Triforce v0.5.2

The new launcher (Windows and Linux: the game's own look, music and settings, saves and mods pages, smaller updates in parts), benchmark mode (Find my best settings: the game tours four places by itself and writes its results), Quit and Restart in the pause menu under Mods, Squadala! in the Mods menu, battle music above 60 fps, menu text, the controller in the menus with the game in the background, the whole GameBanana list, RX 580 and RX 5000/6000 shader fixes, the pipeline cache kept through an update, async present black flashes, and faster builds.
This commit is contained in:
ruckus7382sfgsgdf
2026-11-19 10:52:50 -06:00
parent a750a21025
commit fb26495475
167 changed files with 21552 additions and 293 deletions

View File

@@ -435,7 +435,22 @@ bool create_device() {
logf("vk: vendor fixes for %s (%s): out-of-bounds texel fetches 0 %d, NaN tests by bits %d, gather offset %d",
drv.driverName, drv.driverInfo, (int)dev.fixes.fetch_bounds, (int)dev.fixes.nan_bits, (int)dev.fixes.gather_offset);
}
dev.eds3_a2c = dev.dynamic_state3 && eds3.extendedDynamicState3AlphaToCoverageEnable;
// ((Triforce, 11-18, EstebanGGM's port of NX2navi issue #67) AMD's older proprietary shader compiler (vk_vendor.h
// amd_older_compiler) applies alpha to coverage while the dynamic state says off: white stippling over foliage, grass, Link's
// hair and tunic on RX 470-590 (#bug-tracker "Texture glitch again ... (RX 580)", 11-18 screenshots). There the state goes in
// the pipeline instead. Only where the driver has no fast-linking pipeline libraries (the RX 580's 26.5.2: "pipeline
// libraries no"): fast pipeline link needs this dynamic state (gpl_wanted), and an RX 6000 card on the same driver keeps it.
// Those cards' pipelines are new to the driver's caches once (the state is in them now). RX_VK_DYN_A2C=0/1 decides outright
{
const bool eds3_a2c = dev.dynamic_state3 && eds3.extendedDynamicState3AlphaToCoverageEnable;
bool want = !(amd_older && !dev.graphics_pipeline_library);
if (const char* e = getenv("RX_VK_DYN_A2C"); e && *e) want = atoi(e) != 0;
dev.eds3_a2c = eds3_a2c && want;
if (eds3_a2c && !dev.eds3_a2c)
logf("vk: alpha to coverage in the pipelines, not dynamic (%s; RX_VK_DYN_A2C=1: dynamic)",
getenv("RX_VK_DYN_A2C") && *getenv("RX_VK_DYN_A2C") ? "RX_VK_DYN_A2C=0"
: "AMD's older compiler draws it while it is off");
}
dev.vk12 = f12;
dev.vk13 = f13;
{ // (hookshot, after TotK Recomp's vk_caps.c) NVN's programs are written for 32 lanes: the translator is told 32, and the lane