also returned for vs2v991, vs299,
    vs299a, and vs299b, helping them run.

Commit 8835d6f on 2013-11-30 by Nik Henson

    Some updates to Supermodel made at beginning of the year but only now got
    around to checking in (better late than never...):
    - hooked up the remaining controls in Supermodel (except for Magical Truck
    Adventure which does not work at all yet).  The new controls are:
     * InputAnalogJoyTrigger2 and InputAnalogJoyEvent2 for the additional second
    trigger and event buttons that were missing from Star Wars Trilogy,
     * InputRearBrake and InputMusicSelect for the rear brake and music
    selection buttons that were missing from Harley Davidson,
     * InputAnalogGunXXX, InputAnalogTriggerXXX, InputAnalogGunXXX2 and
    InputAnalogTriggerXXX2 for the analogue guns of Ocean Hunter and LA
    Machineguns (NOTE: these controls must be calibrated in the games' service
    menus otherwise they will not work properly.  Also, the alignment of the gun
    cursor does not line up very well with the mouse position at the moment, but
    at least the games are a bit more playable now, although still with numerous
    graphical glitches...)
     * InputSkiXXX for the controls of Ski Champ, making the game playable now.
    - hooked up existing InputViewChange control to Harley Davidson's view
    change button
    - improved the handling of InputGearShiftUp/Down inputs so that they work
    better with the driving games.  With Dirt Devils, ECA, Harley and LeMans
    this means they map directly to the game's own shift up/down controls, while
    with the 4-speed games such as Daytona 2, Scud Racer and Sega Rally 2, they
    simulate the user shifting up and down through the gears
    - added defaults for the new controls to Supermodel.ini
    - other small code tweaks:
     * fix small bug with handling of pos/neg inputs mapping to a control with
    inverted range (0XFF to 0x00) - this was needed to get Ski Champ's X-axis to
    work properly
     * removed Wait method from InputSystem and added to CThread as
    CThread::Sleep instead
     * added FrameTimings struct to hold all frame timings in a single place No
    networking code yet as just haven't had a chance to work on it since initial
    progress at the beginning of the year - am *hoping* might have some time to
    pick it up again over Christmas...


Commit b017b24 on 2012-09-06 by Bart Trzynadlowski

    Changed output "Pause" to "pause", as it is spelled in MAME, for MAMEHooker.

Commit 08d347c on 2012-07-24 by Nik Henson

    Small tweak to get to compile under GCC (Unix/MacOS)

Commit a3c15e8 on 2012-07-24 by Nik Henson

    Slight tweak to changes made to sound board synchronisation code in previous
    commit.  The previous change was necessary to ensure no deadlock when
    waiting for sound board thread to exit but it introduced unwanted buffer
    overruns (with the end result being that the sound was often speeded up) so
    have added an extra check in WakeSoundBoardThread to prevent this.


Commit 108ac64 on 2012-07-23 by Nik Henson

    - CModel3::StopThreads was rather gracelessly killing threads while they
    were waiting on their semaphores, which works okay Windows but not on Mac
    OS-X where it was preventing Supermodel from exiting properly.  Hence, have
    instead altered StopThreads to signal to threads that they should exit and
    wait until until they have all done so.  This is much better and is what I
    should have done the first time around had I not been so lazy :-)!
    - Also fixed a small race condition in WakeSoundBoardThread.  Previously if
    it happened to be called just before the sound board thread had finished
    processing a frame's worth of audio but before it had got around to waiting
    on the sync condition variable then the wake notification would be lost.


Commit f1e93bc on 2012-07-16 by Bart Trzynadlowski

    - Fixed Win32 GCC Makefile.
    - Reverted Nik's polygon palette index change in Models.cpp. Subtracting 1
    was breaking Scud Race.
    - Added an explicit typecast in Inputs.cpp, which GCC requires for some
    reason.

Commit 183dca5 on 2012-07-15 by Nik Henson

    Committing various small updates that have been hanging around in my source
    tree for a while now:
    - Added 'crosshairs' command line and config option.
    - Added 'vsync' command line and config option (so far only tested on NVidia
    cards on Windows 7 - other graphics drivers, O/Ss or driver settings may
    simply chose to ignore this).
    - Added fullscreen toggle within game using Alt+Enter key combination.
    - Added framework for lamp outputs and 'outputs' command line and config
    option.  So far only the lamps for driving games are hooked up in the
    emulator (others to be added later).
    - Added an initial outputs implementation for Windows that sends MAMEHooker
    compatible messages (-outputs=win to enable)
    - Fixed fps calculation in Main.cpp that was producing incorrect results and
    so giving the impression that frame throttling wasn't working properly when
    in fact it was.
    - Fixed palette indexed colours as the index was always off by one, causing
    incorrect colours in various games, eg drivers' suits and flashing Start
    sign in Daytona 2.
    - Altered caching of models so that models with palette indexed colours use
    the dynamic cache rather than the static one.  This is so that changes in
    palette indexed colours appear on screen, eg the flashing Start sign on the
    advanced course of Daytona 2 (although currently the START message itself is
    not visible due to other problems with texture decoding).
    - Fixed small bug in TileGen.cpp which meant both palettes were being
    completely recomputed pretty much with every frame.  This was a significant
    performance hit, particularly as palette recomputation is currently being
    done in SyncSnapshots (it should be moved out of here at some point,
    although for now it's no big deal).
    - Made sure all OpenGL objects and resources are deleted in Render2D/3D
    destructors, in particular the deleting of the VBO buffer in
    DestroyModelCache.
    - Made sure that GLSL uniforms are always checked to see if they are bound
    before using them in order to stop unecessary (but harmless) GL errors.
    - Altered the default texture sheet handling to use a single large GL
    texture holding multiple Model3 texture sheets rather than multiple GL
    textures as before (if required, the old behaviour can still be selected
    with the mulisheet fragment shader).  I believe this fixes the disappearing
    crosshairs/corrupt GL state problem which the multisheet fragment shader
    seemed to be triggering somehow.
    - Fixed a bug in debugger which meant memory watches were not triggering
    properly 


Commit ac29fc1 on 2012-07-12 by Bart Trzynadlowski

    Somehow John's Build.bat and Clean.bat did not get added in r262. Here they
    are.

Commit 772c94d on 2012-07-12 by Bart Trzynadlowski

    - Committing Nik's timing changes: the PowerPC's DEC and TBR now count at
    the correct rate and time is accurately maintained between calls to the
    emulator. A VBlank bit was identified in the Real3D status register space
    and is now timed. Unfortunately, there are still some problems and not all
    games are perfect, but FV2, VS2, and others are working much better or
    perfectly.
    - Added a new dump of Scud Race Plus (scudp1). Strangely, this one uses an
    MPC106 instead of the MPC105 used by every other Step 1.5 game... 

Commit 3963a7a on 2012-07-12 by John Peterson

    Applied a series of patches by John Peterson:

    - Fixed a printf statement in ConsoleDebugger.cpp.
    - Fixed memory watches not breaking in CPUDebug.h.
    - Added some helpful batch files to the VS2008 directory.
    - Added support for automatic loading of parent ROM sets.

Commit 93b71c8 on 2012-03-31 by Bart Trzynadlowski

    Applied Nik's fix for cyclical scene graphs. Improves Harley Davidson, Ski
    Champ, Emergency Call Ambulance, and Ocean Hunter.

Commit 1bce444 on 2012-02-28 by Bart Trzynadlowski

    Brought in the near clipping plane for Step 1.5+. Objects close to camera
    are no longer cut-off (e.g., driver arms and steering wheels in cockpit
    views). Note: This change should probably be applied to Step 1.0 as well in
    the future. For now, I am leaving the old distance because Step 1.0
    coordinates need closer investigation. Bringing in the near clipping plane
    extends the view frustum depth and will worsen Z-buffer accuracy. Be on the
    lookout for Z-fighting that did not previously exist.

Commit f9c3b9b on 2012-02-27 by Nik Henson

    DirectInputSystem's initial state wasn't being initialized properly, which
    would sometimes result in invalid input values until keyboard or mouse was
    successfully acquired

Commit 69610ef on 2012-02-25 by Bart Trzynadlowski

    Added a clear() method to VBORef and removed unsafe memset() clear.

Commit 3b41239 on 2012-02-23 by Nik Henson

    This update fixes (hopefully) the last of the texture upload lags that were
    affecting some games, in particular Daytona 2 and Spikeout. To achieve this
    UploadTextures no longer clears the model cache when called.  Instead the
    cache is kept in-tact (which should help improve cache hits) and all
    textures referenced by models being rendered are (re-)decoded with every
    frame. To help with tracking all the unique texture references contained in
    a model a new class TextureRefs has been added.


Commit aba801e on 2012-02-23 by Bart Trzynadlowski

    Byte swapped sample ROMs. This should fix the audio distortion that many
    users noticed. The differences are particularly noticeable in Scud Race and
    Daytona 2.

Commit 16e4fca on 2012-02-23 by Bart Trzynadlowski

    Forgot to update to-do list in previous revision. Added a comment to mark
    CRender2D::WriteVRAM() as deprecated.

Commit fcf1d95 on 2012-02-22 by Bart Trzynadlowski

    Previous "optimization" contained a mistake and made things worse. Removed
    unnecessary clear of the bottom layer buffer. Now, if there is no bottom
    layer, it will not be displayed at all and glClear(GL_COLOR_BUFFER_BIT) is
    used instead.

Commit c085be5 on 2012-02-20 by Bart Trzynadlowski

    Further optimization/simplification of the 2D renderer's main loop.

Commit 453df4f on 2012-02-20 by Bart Trzynadlowski

    - Finished cleaning up and optimizing the 2D renderer.
    - Fixed up color offset register support for new 2D rendering system. Now
    maintains 2 computed palettes for layers A/A' and B/B'.
    - Fixed a minor bug in InitPalette(); VRAM was not being typecast properly.
    - Fixed specular lighting bug that occurred on some OpenGL drivers because
    integers were not being interpreted as floats in the vertex shader.
    - Began to update copyright date in some files.
    - Graphics modules now use the C++-style names for C standard library
    headers (e.g., stdio.h -> cstdio) consistent with the rest of Supermodel.

Commit 0257f1f on 2012-02-17 by Bart Trzynadlowski

    Contour processing (discarding of pixels when alpha==0) enabled for A4L4
    texture formats. Improves smoke and fire in Daytona 2.

Commit 5709010 on 2012-02-14 by Bart Trzynadlowski

    Default SCSP balance set to 0. Was erroneously set at +50%.

Commit 79d24d4 on 2012-02-14 by Bart Trzynadlowski

    - NOTE: In this revision (possibly earlier), I've started to notice some
    intermittent geometry glitches in Spikeout and slowdown while playing
    certain characters.
    - Added multi-texture fragment shader to repo.
    - Added a multiTexture option (enabled by default) to use multi-texturing to
    decode textures.
    - Added some comments regarding timing to the PowerPC execution loop.

Commit 30130f2 on 2012-02-13 by Nik Henson

    Updated fragment shaders' floating point comparisons and also some knock-on
    changes from SVN revision 246.


Commit 8a4ea64 on 2012-02-13 by Nik Henson

    Added OSD/Video.h to provide OSD-dependent video functionality.  The static
    functions in here get called by Model3.cpp from within the render thread so
    that any OSD related rendering can also be run in parallel for a further
    speed increase. Updated OSD/SDL/Main.cpp to provide the SDL implementations
    of the functions in OSD/Video.h.


Commit 84eb017 on 2012-02-13 by Nik Henson

    Added support for multiple texture sheets (with up to one per Model 3
    texture format) as a rather brute-force way to handle overlapping texture
    formats in the current 3D engine.  This fixes some corrupt textures in
    Daytona 2 and Virtua Striker 2 (and possibly other games) and also offers a
    small speed increase when some scenes load multiple overlapping textures. 
    This feature only enables itself when a compatible shader script is loaded. 
    Since none have been checked in yet this means it is currently disabled.


Commit 35a47bc on 2012-02-11 by Bart Trzynadlowski

    Added command line options to force windowed mode and to control GPU
    multi-threading.

Commit 1b87401 on 2012-02-10 by Bart Trzynadlowski

    Added a wide screen hack that can be enabled with the -'wide-screen'
    argument or WideScreen in the config file.

Commit 40a5128 on 2012-02-09 by Bart Trzynadlowski

    Tweaked help text slightly.

Commit 09f9bd8 on 2012-02-09 by Bart Trzynadlowski

    Fixed left/right stereo channels -- they were reversed by default.

Commit c5d0cb0 on 2012-02-09 by Bart Trzynadlowski

    Added front/rear speaker balance (actually the master/slave SCSP balance).
    Balance setting is stored in CSoundBoardConfig and can be set on the command
    line (-balance) or config file (Balance). Valid range is 100 (front speaker
    volume doubled, rear muted) to -100. The default is 0, both sets at full
    volume. Balance is applied in the SCSP core before overall sound and music
    volume settings.

Commit 6814ef1 on 2012-02-06 by Bart Trzynadlowski

    Added preliminary specular lighting to shaders. The "no spotlight" shader
    has not yet been updated.

Commit 9fc0d7b on 2012-02-01 by Bart Trzynadlowski

    Updated some comments regarding the VF3 ROM patches. It seems many of them
    are no longer necessary. The single necessary patch affects timing (note how
    the warning screen stays up longer when it is removed). This needs to be
    investigated -- perhaps it can lead to a timing fix for vf3, vs2, and
    fvipers2?

Commit 0fcb968 on 2012-01-27 by Bart Trzynadlowski

    Documented the tile generator.

Commit 8a8609f on 2012-01-27 by Bart Trzynadlowski

    - Added specular highlight support to the 3D engine. A shininess parameter
    is now be passed to the shaders. Does not work correctly and I will save the
    shaders themselves for a future commit.
    - Added support for layer priorities. Highly unoptimal! This absolutely
    needs to be polished up before any release (and properly documented). For
    now, priorities in the popular games seem to be fixed but not all possible
    priority settings have been figured out yet.

Commit c38ec7d on 2012-01-26 by Bart Trzynadlowski

    Removed unused dirty rectangle code.

Commit e00a372 on 2012-01-26 by Bart Trzynadlowski

    Removed unused dirty rectangle code.

Commit b7980aa on 2012-01-26 by Bart Trzynadlowski

    Removed unused dirty rectangle code.

Commit 93d68c5 on 2012-01-22 by Nik Henson

    Added missing library WbemUuid.lib to VS project linker options


Commit 3315fb4 on 2012-01-22 by Bart Trzynadlowski

    Small correction to lighting model in shaders. Light intensity is no longer
    clipped.

Commit b57ba51 on 2012-01-18 by Bart Trzynadlowski

    Removed unused (and non-functional) 'stack machine' scene graph traversal.

Commit 4c7fe2f on 2012-01-18 by Bart Trzynadlowski

    MinGW Makefile now uses cmd prompt commands, not UNIX/msys.

Commit 1d64fc2 on 2012-01-18 by Bart Trzynadlowski

    Added a MinGW Makefile for users of GCC on Windows.

Commit dd4a132 on 2012-01-18 by Bart Trzynadlowski

    Removed deprecated, unused Turbo68K stuff.

Commit 611ef43 on 2012-01-18 by Bart Trzynadlowski

    Now compiles with MinGW. Among other things, the use of __uuidof() was
    removed.

Commit d1d5175 on 2012-01-16 by Nik Henson

    New multi-threaded rendering changes that parallelise graphics rendering and
    PPC execution in order to increase performance on multi-core machines. New
    gpuMultiThreaded config option to enable/disable multi-threaded rendering
    (enabled by default, disabling it reverts to previous behaviour). Other
    rendering optimisations:
    - texture uploads now only affect appropriate region in the texture sheet,
    rather than uploading the whole sheet each time
    - performance of clearing the model caches has been improved New Alt+O key
    input added to toggle outputting of frame timings for debugging purposes.


Commit 0835e38 on 2012-01-16 by Nik Henson

    Removed unecessary SDL include


Commit a1a9939 on 2012-01-16 by Nik Henson

    Added missing shader source files into VS2008 project (not needed for build,
    just for reference)


Commit b7242d2 on 2012-01-16 by Nik Henson

    Added CThread::Sleep && CThread::GetTicks methods


Commit 6703601 on 2012-01-02 by Nik Henson

    Fixed Unix Makefile to link okay on Ubuntu 11.10 (archive libraries needed
    to be referenced at end of linker options)


Commit 2ece814 on 2011-12-27 by Bart Trzynadlowski

    Assigned new version number: 0.3a-WIP.

Commit 532eafe on 2011-12-26 by Bart Trzynadlowski

    Fixed a bug w/ Read16() -- was not casting to a pointer properly, caused
    Scud Race to crash.

Commit bdf11c7 on 2011-12-26 by Bart Trzynadlowski

    Preliminary backface culling fix. Matrices which effectively change the
    polygon winding without changing the Z component of the polygon normal are
    tested for when appending items to the display list. If necessary, the
    polygon winding is changed on the fly to ensure visibility is correct. This
    code desperately needs to be optimized. Note: Not yet thoroughly tested.

Commit a71af94 on 2011-12-23 by Bart Trzynadlowski

    Fixed ROM banking -- appears to fix the sounds in Daytona 2 PE and VON2 but
    otherwise has not been extensively tested. The high 8MB of the 68K address
    is treated as a single window into the 16MB sample ROM, and can point either
    to the low or high 8MB bank.

Commit 4c3d6e1 on 2011-12-23 by Bart Trzynadlowski

    Fixed a MAGIC_NUMBER comment typo.

Commit 885d946 on 2011-12-23 by Nik Henson

    Small bug fix


Commit 6855410 on 2011-12-23 by Nik Henson

    Small tweak to improve startup time


Commit b9375bb on 2011-12-23 by Nik Henson

    Optimisations to improve performance when running with debugger.  Now runs
    at about 2/3 speed of full speed of emulator, rather than 1/3 like before.


Commit 2e33745 on 2011-11-06 by Nik Henson

    Added missing DirectX include and library paths to project

Commit 6220468 on 2011-11-06 by Nik Henson

    Added quotes around paths in post-build events to allow for spaces in
    directory names

Commit a6b31b7 on 2011-11-05 by Nik Henson

    Fixed incorrect ZLib paths


Commit 8f0e0ed on 2011-11-05 by Nik Henson

    Updated comment in DIR.txt (again!)

Commit 4f7c038 on 2011-11-05 by Nik Henson

    Updated comment in DIR.txt.


Commit 4c26932 on 2011-11-05 by Nik Henson

    Removed sub-directories of Libraries as are created when user unzips source
    archives, so might be confusing.


Commit 1451f8c on 2011-11-05 by Nik Henson

    Added DIR.txt to explain purpose of Libraries directories


Commit fc27908 on 2011-11-05 by Nik Henson

    Added place holder directories for third-party libraries (eg ZLib and SDL)


Commit 663a7f9 on 2011-11-05 by Nik Henson

    Changes to Supermodel VS2008 solution to ensure it will now build completely
    from scratch:
    - added projects to build third-party libraries ZLib and SDL (expects source
    code to be present in Libraries directory)
    - moved Musashi68K into its own directory
    - tidied up some config options


Commit dfa77e9 on 2011-11-01 by Nik Henson

    Added support for joystick sliders to input system. Added extra debugging
    output to calibration input detection routine (activated by pressing Shift).


Commit e431b79 on 2011-11-01 by Nik Henson

    Fixed small bug which meant wasn't correctly clearing SCSP structures on
    initialisation.


Commit a18c260 on 2011-11-01 by Nik Henson

    Added separate Musashi68K project that compiles and runs m68kmake to build
    required Musashi68K file. Updated solution so that with above change it will
    now build Supermodel completely from scratch.


Commit e6fd2e2 on 2011-10-02 by Nik Henson

    Added DirectInputConstForceLeftMax and DirectInputConstForceRightMax options
    to allow different force feedback strengths in each direction. 
    DirectInputConstForceMax still works as before but these new options will
    override it for the given direction if they are specified.


Commit 46211fd on 2011-09-30 by Bart Trzynadlowski

    Fixed a comment in default ini file.

Commit 0c4f938 on 2011-09-25 by Nik Henson

    Force feedback fixes in DirectInputSystem.cpp:
    - Fixed bug which meant right constant force was not being capped to
    DI_EFFECTS_MAX
    - Made sure all values are being clamped to sensible ranges

