1385 Commits
Author SHA1 Message Date
Valiantsin DzerakhandClaude Opus 5.5 cad5b0cc39 Keep the game step at 1/240 s or longer and stop the frame limit at 240
With a step under 1/240 s (ratio below 0.125) 1.4.3 aborts with Havok "Out of memory" from the physics pre-solve
buffer resize (main+0x1261630) within seconds of loading some saves, so a 280 Hz monitor with fps_limit = display
crashed on load (issue #241). frame_step_present floors the averaged step at 1/240 s, and the limit, including the
monitor's rate, stops at 240; a higher fps_limit now logs and gives 240 instead of 60. Limits of 240 and below step
exactly as before.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-31 22:59:02 +03:00
Wellington Dias dbe8daf25c Record the shader variants the driver is building during preparation, skip them after a crash and build the rest again 2026-10-31 16:39:38 -03:00
Wellington Dias c05b083130 Read the extracted game from the work folder in build_game.py and stop when the RomFS is missing 2026-10-31 16:23:28 -03:00
Wellington Dias 1a07a8d46f Fix the sun shadow map at one size, so foliage shadows stop turning blocky and jumping.
Each frame the engine sizes its shadow cascades to the area they cover (main+0xcebdf4 in 1.2.1): from 960 down to about
400 near the ground, so leaf shadows turn coarse, and every change of size moves the shadow texels, so they jump instead
of sliding. The Switch does the same; the community mods force a fixed size. The shadow patch set loads the size from
patch page +92 just before the engine rounds it (main+0xcec22c in 1.2.1, 0x683604 in 1.4.3), and the menu's Shadow map size
(shadow_res in totk.ini) chooses it: game, 512, 768, 1024 (default), 1536 or 2048, with the texture heap grown above 960. A write
watch on the size field found the one writer; with the setting, every shadow map in both versions is the fixed size.
recomp/patches.py lists the patch page words in use.
2026-10-31 15:59:53 -03:00
shenjackanddeepseek-v4.1-flash-1017 73886c4de6 Name 1.1.1 where the docs list the updates
The installer takes the 1.1.1 update since #220, but the README, the installer guide and the tester guide still name 1.2.1
and 1.4.3 alone, so a player on 1.1.1 reads that their update is not supported.

The saves note stays with 1.2.1 and 1.4.3: carrying a save across those two is what has been tested.

Co-authored-by: deepseek-v4.1-flash-1017 <noreply@deepseek.com>
2026-11-01 02:48:57 +08:00
shenjackanddeepseek-v4.1-flash-1017 4223adcb72 Check the Chinese menu catalog in probe-services
The probe asserts a zh-Hans row and the language setting that reaches it, so a catalog that loses its strings or a
language map that stops resolving fails the self test.

Co-authored-by: deepseek-v4.1-flash-1017 <noreply@deepseek.com>
2026-11-01 02:34:52 +08:00
shenjackanddeepseek-v4.1-flash-1017 b94bacd6d5 Name zh-Hans in the MENU_LANG reference
The runtime accepts it; the ini comment and its example line now say so instead of listing en and ru alone.

Co-authored-by: deepseek-v4.1-flash-1017 <noreply@deepseek.com>
2026-11-01 02:34:52 +08:00
shenjackanddeepseek-v4.1-flash-1017 371859c9c1 Merge the Chinese menu glyphs into the font atlas and offer the language
The atlas takes the glyphs of the four CJK rows (ja, ko, zh-Hans, zh-Hant), so MENU_LANG switches while the game runs
without rebuilding it, and the Menu tab lists Simplified Chinese next to English and Russian.

Co-authored-by: deepseek-v4.1-flash-1017 <noreply@deepseek.com>
2026-11-01 02:34:52 +08:00
shenjackanddeepseek-v4.1-flash-1017 ad305912ef Add a Simplified Chinese catalog for the settings menu
Every label and row description in zh-Hans, the catalog rows of the software keyboard included. A menu id without a
Chinese entry falls back to English, so the menu never shows an empty row.

The row also covers the menu language: TOTK_MENU_LANG names a catalog language ("en", "ru", "zh-Hans") and an unset
setting follows the system UI language through the same code map the game language uses ("zh-CN" -> "zh-Hans"). A
language whose row has no menu text still shows English.

Co-authored-by: deepseek-v4.1-flash-1017 <noreply@deepseek.com>
2026-11-01 02:34:52 +08:00
shenjackanddeepseek-v4.1-flash-1017 24c4a27e0b Keep the game's input blocked until the sticks are released
The menu and the software keyboard hold the game's input until everything they saw is released again, and that release
test counted buttons only. With the walk key a keyboard stick moves part way without setting a direction bit, so the
movement went through the moment the menu or the text entry closed; the four axes count as held now.

The walk key's conflict note reads the bindings from a snapshot rather than the live table that the polling thread
rebuilds, and it looks at every key of an action instead of the two its cells show, so a key bound third still names the
action it presses.

Co-authored-by: deepseek-v4.1-flash-1017 <noreply@deepseek.com>
2026-11-01 02:33:39 +08:00
shenjackanddeepseek-v4.1-flash-1017 6e73cd813d Add a walk key that tilts the keyboard's sticks part of the way
A keyboard key is all or nothing, so the game only ever saw a full tilt from it and the keyboard could not walk. While
the key named by walk_key is held, the two keyboard stick samples are scaled to walk_scale, the same thing yuzu and eden
do with their modifier setting; the gamepad's axes and the mouse camera keep their own values.

0.51 rather than half: a sample of 32767 times 0.5 is 16383, one short of the 16384 the game's stick-direction bits
need, so half a tilt would leave the direction clear while the axis moved. No key is on by default, because Shift is B
in the default bindings. The Controls tab lists the keys players pick, steps the scale in hundredths and names the
action a key also presses.

Co-authored-by: deepseek-v4.1-flash-1017 <noreply@deepseek.com>
2026-11-01 02:33:39 +08:00
Valiantsin DzerakhandClaude Opus 5.5 f469c5a774 Raise the version to 0.2.8
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
v0.2.8
2026-10-31 21:17:15 +03:00
Valiantsin DzerakhandClaude Opus 5.5 1d6baac9d2 Merge pull request #233: Add colour profiles and brightness, contrast, saturation and gamma for the picture (Lucas-blip-png)
With TOTK_COLOR_* away from their defaults, present_image runs one compute pass (vk_color.c) that writes the game
picture's crop, adjusted, into an RGBA16F image, which FSR 1 or the blit then scales. At the defaults nothing is
recorded and the present takes the same source, format and rectangle as before. The Screen tab gets a Colour group
(profile and four sliders) after the field-of-view slider of #248; cfg.c and totk.ini.example keep both settings.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-31 21:15:35 +03:00
Valiantsin DzerakhandClaude Opus 5.5 b768567f8d Merge pull request #248: Add a field-of-view setting for 1.4.3 (Lucas-blip-png)
TOTK_FOV (70..140 percent, Screen tab, live) multiplies the camera's vertical view angle through the new patch set fov
at the two camera routines the aspect set patches; page +60 holds the factor. At the default 100 the set is off and the
camera runs the game's own instructions. Both fov sites and the rest of the 1.4.3 table pass patches.check against the
1.4.3 main. 1.2.1 and 1.1.1 have no fov sites and log that the setting does nothing.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-31 21:14:50 +03:00
lunar-meandClaude Opus 5.5 bdad89e0e4 Say "an update file other than the recorded one" in the installer guide (#247)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-31 21:14:41 +03:00
Valiantsin DzerakhandClaude Opus 5.5 db4d04a6e5 Merge pull request #269: Trust the narrowed shader walk when it finds the version's shaders (shenjackyuanjie)
The shader list covers several versions, so most of its rows are absent from any one game and the old test fell back to
reading the whole RomFS on every install: 23,825 of 41,414 rows on 1.2.1, 23,370 on 1.4.3. The fallback now runs only
when the narrowed walk finds next to nothing. On 1.2.1 and 1.4.3 the narrowed walk finds the same shaders as the full
one (27,576 and 28,667, none only in the full walk) in 2 s instead of 51 s.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-31 21:14:34 +03:00
Valiantsin DzerakhandClaude Opus 5.5 407c6915b5 Merge pull request #220: Support the 1.1.1 update as a third game version (shenjackyuanjie)
1.1.1 (update 131072) gets its boot constants, expected hashes, the 60 fps
RomFS overlay, its own code patch table, the scene module slot, loader probe
values and shader census lookup, and the setup accepts it. 1.2.1 and 1.4.3
keep every value: the runtime branches on RT_V111, and an install without a
recorded version still falls back to 1.2.1.

1.1.1 has no aspect patch set; outputs other than 16:9 are scaled to the
window on that version.

Fixes #173

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-31 21:03:32 +03:00
Valiantsin DzerakhandClaude Opus 5.5 f408f26363 Merge task/savecommit: no save or backup for a Commit that changed nothing, a short swap retry (#260, #263)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-31 21:03:24 +03:00
Valiantsin DzerakhandClaude Opus 5.5 d7266301d2 Skip the save and backup of a Commit that changed no byte, and retry the swap briefly
At start the game rewrites its save files with the same bytes and commits
twice; each of those commits made a new committed save and a backup, so one
start pushed every older backup out of save-backup. A Commit whose copy is
the newest save file for file (all hard links, same names) now only retries
a waiting swap.

Renaming the committed save away is retried for 200 ms instead of 2 s: a
failure no longer loses the commit, the next Commit or start retries, and the
game's next change to its save data waits for the commit thread.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-31 21:00:01 +03:00
shenjackanddeepseek-v4.1-flash-1017 55c7411de0 Give the loader probe and the shader census the 1.1.1 values
Three places still assumed the two older versions:

- loader.c compared a 1.1.1 load against 1.2.1's expectations, so --probe-loader failed every time:
  main has 10,668 constructors against 1.2.1's 10,673, and its relocation kinds and totals differ.
  1.1.1 has its own table; subsdk0 and sdk match 1.2.1 down to every count.
- gfx_shader_probe.c looked for build/v121 and build/v143 only, so a census built for 1.1.1 was never
  found and --probe-shader failed on a checkout that has only that version. It looks for the running
  version first, then the other two.
- layout.py put 1.1.1 first in SUPPORTED while steps.py took SUPPORTED[0] as the version of an install
  made before versions were recorded, and such an install is 1.2.1: updating one picked 1.1.1's hashes
  and invalidated its extraction. The tuple keeps 1.2.1 first and the fallback names 1.2.1 outright.

--probe-loader now reports 0 failures on a 1.1.1 install.

Co-authored-by: deepseek-v4.1-flash-1017 <noreply@deepseek.com>
2026-11-01 01:49:40 +08:00
shenjackanddeepseek-v4.1-flash-1017 1ea8813bed Read the scene module of 1.1.1 from its own slot
frame_rate.c finds the player through a global pointer in main's data segment: 1.4.3 keeps it at
main+0x3add238, 1.2.1 at main+0x4728538, and 1.1.1 was borrowing 1.2.1's address, which points at
something else in that build.

The 1.1.1 slot is main+0x4630a68. It was measured from the build itself: one function there walks
the whole chain this code walks (module+0x1e8 scene, the 14-component check at +0x50, component 13
at +0x58, the 0x70 link stride, the name compare) and 181 scene accesses read the same slot, which
makes it the scene module and not one of the single-instance pointers next to it.

Co-authored-by: deepseek-v4.1-flash-1017 <noreply@deepseek.com>
2026-11-01 01:49:39 +08:00
shenjackanddeepseek-v4.1-flash-1017 6b10489e59 Let the setup take the 1.1.1 update
The update's hashes, the patch table and the boot constants are in place; the setup itself still refused the
version. SUPPORTED lists it now, so a player with that update can build.

The labels and messages that named the two supported updates name this one as well, so a player whose update is
1.1.1 is no longer asked for a file they do not have.

Co-authored-by: deepseek-v4.1-flash-1017 <noreply@deepseek.com>
2026-11-01 01:49:39 +08:00
shenjackanddeepseek-v4.1-flash-1017 6d27cf8202 Add the 1.1 code patch table with its spin waits, hair and cloth sites
The 60 fps frame step and ratio, the render resolution, the slow-motion check and the actor animation clock, at the
addresses this build has them: each site was located again by instruction-window matching and its words checked against
the 1.1.1 main. The spin-wait back-edges are the ones that only another thread can end, and the last five sites are the
two inlined clocks of the animation update with the shared call they hand their seconds to.

Co-authored-by: deepseek-v4.1-flash-1017 <noreply@deepseek.com>
2026-11-01 01:49:38 +08:00
shenjackanddeepseek-v4.1-flash-1017 93eeb8b29f Record the expected hashes of the 1.1 update
The ExeFS digests, the merged RomFS size and hash and the six shared fonts, so extract.py can tell this update from any
other. No rewritten_sha256: that field names MacNativeGames' NSO rewriter output, this dump never went through it, and
nothing reads it.

Co-authored-by: deepseek-v4.1-flash-1017 <noreply@deepseek.com>
2026-11-01 01:49:38 +08:00
shenjackanddeepseek-v4.1-flash-1017 d8a211af7f Write the 60 fps RomFS overlay for 1.1
Its Common archive carries the same version number as 1.2.1, so make_romfs_patches.py takes v111 and reads
Common.Product.110, moving the quick-menu background captures to animation frame 0 where the world still draws.

Co-authored-by: deepseek-v4.1-flash-1017 <noreply@deepseek.com>
2026-11-01 01:49:37 +08:00
shenjackanddeepseek-v4.1-flash-1017 2b16578f69 Boot the 1.1 game: its init_array count, checkpoint addresses and version name
The build knows the player's update as v111: DISPLAY_VERSIONS maps it for the version line of the settings menu, and
boot.c carries its two checkpoint return addresses (0xeb8c14 for SetFocusHandlingMode, 0xebcdec for the first
nvnBootstrapLoader call) beside the 1.2.1 and 1.4.3 ones, with init_array counts of 10668, 45 and 116 for main, subsdk0
and sdk. The build guide lists v111 among the versions every step takes.

Co-authored-by: deepseek-v4.1-flash-1017 <noreply@deepseek.com>
2026-11-01 01:49:37 +08:00
shenjackanddeepseek-v4.1-flash-1017 157e619084 Trust the narrowed shader walk when it finds the version's shaders
The fallback read the whole RomFS whenever the shader list named shaders this version does not have. The list covers
several versions, and 1.1.1 holds 17,587 of its 41,414 rows, so every 1.1.1 install read all 304,469 files again:
23827 of the 41414 listed variants are not among the 27576 shaders found, and the narrowed walk costs 2.8 s against
70.6 s for the whole RomFS.

What says the walk fell short is a walk that found next to nothing, which is when the shader directories have moved.
Counting absent list rows does not: for 1.1.1 most of them are absent because the game does not have them.

Co-authored-by: deepseek-v4.1-flash-1017 <noreply@deepseek.com>
2026-11-01 01:49:32 +08:00
Valiantsin DzerakhandClaude Opus 5.5 47485da7d9 Merge pull request #262: Linux fixes (wdiasjunior)
Shader preparation reads MacNativeGames' whole shader list history,
release downloads resume from a .part file, appimagetool and the type-2
runtime come from release tags, and the preparation summary counts the
vertex and fragment variants it skips with the graphics pipeline library
off.

The gettid cache in runtime/platform/linux.c conflicted with the same fix
already on dev (8dd7e20, #236); dev's version is kept.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-31 20:45:06 +03:00
Valiantsin DzerakhandClaude Opus 5.5 1ba0424774 Merge task/savecommit: a save commit that cannot replace the committed save waits in save-tx instead of being lost (#260, #263)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-31 20:35:26 +03:00
Valiantsin DzerakhandClaude Opus 5.5 65fd631fea Keep a save commit that cannot replace the committed save instead of dropping it
Windows refuses to rename a directory while any file under it is open, so a
virus scanner, Explorer's preview or a sync tool holding a file of
save/<uid> made every Commit of a session fail after its 2 s of retries.
The game had its Commit return 0 and showed the new slots from the working
copy, and the next start deleted that copy: hours of play were gone (#260,
#263, log line "cannot move the committed save (error 5)").

A Commit now renames its finished staging copy to save-tx/ready-<n> before it
touches the committed save. That rename is the commit point. When the swap
into save/<uid> fails, ready-<n> stays the newest save: later working copies
and the next start are made from it, and the next Commit or start retries the
swap. The failure log names the programs holding files in the committed save
(Restart Manager) or says the folder refuses the rename (no DELETE access).
--probe-fs holds a committed file open across two Commits and restarts.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-31 20:20:50 +03:00
Valiantsin DzerakhandClaude Opus 5.5 8a4ab178ec Raise the version to 0.2.5
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-31 20:04:32 +03:00
Valiantsin DzerakhandClaude Opus 5.5 8041ded696 Merge task/crashdiag: menu without the core gate, system font check, abort reports, command buffer diagnostics (#212, #176, #94)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-31 19:57:26 +03:00
Valiantsin DzerakhandClaude Opus 5.5 e84c61f16d Say why a command recording buffer could not grow
A player's game ended with "nvn: command recording buffer of 1048576 bytes" during a dive into the Depths (#94):
realloc refused 1 MiB with commit to spare. That leaves a buffer pointer that is no block of ours, from a command
buffer object the game overwrote or initialized again without Finalize. The fatal line now gives the object, its
magic, the old buffer and its size, the bytes used and asked, and Initialize logs (eight times at most) when it
replaces an object that was not finalized.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-31 19:57:09 +03:00
Valiantsin DzerakhandClaude Opus 5.5 e451ce6a0d Report abort and invalid C runtime parameters as crashes on Windows
abort() (a failed assert, std::terminate) and an invalid parameter to a C runtime function raise no exception: the
C runtime ends the process with 0xc0000409 and the log stops without a FATAL line. A SIGABRT handler and an
invalid-parameter handler now send both to the crash hook with the host return addresses in totk.exe, and the
report lists them as totk+offset frames after the guest registers. Checked with win32.c in a test program: assert,
abort and fclose(NULL) each give one report whose innermost frame is the failing call.

lifted_function no longer names the nearest lifted function for an address above the lifted code, so a host frame
of the runtime is not reported as game code.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-31 19:57:09 +03:00
Valiantsin DzerakhandClaude Opus 5.5 637b98dea7 Check a system font before the menu's font atlas takes it
The menu merges the Japanese, Korean and Chinese language names from a system font. Dear ImGui asserts on a font
stb_truetype cannot parse, and the assert ends the process with 0xc0000409 (exit code 127 in a shell) and no line in
the log, a few seconds after the window opens: the report of an AMD laptop on Windows 11 Home Single Language (#176)
matches that, and a test build of our ImGui with a damaged font exits the same way. A cut-off font passes the parse
and faults later, when the glyphs are made, because stb_truetype reads tables without bounds checks.

fonts_up now runs each system font through a private copy of stb_truetype whose asserts only mark the font bad, after
checking that every table of its directory lies inside the file, and leaves out a font that fails with a log line.
--probe-window checks that the test takes the embedded Roboto and refuses a cut-off copy, one without a character
map and one with an unknown tag.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-31 19:56:57 +03:00
Valiantsin DzerakhandClaude Opus 5.5 a554a7c13c Take the virtual core gate out of the settings menu
The row was labelled "Virtual core gate" with the text "Lets only one thread run per virtual core, as on the
console", but switching it on wrote free_cores = 1, which lifts the gate. Without the gate guest threads run in
parallel on one virtual core, and TotK keeps its per-core data without locks: an Arc B580 player who turned it on
crashed at every start with "CallCommands: command handle ... is not a live recording" (#212). free_cores stays an
ini key for experiments; the log now says when it is set, and totk.ini.example warns that the game can crash.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-31 19:56:48 +03:00
Valiantsin DzerakhandClaude Opus 5.5 c346c7fde2 Merge task/camlabel: write back GPU renders into textures that share their page (#13, #63, #118, #252)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-31 19:55:56 +03:00
Valiantsin DzerakhandClaude Opus 5.5 de3ed42a4c Write back GPU renders into textures that share their page when the texture's own bytes are unchanged (#13, #63, #118, #252)
The page write tracking dropped the render of any texture smaller than a page whenever the CPU wrote another
part of that page. The Purah Pad camera renders the ID, screen rect and depth of what is under the reticle into
4x4 targets in two slots and reads them a frame later. On 1.4.3 one slot's rect target shares its page with data
the CPU writes every frame, so the game read a zero rect every other frame and the target's name label blinked; on
1.2.1 the targets of both slots were dropped and nothing was recognised. A texture that fills no page now keeps a
copy of its storage bytes from the GPU write, and a page write counts only when those bytes changed.
TOTK_WB_EXACT=0 restores the old decision.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-31 19:55:56 +03:00
DizzyMan1337andClaude Opus 5.5 54337ce6a4 Let DLAA copy the camera block out of the uniform snapshot buffer
mv_pass copies the scene's camera block from the snapshot arena with vkCmdCopyBuffer, which needs
VK_BUFFER_USAGE_TRANSFER_SRC_BIT on the source (VUID-vkCmdCopyBuffer-srcBuffer-00118). Reported by
Lucas-blip-png in #228 and SchulzCode in #253.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-01 00:16:47 +08:00
Valiantsin DzerakhandClaude Opus 5.5 da6cd4009e List the bundled libXft and libXrender in the third-party notices
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-31 19:05:30 +03:00
Valiantsin Dzerakh 14cc5f1d9b Bundle libXft and libXrender with the Xft Tk so the installer opens on hosts without them 2026-10-31 18:56:40 +03:00
Valiantsin Dzerakh e4eecef0f7 Pick the installer log font next to the UI font 2026-10-31 18:52:38 +03:00
Valiantsin Dzerakh ec34382f0a Use the platform UI font for the idle step rows in the installer 2026-10-31 18:52:38 +03:00
Valiantsin Dzerakh f8d8a4c553 Merge pull request #230: Restore Russian text in the Linux installer with Xft (lcd1232) 2026-10-31 18:52:09 +03:00
Wellington Dias 9865e7ab9a Count the vertex and fragment variants preparation skips with the graphics pipeline library off, and say so in its summary instead of reporting them as built. 2026-10-31 12:48:36 -03:00
Valiantsin Dzerakh e34909b40d Merge pull request #229: Find the shaders by their directories instead of reading the whole RomFS (shenjackyuanjie) 2026-10-31 18:48:09 +03:00
Wellington Dias 02d79bd378 Read a thread's id once on Linux, where gettid is a system call and every recursive mutex lock asked for it. 2026-10-31 12:41:49 -03:00
Wellington Dias 42e5c98395 Pin appimagetool 1.9.1 and the type-2 runtime 20251108 from their release tags, so a new upload under continuous no longer stops the AppImage build. 2026-10-31 12:41:49 -03:00
Valiantsin DzerakhandClaude Opus 5.5 e38eaa8224 Turn the recorder thread back on by default on AMD GPUs
The AMD device losses with the recorder (#5, #40) stopped after objects
were kept until their submission completed and the recorder began ending
its own parts; RDNA 1 and 2 cards in #5 ran with vk_recorder on and off.
vk_recorder = 0 still turns it off.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-31 18:41:38 +03:00
Wellington Dias 522c69e951 Download a cached release file again when its hash no longer matches its pin, in the same run, instead of stopping. 2026-10-31 12:40:55 -03:00