Merged on top of #566. The camera motion is #594's: its blocks in camb also feed the per-object pass, and
dlaa_mv.comp stays as on main, fed the identity and the reprojection made on the CPU in double. #566's
vk_usnap_host goes, vk_usnap_bytes serves the same. #566's scaling-draw swap, camera rune guard and output-size
scaling pass stay in vk_dlaa.c. Translator patch 0012 becomes 0015 after dev's 0012-0014.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A connection cut mid-file raises http.client.IncompleteRead, which is not an OSError, so it failed the whole setup at
the DLSS step instead of warning. The download box is ticked by default, so every player without an RTX GPU got a
warning for a download they never chose; that case is a log line now. The game file dialog keeps its "All files"
filter: the check tells an NSZ or XCZ by its content, not its name.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
dlaa and the DLSS modes need nvngx_dlss.dll next to totk.exe, which the release does not ship, so most players
never had DLSS. A new setup step after prepare downloads NVIDIA's own copy from the DLSS SDK repository
(github.com/NVIDIA/DLSS, tag v310.9.1, NVIDIA RTX SDK license) and checks its size and SHA-256.
The window has a tick box for it, on by default and named with its source and licence; the CLI takes --dlss. The
step runs on Windows with an NVIDIA RTX GPU (the host check's Vulkan devices), keeps a DLSS library the player
already put there, and on a failed download warns and lets the game run without DLSS; the next setup run tries
again. Texts in English, Russian, Brazilian Portuguese and Simplified Chinese, docs/installer.md step 7, tests in
tests/installer_test.py (DlssTest, with a file: URL).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Solid scene draws whose position reads an object's own uniform blocks (bones c5, place c7, c6) are drawn a second time
with a vertex shader variant (translator patch 0012) that computes last frame's position from last frame's blocks; the
game writes those into two copies that take turns, and the uniform snapshot keeps both (vk_mvprev.c). The draws are
listed during the replay and recorded through the recorder thread with repeated state skipped. Blocks whose last-frame
copies measured wrong (c9 and c11 of walls and cliffs) keep this frame's; positions without an object block (terrain,
grass) keep the camera's motion, which measured as good.
The camera motion is reprojected in double precision on the CPU: the float inverse made noise on far surfaces and a
still camera that DLSS followed into a wobble. DLSS gets Ultra Performance (a third of the output, also in the F1 menu)
and NVIDIA's recommended preset per mode (K, M for Performance, L for Ultra Performance; TOTK_DLSS_PRESET overrides).
Test switches: TOTK_MV_CHECK measures the motion vectors and DLSS's output against the picture, TOTK_DLAA_TIMES times
DLSS and the motion passes on the GPU. A reactive mask pass for transparents is included but off (TOTK_DLSS_REACTIVE):
its mask covers sky and fog too.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
replay_async was off because no timing on our machines showed a gain. Where the game waits for the GPU
it does: a GTX 1650 in the Depths went from 450-686 frames over 50 ms a minute to 10-13, and the 1% low
from 10-12 to 13-24 fps (#88). An RTX 5090 ran slower with it off (#459). Where the replay or the game
threads bound the frame it is neutral (machines A and B, the Steam Deck in #308), and no crash report
ties to it. MacNativeGames has its translation thread on by default for TotK. It adds up to a frame of
latency; replay_async = 0 turns it off.
The fight scene on an RX 9070 with it on: 60.0 fps mean, 1% low 56.3 in the last window, no frame over
50 ms, 0 validation errors.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- BFE / BFI (patch 0012): a field past bit 31 is cut there, one that starts past it extracts 0 (signed: bit 31
repeated) and inserts nothing, as PTX bfe / bfi. Clamped only where the field is not a constant inside the word.
1.4.3's Depths terrain vertex shaders (42 variants, 134 extracts) compute the position as ((2y + x) * 8) & 0xff,
up to 248; AMD's bitfield instructions take it modulo 32. UltraTears saw the Depths walls and ground missing on
RADV (Steam Deck, RX 570).
- SHF by 0 (patch 0013) gives the unshifted word instead of shifting the other one by 32. Every register SHF of
1.4.3 pairs its word with 0, so 1.4.3 results do not change.
- Function-local variables start at 0 (patch 0014): a register read on a path that never wrote it is no longer an
undefined value Mesa may fold.
All 12,454 1.4.3 shaders translate and pass spirv-val. RX 9070, Depths save: the same picture before and after.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The settings menu offers 1920x1280, 2160x1440, 2256x1504, 2880x1920 and 3000x2000 (the 3:2 screens of Surface and
Framework laptops and tablets), and the setup builds romfs-patches-<w>x<h> for each, so the HUD moves to the top and
bottom edges as it does at 16:10 instead of being stretched.
The check step tells a compressed file by its content (an NSP or XCI holding NCZ files), whatever its name, and
unpacks it into the temp folder with tools/ncz.py before the extraction reads it. Every NCA comes back byte for byte
and is checked against the hash in its name. The file dialog has one filter for all game files, and the form says
that compressed files are unpacked automatically, in every setup language.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
-march=x86-64-v2 does not reach inline assembly, target attributes or intrinsics, which is how kern_ap_stub's YMM
saves got into libtotkrt-v2.a (issue #454). tools/check_isa.py disassembles objects and lists every function with
VEX/EVEX, BMI1/2, LZCNT, MOVBE, ADX, SHA, AES, PCLMUL or RDRAND/RDSEED code or YMM/ZMM/mask registers, except
functions in its DISPATCHED list (sha256.c blocks_ni, behind __builtin_cpu_supports). build_game.py --cpu v2 runs it
over the runtime objects after compiling and before linking or archiving, so make_release.py's v2 library build fails
too. Objects unchanged since the last clean scan are skipped; a full scan of the runtime takes about a second. The
release ships the script, and the installer's runtime fingerprint includes it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The replay marks the start and the end of every render pass and dispatch
(vk_exec.c vk_exec_pass_begin and vk_exec_pass_end, vk_exec_draw.c
op_dispatch and op_dispatch_ind), outside the pass so no pass is broken.
With VK_AMD_buffer_marker the GPU writes the last started and finished mark
into words 2 and 3 of the breadcrumb buffer; with
VK_NV_device_diagnostic_checkpoints the same points are checkpoints, read
with vkGetQueueCheckpointData2NV after the loss. A ring of 4096 records maps
a mark to its submission, index, target formats and size, draw count and the
input ids of the first and last draw's shaders. vk_device_fault_report logs
both marks with their records. The calls go through the recorder thread.
TOTK_PASS_CRUMBS (menu: GPU pass marks): unset, AMD buffer marks only; 1,
also NVIDIA checkpoints, whose extension is enabled only then; 0, none.
The design follows Savage Wind Recomp 3fae40c; the code is ours.
--probe-vkexec gets a pass marks part: pass, dispatch, pass in one
submission; the marks read back after the fence name the last pass and the
report text says so.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
TotK's terrain shaders fill the gradient registers of a texture-with-derivatives instruction per coordinate (du/dx, du/dy, dv/dx, dv/dy), while the translator takes the first two registers as dPdx and the last two as dPdy. The mip level then came from the wrong derivatives and was too sharp on terrain faces turned across the screen, so patches that followed the faces crawled when the camera moved.
After a shader loads, every two-dimensional gradient pair has its inner components swapped by two new composite constructs, so the shader cache is untouched. TOTK_TXD_TRANSPOSE=0 turns it off. The finding is row 2.13 of docs/maxwell.md.
Mods were described only in one dense list in the tester guide. docs/mods.md walks a player through installing a mod,
its order and switches, TKMM merges, code patches and their rebuild, what renaming a mod folder does, the reasons the
Mods tab gives for an unused mod, and what to check when something looks wrong. docs/cheats.md gains a folder example
and the rename note.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The Brazilian Portuguese and Simplified Chinese setup texts now name the 1.1.0 update beside 1.2.1 and 1.4.3, as the
English and Russian ones do.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The rebuild command puts the release folder on sys.path itself: totk.exe starts it in the install folder, and only the
Windows release Python's ._pth names that folder. In a repository the window and its setup run use the Python that
runs the setup, since the tools' embeddable Python there has neither tkinter nor the repository on its path.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The update this port builds for has version value 0x20000, which is Tears of the Kingdom 1.1.0, not 1.1.1: the upstream expected-v1.1.2.json carries 0x40000 and the other expected files step 0x10000 per release, so 0x20000 is 1.1.0 and 0x30000 is 1.1.1. Pull request #220 introduced the wrong label in expected-v1.1.1.json.
Expected hashes, patch sites, boot constants and the RomFS layout are unchanged; only names change:
- tools/data/expected-v1.1.1.json -> expected-v1.1.0.json, display_version 1.1.0
- recomp/patches-v111.txt -> patches-v110.txt
- installer/layout.py SUPPORTED and tools/build_game.py DISPLAY_VERSIONS
- make_romfs_patches.py --version choices and its COMMON table key (the RomFS archive name stays Common.Product.110)
- runtime RT_V111 -> RT_V110, the inits_111 and expect_111 tables, the probe's census folder v110
- README, docs and the installer texts
Co-authored-by: deepseek-v4.1-flash-1017 <noreply@deepseek.com>
A mod that patched the game code (.ips, .pchtxt) stayed off as a whole.
tools/exefs_patches.py now reads IPS, IPS32 and IPSwitch files from each
mod's exefs folder and from atmosphere/exefs_patches/<name>/. It keeps the
ones whose build id is main's and turns their writes into main's code into
one patch set per file. The writes go through the same rules as the code
cheats (tools/cheats.py convert). A file that writes main's data, or that the
rules cannot express, is unsupported as a whole. The setup's lift runs it
after cheats.py and writes build-cache/exefs.status.
runtime/mods.c checks each patch file against that status:
- a built file's set is switched on with its mod;
- a file the build lacks keeps the mod off and starts the rebuild at the next
start, like a code cheat does;
- an unsupported file keeps the mod off, and the Mods tab shows the reason;
- a file for another version is counted and left out.
A mod patch on a site of one of the port's own sets turns that set off while
the mod is on, and the Mods tab names it. The Mods tab also gets the "Rebuild
the game now" row.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Code cheats need the game lifted and compiled with them. Running the setup
again did that, but it lifted and compiled everything and prepared the
shaders again.
- The setup keeps the generated C and the objects in its temp folder (about
1.3 GB). gen.py rewrites only the chunks whose C changed, so Ninja
recompiles only those. One new code cheat on 1.4.3 took a 47 s lift and
19 s of compile and link on 24 cores.
- Prepare's key no longer follows the compile key, only the runtime, the
translator and the game version, so a rebuild keeps the prepared shaders.
- The cleanup writes build-cache/rebuild.txt, the command that opens the
setup window in a new --rebuild mode on the install folder. That mode runs
at once and starts the game again when it succeeds.
- At start, runtime/cheats.c runs that command and exits when an enabled
mod has code cheats the build lacks (plat_spawn, new for Windows and
Linux). build-cache/rebuild.skip keeps a failed or cancelled rebuild from
starting again for the same cheats. "Rebuild the game now" on the Cheats
tab tries again, and [mods] mods_rebuild = off turns the rebuild off.
Scripted runs never start it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The game sizes the buffer for a file from its entry in
System/Resource/ResourceSizeTable.*.rsizetable.zs, so a mod file that grew
could fail to load, and a file a mod added had no entry. With a player mod
enabled, romfs.c restbl_overlay now rebuilds the table at start from:
- the game's table;
- every table of this version that a mod (or the Xbox prompts folder) ships,
where the larger entry wins;
- a mod's table for another version, for that mod's own files only;
- each served mod file: a grown file's entry rises by the growth times its
type's factor, and an added file gets an estimate.
The factors come from the 1.4.3 table measured against its own files (most
types need the size plus at most 8 KiB; bfres 1.5x, bkres 2x, bgyml 3.85x).
The result is served from <profile>/mods/ and takes about 90 ms. A mod's
table no longer counts as a clash with the Xbox prompts. Game entries are
never lowered.
tests/mods_romfs_check.py checks the merged table and now passes probe-fs a
profile of its own.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A cheat file is now part of a mod: mods/<mod>/cheats/<build id>.txt, or the
Atmosphere places inside the mod folder, the file name in any case. The port
ships no cheat files any more (the MyNXCheats files and their notice row are
gone).
- tools/cheats.py --mods builds the code cheats of every mod folder; a
cheat's key hashes "<mod folder>/<name>", so two mods' cheats of one name
keep their own switches.
- The setup reads the cheats from <install>/mods and writes the status to
build-cache/cheats.status.
- runtime/cheats.c reads the files of the enabled mods the port can use, in
the Mods tab order, and the status from build-cache.
- A mod with only cheats is usable. The Cheats tab groups the cheats by mod,
and the Mods tab says how many cheats a mod brought.
- The cheat texts are in all four menu languages.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
make_release.py copied build/xlate as it found it, so a translator change reached no Windows release until someone rebuilt it by hand, and nothing failed when they did not: 0.2.1 and 0.2.8 shipped a translator from before the gather bias while the runtime logged that the bias was on. The Linux release builds its own.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Merged onto the mods folder (task/mods1); the Cheats tab sits after Mods. The cheats' menu text and the cheat file's
place are reworked for the mods folder in the commits that follow.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Each mod is a row with its switch, the files it serves and why it is unused when it is; the top and left face buttons,
Ctrl+Up and Ctrl+Down or the drawn arrows move it. The real-time step, cloth and high-fps fix switches move here from
the Game tab under the port's own patches. The text is in the catalog in English, Russian, Brazilian Portuguese and
Simplified Chinese. The tester guide says where mods go and how TKMM's merged output fits in.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
runtime/cfg.c: the known names take both sides, the per-stick dead zones beside the shadow and
DLSS switches.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
0.2.8 turned the recorder on for every GPU (e38eaa8). The AMD device losses
and hangs came back the same day: a black screen or VK_ERROR_DEVICE_LOST at
startup on RDNA 2, 3 and 4, under AMD's Windows driver and under RADV
(#277, #283), and vk_recorder = 0 still gets past it.
The submission recorder goes back to defaulting off on vendor 0x1002;
TOTK_VK_RECORDER=1 still forces it on for an A/B run, and the log says which
path was taken. The rec_end_cb line now prints only when a part really was
recorded by the recorder, so a log no longer suggests the setting was
bypassed (#315).
The menu side is dev's: it has Simplified Chinese since #243, glyphs included. Simplified Chinese
joins the setup window as a fourth language beside Brazilian Portuguese, in the language switch,
the Windows and locale pick and the launcher; zh-Hans.json gets the nineteen texts added since
(fonts step, CPU level, base game checks, Linux checks) and the {game_exe} placeholder.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The stick reading moved from input_win.c to input.c in dev, so the per-stick dead zone and
input_pad_stick_percent are applied there, under the platform mutex. The rows sit beside the
stick camera speed and walk key rows; pt-BR and zh-Hans get the per-stick texts.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Conflicts with dev resolved by keeping both sides: the fonts step sits between check and extract,
beside the x86-64-v2 check, the pt-BR window language and the xbox-label patches. pt-BR gets the
thirteen firmware and fonts texts.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The sampler keeps both the DLSS mip bias and pull request #321's anisotropic level; DLAA's pass hook runs after the
contact shadows'; the extension arrays grow for NGX's extensions beside HDR's and the calibrated timestamps.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The setup takes an optional --firmware directory (an unpacked firmware or an emulator's Contents/registered) and, for Chinese, Korean and Traditional Chinese, runs the vendored font extractor into <install>/shared-fonts. Without the folder it warns as before and the install still completes. The fonts step is keyed by the language and a cheap firmware fingerprint; the window gets a folder picker and help text, and the command line takes --firmware.
Reverts dafc35f: free_cores = 1 in totk.ini lifts the virtual core gate again, as in 0.2.8. The mode stays as an
experiment, and every place a player meets it says the game crashes: totk.ini.example, a WARNING line in the start log,
and a line under every crash report and FATAL while the gate is off (kern_free_cores_note), so a log from #212 names
the cause at once.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>