# nx2navi.ini: settings for nx2navi.exe. Copy this file to nx2navi.ini next to nx2navi.exe and remove the '#' of a line to change it.
# Each line shows the default; on a Steam Deck the device profile ([display]) changes a few of them. A NX2NAVI_*
# environment variable of the same name (NX2NAVI_FPS_LIMIT for fps_limit) overrides the file. Keys are case-insensitive;
# '#' and ';' start a comment; [sections] only group lines.
# Only the keys in this file are settings. A line with any other key does nothing: the game's log names it at the
# start ("no setting of this version") and the settings menu (F1) shows it in its bottom row.
# Updates and rebuilds of the game never rewrite nx2navi.ini; the settings menu changes only the lines of the settings
# it changes and keeps every other line as it is.

[frame rate]
# Frame rate limit: 30, 40, 45, 60, 90, 120, 144, any whole rate from 30 to 240, display (the refresh rate of the
# window's monitor, at most 240) or unlimited (as fast as the PC goes, at most 240 frames of the game a second). A
# higher rate gives 240: far above it the game's own code fails (its quick menu crashed at 700 fps). Above 60
# is experimental and needs realtime = 1. Changes while the game runs take effect at once.
# With mailbox or fifo presentation (vk_present, the default) a frame reaches the screen at a vertical blank, so a limit
# that does not divide the monitor's refresh rate keeps each frame on screen for a different number of refreshes. On a
# 320 Hz monitor 80, 64, 40 and 32 divide it and 90 does not: 90 fps alternates 3 and 4 refreshes per frame, which reads
# as judder while the frame counter still shows 90. Likewise on 240 Hz 60, 48 and 40 divide it and on 144 Hz 72, 48 and
# 36 do, while 60 does not. vk_present = immediate shows a frame as soon as it is presented, without this quantisation,
# at the cost of tearing. A monitor running with variable refresh (FreeSync, G-Sync) follows the frames instead, so
# there any limit inside its range is even.
# With frame generation on (framegen, [display]) the limit still counts the frames the game draws, and the generated
# frames come on top: 60 at 4x is 60 drawn and 240 on screen. Only display counts the frames on screen, so there the
# game draws the monitor's rate divided by the multiplier.
# fps_limit = 60
# Where RivaTuner Statistics Server is installed and running, the limit goes into its profile for the game's
# executable and RTSS does the limiting instead of the game; closing the game puts back what the profile held.
# Changing that profile needs the game run as administrator (RTSS keeps its profiles under Program Files); without
# it, or with frame generation on, the game limits by itself, and also when RTSS's own frame counter shows the game
# well past the limit (RTSS then hooks the game without holding it back). rtss = 0 always leaves it to the game.
# rtss = 1
# 1: asks RTSS's EnumProfiles whether the game's profile exists, instead of looking for its file. EnumProfiles crashes
# the game (RTSS keeps a pointer of the game's in 32 bits); only for testing an RTSS that fixed it.
# rtss_enum = 0
# Older key, read only while fps_limit is unset: fps60 = 0 means a limit of 30.
# fps60 = 1
# Real-time frame step: game speed follows the clock instead of the frame count.
# realtime = 1
# Cloth simulation on the real-time clock. Needs realtime = 1.
# cloth_realtime = 1
# Fix for physics that bounce at high frame rates.
# highfps_fix = 1
# Above 60 fps the game's physics solved contacts and friction in fewer steps a second than at 60, and in none at all
# in a frame shorter than 1/240 s, so things standing on moving platforms were left behind. 1 rounds the step count
# up instead; 0 is the game's own count.
# physics_steps = 1
# The number in the window that shows a pick-up (a rupee) counts up by whole steps a frame, and above 60 fps the game
# rounded that step down to nothing: the window stayed until the area was left and its counting sound went on. 1 lets
# it count at least one a frame; 0 is the game's own count.
# pickup_count = 1
# The step the game is given each frame: 1 keeps it steady (the same value from frame to frame while the frame rate
# holds, exactly one 60 Hz frame at a 60 fps limit) and makes up for a late frame over the next second; 0 is the mean
# of the last five frame times, which changes with every frame.
# step_steady = 1

[performance]
# The keys of this section and of [diagnostics] are for testing and troubleshooting, and take effect at the next start.
# The Advanced tab of the settings menu (F1) shows all of them except zstd_native and zstd_check.

# Scheduling.
# A thread waiting for a virtual core interrupts the owner when it runs game code: 1 = on, 2 = priority inheritance
# without the interrupt, 0 = off (the old scheduling).
# async_preempt = 1
# Replay NVN submissions on their own thread, at most one frame behind the game, 0 or 1. Default: off. Helps where the
# game waits for the GPU and in heavy scenes (15 to 18 percent more frames per second in a village); adds up to a frame
# of latency, and can show a dark or flickering frame now and then: the other thread reads vertex and buffer memory
# the game has sometimes rewritten by then.
# replay_async = 0
# Record the large groups of draws of a submission on this many replay worker threads beside the submitting one, 0 to
# 16. Default: 3 (Hyrule Field standing on an RX 7900 XT: 51.7 instead of 43.6 fps). 0: the replay records everything
# on one thread, as before; try it if a picture looks wrong. With 1 a worker records each long run of draws while the
# submitting thread waits for it: no faster, it checks that the pictures stay the same.
# replay_workers = 3
# EXPERIMENTAL, THE GAME CRASHES WITH 1. 1 lifts the virtual core gate: guest threads run without the console's limit
# of one thread per core, and the game keeps per-core data without locks. With 1 the game crashes or hangs,
# usually at start. Keep 0 to play.
# free_cores = 0
# Windows timer resolution of 1 ms for steadier frame pacing (0 = the Windows default).
# timer_1ms = 1
# The thread that records the GPU work (the first to spend half a second doing it, not the one that submits first) keeps
# the highest scheduling priority (recorder_prio), and can get a core of its own that the process's other threads stay
# off (recorder_core: 0 = never, the default; auto = with 12 or more logical processors, unless the game runs on the
# cache die below; 1 always). That is a fast core, with both of its logical processors: on Windows the highest-numbered
# core of the fastest efficiency class; on Linux the performance core that clocks highest on a hybrid CPU (else the core
# of the highest capacity or clock), and none when fewer than four cores are that fast. The log names the thread and
# its logical processors. Default 0: on an i9-14900KF the reserved core left the woken replay thread unscheduled for
# about 200 ms roughly once a second.
# recorder_prio = 1
# recorder_core = 0
# Submissions handed to the GPU queue together in one call, 1 to 16 (1 = each one at once).
# submit_batch = 4
# Mesa's Vulkan drivers (RADV, ANV, NVK): every submission to the GPU queue goes to a thread of the queue's own, so its
# kernel call leaves the thread that prepares the frames, 0 or 1. Default: on (Linux), off (Windows). 0 leaves the
# driver's own choice. Unused when MESA_VK_ENABLE_SUBMIT_THREAD is set.
# submit_thread = 1
# Occlusion query results are read back from the video card in batches, 0 or 1. Default: on. 0 reads each query with a
# call of its own, as before, which on AMD's Windows driver held every frame's "GPU done" back by about 4 ms.
# oq_batch = 1
# How many milliseconds a submit call waits for the game's own signal that the frame's data is written, when the
# frame's commands begin with a wait for that signal (0 to 100; 0: the frame is set aside at once and drawn at its
# present, as before).
# gate_wait_ms = 10
# Dynamic resolution, 0 or 1, applied at once: with a DLSS upscaling mode, the scene resolution follows the GPU time
# per frame toward dyn_res_target milliseconds (1 to 100), one DLSS step at a time, each held for dyn_res_step frames
# (8 to 600), from the DLSS mode set down to performance. Default: off.
# dyn_res = 0
# dyn_res_target = 16.7
# dyn_res_step = 45
# Processors. On a processor whose dies have L3 caches of different sizes (AMD's two-die X3D processors, with 3D
# V-Cache on one die) the game runs on the die with the larger cache when that die has 8 cores or more: 1 = on,
# 0 = Windows places the threads.
# cache_die = 1
# One logical processor of each physical core, so no two busy threads share a core: 1 = on, 0 = all of them.
# one_per_core = 0

# Shaders and pipelines.
# Draws whose pipeline is still building: default waits on a reserved worker; draw is the previous scheduling.
# pipe_wait = default
# Which draws are dropped instead of waiting for their pipeline: strict (only draws into per-frame passes), any (also
# draws into passes whose target a recent frame rendered; full-screen draws wait, from pipe_frame_wait_ms) or fullscreen
# (those and full-screen draws: no wait, but a lighting or fog pass that is dropped shows as a black or unlit frame).
# pipe_skip = any
# How many milliseconds one frame may wait, in all, for the shaders and pipelines of its full-screen draws and of
# draws into pictures drawn for the first time (0 to 500). What is not ready by then is left out of that frame and
# built in the background.
# pipe_frame_wait_ms = 50
# A picture the game has drawn in many frames counts as drawn every frame after a pause too (a loading screen, a
# menu), so the first frame of an area leaves a draw out while its pipeline is built instead of stopping for it.
# 0: as up to 0.7.1, such a frame waits for each new pipeline in turn.
# pipe_skip_steady = 1
# How many milliseconds the draws of one frame may wait, in all, for a missing pipeline to be linked from the shader
# libraries in the graphics driver's cache before they are dropped (0 to 100; 0: dropped at once, as before).
# pipe_link_ms = 6
# How many milliseconds one missing pipeline is waited for, counted from the first draw that needed it (1 to 50).
# With its shader libraries in the driver's cache a pipeline takes 2 to 8 ms; what is not ready by then is left out
# of the frames until it is. A longer wait leaves fewer draws out and makes those frames longer.
# pipe_link_draw_ms = 3
# When a draw finds its pipeline missing, the rest of the commands the game submitted with it are read ahead and the
# shader libraries of every other pipeline they will need are built at once on background threads, so the later draws
# find theirs ready instead of waiting one after another or being left out of a frame. 0: off, as before.
# pipe_ahead = 1
# A pipeline whose fragment shader reads the picture it draws into (soft particles, water, some full-screen passes) is
# linked from the shader libraries like any other. 0: as before, such a draw waits for the whole compile.
# gpl_feedback = 1
# The threads that build shader libraries also link the pipelines of draws that were left out of a frame, several at
# once, instead of one thread linking them one after another while the others build optimised pipelines. 0: as before.
# pipe_fast_workers = 1
# The graphics driver's cache of shader libraries is checked at every start and, when it lost them (a driver update,
# a cache the driver emptied), filled again in the background while the game runs. 0: no check; force: at every start.
# After two cold starts in a row warming stops being tried on that driver; once its cache can hold the libraries (on
# NVIDIA under Windows, Shader Cache Size Unlimited in the NVIDIA app), start once with force, or delete
# shader-warm.txt, to warm again.
# lib_warm = 1
# Draws into a picture drawn for the first time (a lookup table, a new area's target) wait for a pipeline built for
# the first time from pipe_frame_wait_ms; one left over is drawn the next frame its picture is drawn. 0: as up to
# 0.7.1, each such draw waits up to oneshot_wait_ms milliseconds, which stops the game for as long as the pipelines
# of a new area take to build.
# oneshot_budget = 1
# oneshot_wait_ms = 30000
# Build shader stage libraries in the background ahead of time, and the threads for it (1 to 16; default 2). More
# threads mostly wait on the graphics driver's lock: 16 on NVIDIA spent about 8 cores spinning in the driver.
# lib_prebuild = 1
# lib_prebuild_threads = 2
# How many compute pipelines the start builds and keeps: those of the compute shaders the game used most recently.
# Every other known one is built into the pipeline cache and released again, and made from the cache when the game
# first uses it (well under a millisecond). Very large compute shaders the game has never used are not built at all.
# all: the start builds and keeps the pipeline of every known compute shader, as versions up to 0.6.1 did, which held
# more than 2 GB of video memory. 0: none are kept.
# compute_keep = 512
# 1: very large compute shaders that no cache holds are built on one background thread after the start. 0: each is
# built when the game first uses it, which can hold that frame for a second or two.
# lib_prebuild_big_bg = 1
# The commit headroom in MB those background builds wait for (0: no check). On Linux the headroom is the smaller of
# the room left under the cgroup's memory limit (memory.max, memory.high) and the memory available on the machine.
# big_build_free_mb = 8192
# How far those background builds may raise the process commit in MB before the rest wait for their first use instead
# (0: no limit). Each large shader costs the driver 1-4 GB; on a 31 GB machine four of them took the process from 15 to
# 29 GB and the game stuttered, so raise this only on a machine with room to spare.
# big_build_max_mb = 4096
# Memory kept free of pipelines built ahead of their first use (at the start and in the background during play), in MB
# of commit headroom: RAM plus page file not yet promised to a program. Such a build starts only while this much stays
# free beyond what the builds already running are counted for; otherwise it waits or is left to the moment the game
# first needs the pipeline, which can hold a frame once but does not push a full machine into paging. A pipeline a
# draw is waiting for is always built. 0: no check. Default: 4096.
# build_free_mb = 4096
# What each such build is counted for while it runs, in MB: the working memory of the video driver's shader compiler.
# A very large compute shader is counted for 4096. Default: 64.
# build_reserve_mb = 64
# 1: the log says how many pipelines came from pipeline-cache.bin; 2 also logs every compute pipeline built. 0: off.
# pipe_feedback = 1
# pipeline-cache.bin larger than this (MB) starts empty and is rewritten (0: no limit).
# pipeline_cache_max_mb = 512
# 1: the pipelines a run compiles go into pipeline-cache-play.bin, a small file of its own, so pipeline-cache.bin
# (hundreds of MB) is not written again while you play; the small file is folded into the large one at a start once
# it has grown. 0: one file, written whole whenever a run made a new pipeline.
# pipeline_cache_play = 1
# The size (MB) at which pipeline-cache-play.bin is folded into pipeline-cache.bin at a start.
# pipeline_cache_play_mb = 64
# 1: the pipelines FSR 3.1 makes for upscaling and frame generation are kept in the pipeline cache files, so a later
# start does not compile them again. 0: FSR 3.1 creates them without those files, as the graphics driver alone keeps
# them.
# ffx_pipe_cache = 1
# 1: where the driver has VK_KHR_pipeline_binary, compiled pipelines are kept in pipeline-binaries.bin under the
# runtime's own keys, so a renamed or updated exe starts without compiling them again (pipeline-cache.bin is then
# not used). Default 0 (pipeline-cache.bin only) while it is being tested.
# pipe_binaries = 0
# pipeline-binaries.bin larger than this (MB) starts empty and is rewritten (0: no limit).
# pipe_binaries_max_mb = 1024
# Build known pipelines ahead of time.
# pipe_prebuild = 1
# Count the shared pipeline list that comes with the release (seed-pipeline-keys.bin) as known pipelines, so a new
# install builds them at the start instead of at their first draw. 0: only the pipelines this install drew.
# pipe_seed = 1
# 1: the start waits until every known pipeline is built. Default 0: with the graphics pipeline library the start
# builds them for at most 20 seconds and the rest in the background while you play.
# prewarm_full = 0
# Threads that translate known shaders at start (1 to 64; default: half the CPU threads, at most 16).
# prewarm_xlate = 8
# After an update that brings a new shader translator, 1 keeps using the shader cache's translations from the earlier
# translator, so the update adds no shader translation at start or in play: while you play, a background pass at the
# lowest priority translates them again with the new translator for the next start, then deletes the old ones. A shader
# the new translator cannot translate keeps its old translation. 0: shaders the new translator has not translated yet
# are translated before they are used (at start or when the game first draws them), as earlier versions did.
# shader_carry = 1
# Vulkan graphics pipeline library, 0 or 1. Default: on, except with AMD's Windows driver for RX 6000 and older cards
# (RX 400 to RX 6000), where the game ended at every enemy killed with it on. There it is off: a pipeline needed for
# the first time is built whole while its draw is left out for some frames, and the start waits until every pipeline
# of earlier sessions is built. gpl = 1 turns it on again.
# gpl = 1

# Textures and memory.
# ASTC texture decoding: native (the GPU's own ASTC support), gpu (a compute shader), cpu, or auto (native when the GPU
# has it, gpu otherwise).
# astc = auto
# With GPU ASTC decoding, 1 requests BC3 compression. Slight blocking is possible. 0 keeps RGBA8, which uses about
# four times as much memory for these textures and can exhaust memory while loading.
# Leave this setting commented for automatic selection. The renderer requests BC3 when the largest device-local GPU
# heap is under 20 GiB or the system reports less than 47 GiB of RAM (allowing for a nominal 48 GiB PC). If the RAM
# size is unknown, only the GPU heap decides. Look for "NX2NAVI_ASTC_BC unset" in the game log to see the decision.
# astc_bc = 1
# With BC3 on: a texture none of whose texels is see-through is BC1, half the size of BC3 with the same texels (each
# upload checks the blocks as it copies them). 0: every ASTC texture BC3.
# astc_bc1 = 1
# Game memory shared with the GPU: auto (mirror on a discrete GPU, import on an integrated one), import (host memory
# imported into Vulkan), mirror (pools used as buffers get a copy in video memory) or verify (mirror plus a check of
# every copy; slow). On a discrete GPU import can be far slower than mirror, not slightly slower: the GPU reads the
# game's pools over PCIe. A 5070 Ti Laptop went from 6 ms to 135-230 ms of GPU time per frame with import (5 fps on the
# title screen, bound: gpu, every line of the perf window waiting on a fence). The log names the mode in use: "mem:
# mode import" or "mem: mode mirror".
# memory = auto
# A new texture created over a live texture's memory starts with that content instead of empty.
# tex_alias = 1
# A new depth image is cleared at the place in the frame where the game makes it (1, default), not ahead of the
# frame's other work: made ahead, the clear could wipe depth that draws of the same frame still read, and the sun's
# shadow went fully dark for one frame a few times a minute where the game resizes its shadow maps (a town gate).
# 0: cleared ahead, as before. depth_alias_log = 1 logs each new depth image made over a live texture, with its frame.
# depth_init_inline = 1
# Small texture uploads stage in ranges of a few kept buffers instead of a new buffer each (0 = a buffer each).
# stg_arena = 1
# With astc = gpu: the compute shader reads the blocks in the game's tiled layout, so the CPU only copies the
# texture (0 = the CPU untiles the blocks first).
# astc_gpu_deswizzle = 1
# With astc = gpu: 1 keeps the decoder's scratch buffer from one upload batch to the next instead of making it again for
# each, to shorten the hitches when an area loads (0 = one per batch).
# astc_scratch_keep = 1
# With astc = gpu: 1 stages a texture's ASTC blocks in the shared staging ranges instead of a buffer of their own
# (0 = a buffer each).
# astc_arena = 1
# A render target that needs an image of its own takes the image of a released one of the same size and format
# instead of making a new one (0 = always a new image).
# tex_reuse = 1
# A texture the game does not fill from the CPU itself, made over memory the CPU writes (a second description of a
# texture's bytes), starts with those bytes and takes the game's later writes to them until something is drawn into
# it (1, default). 0: it starts empty, as before, and what is drawn with it can come out flat and pale. 2: it gets
# the bytes once and no later write. Costs an upload for each such texture; a render target the game makes over
# CPU memory starts with the last picture left there instead of an empty one.
# tex_split_fill = 1
# 1: shaders mark the textures they sample, so that while video memory or the system's memory runs low, textures no
# shader sampled for a while go, even those the game keeps registered (the areas it has loaded); one sampled again is
# made again within a few frames. The first start after changing it builds the pipelines again. 0: such textures stay.
# 2 (diagnosis): the same marks, and such textures give way to stand-ins in the texture pool, but none is freed.
# tex_disuse = 1
# Whether such textures really go under pressure. Off by default: after such evictions the ground of some areas was
# drawn flat and see-through and stayed so. 1 turns it on (it frees more memory when memory runs short).
# tex_disuse_evict = 0
# Percent of the video memory in use above which textures give way: first those the game released, then those no
# shader sampled for a while (tex_disuse); each is made again when used. The default lies under the 87 percent past
# which game memory gets no copy in video memory (memory = mirror), which costs far more frame time than a texture
# uploaded again. 0: textures never give way for video memory.
# tex_vram_pressure = 80
# Low-end GPU mode (experimental; also in the settings menu's Game tab). For older graphics cards, and ones short of
# video memory: 1 turns on the settings written for them and measured on a Radeon RX 580, for every key below that
# this file does not set itself. Mesh shaders run as compute where the card has none (mesh_emulation), video memory
# stays within what the driver reports free (vram_heap_limit = auto, vram_reserve_mb = 512), large texture upload
# batches go to the GPU early (upload_batch_mb = 256), game memory is copied for the GPU while a frame is recorded and
# the game waits less for the GPU (mirror_stage = 1, submit_finish_gpu = 0), frames are prepared on a thread of their
# own (replay_async = 1, which can show a dark or flickering frame now and then), pools the game refills every frame
# stay in system memory (mirror_hot = 25), and a texture read after another image was drawn over its memory gets
# memory of its own (tex_clobber = 1). 0, the default: each of these stays as it was in 0.9.1, and none of these keys
# is read. Read at start.
# low_end_gpu = 0
# Textures the game released that are kept as they are (1 to 16384); past this count only those that lose nothing
# when dropped go. Default: one per 2 MB of the video memory budget, 2048 to 8192. Read at start.
# tex_released_max = 4096
# Megabytes of this process's commit above which the log says so. Windows counts video memory against the commit;
# with a discrete GPU the video memory in use is left out of the comparison, since it is not in RAM. Textures no longer
# give way at this bound: dropping one lowers both sides of the comparison alike, so nothing they could do brought the
# process under it, and the ground flickered white while the system had commit to spare. They give way only while the
# system's commit runs low. Default: the installed memory, at least 24576. 0: no bound. On Linux it counts the
# process's resident anonymous memory, which stays under the installed memory, so there only a smaller value holds.
# tex_proc_commit_mb = 32768
# 1: a texture the game releases or takes back, with nothing else changed at its address, puts the texture table's
# slots there back in a state they were in before, and they are not all looked up again (large towns release and take
# back thousands a second). 0: every release and take-back counts as a new state.
# tex_gen_undo = 1
# Host zstd decoder in place of the game's (0 = the game's), and a check that decodes every call both ways and logs
# the differences (1 = on; slow). Ocarina of Time 1.1.1 decodes on the host too since a check run over a village and a
# field found no difference in 3,072 files; zstd_native = 0 keeps the game's own decoder there.
# zstd_native = 1
# zstd_check = 0

# Render fixes.
# Wait for the game to finish rewriting level-of-detail data before the GPU reads it: stops objects flashing in the
# wrong place and a flat far field for a frame.
# lod_guard = 1
# Fill the driver-reserved constant bank c0 (storage sizes, dispatch size, point size, base instance) instead of zeros.
# fix_c0 = 1
# New GPU memory pools start zero-filled, as Switch memory does.
# fix_zero = 1
# Build the water's screen coordinates per pixel and keep its absorption distance at 0 or more: without it water is
# drawn as a rectangle over solid ground at the Hyrule Castle moat and the Castle Town canal from some camera
# positions, and can turn into a red beam there. 0: the water's shaders as they are.
# water_projective = 1
# Lines are drawn as wide as the game sets them (it draws two a frame 4.5 pixels wide into a small mask). 0: every
# line 1 pixel wide, as up to 0.9.1.
# line_width = 1
# Optimisations, 0 turns each off: skip bindings that did not change, remember shader lookups, reuse the previous
# draw's pipeline when its state repeats, remember texture descriptors between draws, update descriptors in bulk,
# skip push constants that did not change.
# bind_skip = 1
# shader_memo = 1
# 1 flushes every shader cache file to the disk before it is used, as earlier versions did; 0 (default) flushes only
# the stored game code, which makes shader preparation about twice as fast (a torn file is translated again).
# shader_cache_flush = 0
# pipe_memo = 1
# tex_memo = 1
# desc_template = 1
# push_skip = 1
# More optimisations, 0 turns each off: a draw resolves again only the descriptors whose bindings changed, and records
# only the dynamic states whose values changed since the last pipeline.
# bind_reuse = 1
# dyn_delta = 1
# Submit GPU work in parts every N draws so the GPU starts earlier (0 to 1000; 0 = off), each part at least split_us
# microseconds long (0 to 20000). split_in_pass = 1 also splits inside a long pass (it ends there and begins again in the
# next part); 0 splits between passes only. split_idle = 1 sends a submission's first part at the first pass boundary
# while the GPU has nothing else to run; it made the foliage flicker in Kokiri Forest, so it is off by default.
# split_draws = 64
# split_us = 1000
# split_in_pass = 1
# split_idle = 0

# Vulkan.
# Vulkan validation layer: checks every call and slows the game down a lot. Needs the layer installed (Vulkan SDK).
# vk_validation = 0
# A thread of its own makes the Vulkan calls of each frame, 0 or 1. Default: on, except on AMD GPUs.
# vk_recorder = 1
# Dynamic vertex input, 0 or 1. Default: on, except on Intel's older integrated GPU drivers (Vulkan 1.3.215 and older).
# vk_dyn_vertex = 1
# Dynamic alpha to coverage, 0 or 1. Default: on, except on AMD's driver for RDNA 2 and older (RX 400 to RX 6000),
# which applies alpha to coverage while the dynamic state says off and draws the characters black.
# vk_dyn_a2c = 1
# Pin the shader subgroup size to 32 when the GPU allows it.
# vk_subgroup_pin = 1
# Keep a shader's threads together after divergent code where the game's shaders expect it (maximal reconvergence,
# when the driver supports it). 0 turns it off; leaves then go missing from the foliage's colour pass.
# vk_reconverge = 1
# Ask the GPU driver to keep NaN, infinity and signed zero in the game's shader maths as the Switch does, 0 or 1.
# Without it a driver may assume they never occur. The AMD symptoms it is for: Navi drawn without her orb (only her
# sparkles), black and white blocks on her and on Link, shadows blown up. Default: on, except on NVIDIA's driver,
# which keeps them anyway; 1 also asks NVIDIA's driver. The first start after a change compiles the pipelines again.
# vk_nan_preserve = 1
# Test for NaN in the game's shaders on the bits of the number instead of as a float, 0 or 1, so that no shader
# compiler can leave a test out. Default: on with AMD's driver for RX 5000 and RX 6000 cards (the one whose shader
# compiler reads "AMD proprietary shader compiler" in the log), where the black and white blocks were reported; off
# elsewhere. The first start after a change compiles the pipelines again.
# vk_nan_bits = 1

[display]
# Device profile: auto (the Steam Deck when this is one), desktop or deck. On a Steam Deck the keys this file does not
# set default to fullscreen = 1, window_size = 1280x800, output_res = 1280x800, render_scale = game, vk_present = fifo,
# fps_limit = 40, cores = 4 (the game's six cores share the Deck's four), astc_bc = 1 and spin_handoff_us = 30
# (starting points, not measured yet). Read at start.
# device_profile = auto
# Presentation mode: mailbox (default when the driver offers it, vsync otherwise), fifo (vsync) or immediate. The menu's
# VSync switch sets mailbox when on and immediate when off.
# vk_present = mailbox
# Window client area in screen pixels (640x360 to 7680x4320), and borderless fullscreen. Windows display scaling does
# not apply: at 150% scale, 1920x1080 covers the area 1280x720 covers at 100%. F11 switches fullscreen and saves it
# here. Changes while the game runs take effect at once.
# window_size = 1920x1080
# fullscreen = 0
# Which monitor the game window opens on, centred in its work area; fullscreen covers that monitor. auto (default): the
# system places the window. primary; a number, counting from 1 in the order `nx2navi.exe --list-monitors` prints; or part
# of the monitor's name, ignoring case (on Windows also of a GDI device name, display3; on Linux the name carries the
# connector, such as dp-2). For a virtual display that may not exist yet, use its name: when nothing matches, the log
# lists the monitors and the window is placed as with auto. The command line `--monitor <value>` overrides this for one
# run. Read at start. On Wayland a window cannot place itself: only fullscreen opens on the chosen monitor, and a
# windowed window opens where the compositor puts it.
# monitor = auto
# 1, the default: the game keeps running while you are in another window; the keyboard stays quiet until the window is
# in front again. 0: it pauses when its window loses focus, as it does on the Switch when Home is pressed; a load
# interrupted that way can stop for good. Read at each focus change.
# assume_focus = 1
# Performance overlay in the top right corner: off, fps, or full (frame rate, mean and longest frame time). F10
# cycles the modes in game, as do both stick clicks (L3 + R3) on a controller.
# perf_overlay = off
# Which GPU to use: its index in the log, or part of its name. Default: the first discrete GPU with Vulkan 1.3.
# gpu_index = 0
# gpu_name = NVIDIA
# Game resolution: the size the game draws its picture and interface in (1280x720 to 7680x4320). Above 1920x1080 the
# game's texture heap and memory pool grow with it. Another shape than 16:9, such as 3440x1440 on a 21:9 monitor or
# the Steam Deck's 1280x800 (16:10), gives the camera's view that shape and lays the HUD out for it. The installer
# prepares the UI layout for fifteen sizes: 2560x1080, 3440x1440, 3840x1600, 5120x2160 (21:9), 3840x1080, 5120x1440
# (32:9), 1280x800, 1680x1050, 1920x1200, 2560x1600 (16:10), 1920x1280, 2160x1440, 2256x1504, 2880x1920 and 3000x2000
# (3:2). Any other size
# gets the camera's view but a 16:9 HUD, stretched over the picture, unless you build its layout with
# tools/make_romfs_patches.py --aspect WxH. Read at start.
# output_res = 1920x1080
# Ocarina of Time sizes its picture with these two instead (output_res and render_scale do nothing there). The shape:
# 16:9, window (the window's shape, or the monitor's in fullscreen), 21:9 or 32:9. The height: 1080 (lighter than the
# game's own), 1440 (the game's own), 1800, 2160, 2880 or window (the window's height, at least 1440). The DLSS and
# FSR 3.1 modes render their share of this picture. Read at start.
# game_aspect = 16:9
# game_res = 1440
# 3D render scale: the scene's size relative to the game resolution, game (5/6, the game's own 1600x900 at 1080p) or
# a percentage (25 to 400, at most 4096 pixels a side); above 100 supersamples the scene (anti-aliasing). Read at start.
# render_scale = game
# Discrete GPUs: a game memory pool whose copies to and from video memory each frame reach this percentage of its size
# is read and written by the GPU in system memory instead (0: every pool keeps its video memory copy). For GPUs and
# buses short of bandwidth; on an RX 580 the pools the game refills every frame cost about 8 ms a frame in copies.
# mirror_hot = 0
# 2: the next Rebuild the Game compiles the game's code at -O2 instead of -O1 (faster game threads on a slower CPU, a
# build several times longer). Read by tools/make_game.py; the environment variable NX2NAVI_GUEST_OPT wins.
# guest_opt = 1
# GPUs without mesh shaders (VK_EXT_mesh_shader: AMD before RDNA 2, NVIDIA before Turing): 1 runs the game's mesh
# and task shaders (grass and other foliage) as compute shaders and draws what they make; 0 skips those draws. The
# arena for their vertices and triangles is mesh_emu_mb MiB of video memory, with mesh_emu_tasks task records (3 KiB
# each) per submission; with trace = perf the log says how much a submission used. Read at start.
# mesh_emulation = 0
# mesh_emu_mb = 256
# mesh_emu_tasks = 16384
# Scaling to the window when it is larger than the game resolution: fsr (AMD FSR 1) or bilinear, and FSR's sharpening
# from 0 (off) to 100. Changes while the game runs take effect at once.
# upscaler = fsr
# sharpness = 80
# Picture shape when the window's shape differs from the game resolution's (a 16:9 picture in a 21:9 fullscreen
# window): fit, the default, keeps the shape, with black bars at the sides or above and below, so a 16:9 game in a wider
# window is no longer stretched as it used to be; stretch fills the window. Changes while the game runs take effect at
# once.
# aspect = fit
# Anisotropic filtering of surface textures: game (the level the game asks for, as on the Switch, 1x or 2x) or 2, 4,
# 8, 16. Default: 16. Keeps ground, roads and walls seen at a shallow angle sharp into the distance. Only filtered,
# mipmapped textures that tile are raised; shadow maps, data maps and screen-sized buffers keep the game's setting.
# Changes while the game runs take effect at once.
# aniso = 16
# Texture detail: auto, full, reduced or low. The textures of a loaded area are what fills a card with little video
# memory, and three quarters of a texture is its largest mip level. reduced makes the large surface textures (256x256 and
# more, with smaller levels) without that level, low without the two largest when they are 1024x1024 or more. Surfaces
# seen up close look softer; interface art, small textures, shadow maps and everything the game draws itself keep their
# size. auto picks by the video memory Windows grants the game: full from 8192 MiB (cards of 10 GB and more), low under
# 6144 MiB (cards of 6 GB and less), reduced between (8 GB cards). Where the ASTC textures stay uncompressed (astc =
# cpu or astc_bc = 0) and the budget is under 20480 MiB, auto goes one step lower, as they take four times the memory.
# The log says what was picked ("texture detail:"). Read at start.
# texture_detail = auto
# Sun shadow map size: game (the game's own, which changes with the view and drops to about 400 near the ground, so
# leaf shadows look blocky and jump) or a fixed 256, 512, 768, 1024, 1536 or 2048. Above 960 the texture heap grows with
# it. 256 is coarser than the game's own and is for integrated GPUs short of memory bandwidth. Read at start.
# shadow_res = game

# Shadow map resolution: the game's sun cascades are 960x960 (and its static shadow maps 1440 and 720) whatever the game
# resolution; 2 to 4 draws them that many times larger on each side, for sharper shadows. 1 = the game's. The game's
# shadow edge filter is narrowed by the same factor (translator patch 0011), so close shadow edges don't look doubled,
# and the game's slope-scaled depth bias grows by it, so the larger maps don't shadow the surfaces they are cast on.
# Read at start.
# shadow_scale = 1
# Sun shadow cascade range in percent of the game's (its cascades end at 20, 40, 80 and 160 m), 50 to 400: higher moves
# the steps where shadows turn softer further away, and spreads the shadow map over more ground. Read at start.
# shadow_range = 100
# Screen-space contact shadows, strength in percent (0 = off): after the game lights its scene, each pixel traces a
# short way toward the sun through the depth buffer and darkens where something close blocks it, adding the small
# shadows at feet, under stones and in grass that the shadow maps miss. Read at start.
# contact_shadows = 0
# Soft sun shadows (PCSS), percent of the default penumbra width (0 to 200; 0 = the game's own four-tap filter): sharp
# where an object meets the ground and softer the further its shadow falls. Done in the translated shaders (translator
# patch 0010), so the shader cache is rebuilt once after an update. Read at start.
# pcss = 0

# NVIDIA DLSS / DLAA for the 3D scene, on NVIDIA RTX GPUs with the DLSS model from NVIDIA's DLSS SDK next to the
# executable (Windows: nvngx_dlss.dll next to nx2navi.exe; Linux: libnvidia-ngx-dlss.so.<version> next to nx2navi, with the
# NVIDIA driver's libnvidia-ngx.so.1 installed); elsewhere the setting does nothing. The HUD is drawn after it and is not touched.
# Read at start.
# dlaa = 0
# What DLSS works from: dlaa (the 3D scene at full size), quality, balanced, performance or ultra_performance (the
# scene at 67, 58, 50 or 33 % of the output, scaled up by DLSS; these replace render_scale and sharpen texture filtering
# to match). They apply only with dlaa = 1 and the DLSS model next to the executable; otherwise the scene follows
# render_scale. Changes while the game runs take effect at once.
# dlss = dlaa
# AMD FSR in DLSS's place, on any GPU (Windows): with taa = fsr3 and dlaa = 1 the scene is scaled by AMD FSR 3.1 in the
# mode dlss names (dlaa is FSR's Native AA). The menu's Upscaling row sets all of these. Changes while the game runs
# take effect at once.
# taa = dlss
# AMD FSR 4 in FSR 3.1's place (taa = fsr3, dlaa = 1): the machine-learned upscaler, a sharper and steadier picture
# that costs more GPU time (on an RX 7900 XT at 3440x1440, Quality: 47 fps where FSR 3.1 gives 51) and about 220 MB of
# video memory. It runs where AMD's amd_fidelityfx_loader_vk.dll and amd_fidelityfx_upscaler_vk.dll are next to the
# executable (the setup puts them there) and that library lists FSR 4 for the GPU: Radeon RX 6000, RX 7000 and RX 9000
# cards with AMD's driver 26.10.1 or newer. Elsewhere, or when it fails, FSR 3.1 runs and the log says why. Changes
# while the game runs take effect at once.
# fsr4 = 0
# The menu's Frame generation row lists it as "AMD FSR 4 library". With fsr4 = 1 and framegen = fsr3: 1 takes the
# frame generation of the same AMD library
# (amd_fidelityfx_framegeneration_vk.dll next to the executable) instead of the FSR 3.1 library's. The library picks
# the version for the GPU: the machine-learned FSR Frame Generation 4 on Radeon RX 9000 cards with Windows 11, which
# has not been tested, and 3.1.6 on every other, where it runs at the same frame rate as the FSR 3.1 library's. When
# it fails, the FSR 3.1 library takes over and the log says why (fg4:). Changes while the game runs take effect at once.
# fsr4_fg = 0
# The debug menu, a tool for testing (Ocarina of Time 1.1.1): with 1, F9 opens a window with the game's on/off flags
# by name (the names found in your own game's event files), a teleport to every area's start positions (the warp
# songs' pads and both sides of the loading zones among them), the game's cutscenes and other events to start by name,
# the game's speed, Link's position, a free camera flown with the sticks, and a search of the game's memory with
# values to set and to hold. Changing flags or memory, teleporting and starting events can break a save, and the game
# saves by itself after a change of area: keep a copy of the save first. Do not bind F9 to a game button while it is on. The menu's Game tab has the switch. Takes effect at once.
# debug_menu = 0
# The DLSS model: default (NVIDIA's advice per mode: K for dlaa, quality and balanced, M for performance, L for
# ultra_performance), k (K in every mode) or j (the other transformer model: slightly less ghosting behind moving
# characters, a little more flicker). Any other model by its letter, a to m: on slower RTX cards the older model e is
# cheaper. Changes while the game runs take effect at once (DLSS starts a new history).
# dlss_preset = default
# DLSS keeps one network per picture size (gameplay, the pause screen). 1: one unused for 3,600 frames is freed.
# 0: none is ever freed (about 0.5 GB more video memory after the first pause). 2: as 1, and the networks that stay
# are not made again when another is made (shows striped pause screens; for tests).
# dlss_idle_free = 1
# Per-object motion vectors for DLSS: characters, animals and moving objects get their own motion instead of the
# camera's, so DLSS keeps their history instead of smearing or rejecting it. Costs some CPU time each frame (about 1-2 ms
# on a laptop CPU); 0 leaves camera-only motion. Read at start.
# dlaa_objmv = 1
# Motion vectors of Link's own for DLSS, FSR 3.1 and frame generation (game 1.1.1): Link and the equipment attached to him are
# drawn a second time with last frame's pose, so fast turns, attacks and jumps move by their real motion instead of an
# estimate from colour and depth. 0, the default: the estimate, as before. 1: on, to try; it reads the player's
# equipment from the game's memory at every submission and has not been played on this version yet. Read at start.
# motion_guard_degrees (35, 45 or 60, the default): a part that turns more than this in one frame is checked against
# last frame's depth before its vector is written.
# selective_motion = 0
# Characters' trails: where a character, animal or held item moves without motion vectors of its own (all of them with
# dlaa_objmv = 0), DLSS can drag a faint copy of it. The G-buffer marks those objects; on them and where they were the
# frame before, DLSS is told to favour the current frame by this percentage (0, 35, 65 or 100; 0 off). Higher removes
# more of the trails and leaves those edges a little less smooth. Changes while the game runs take effect at once.
# dlss_char_mask = 65
# Anti-aliasing of the 3D picture: game (the game's own SMAA, as on the console), off, fxaa, smaa (SMAA at its Ultra
# preset in the game's place) or taa (DLSS or FSR at full size: dlaa = 1 with dlss = dlaa, which the menu's
# Anti-aliasing row sets with it, and the game's SMAA). With DLSS or FSR on (dlaa = 1, any mode) their TAA always runs,
# and game, off, fxaa or smaa runs along with it: off leaves the edges to the TAA alone. Changes while the game runs
# take effect at once.
# aa = game
# The FXAA switch of the game this runtime was first written for; Ocarina has no such pass, so it changes nothing here
# (aa above is the setting).
# fxaa = 1
# Frame generation: off, fsr3 (AMD FSR 3.1 on any GPU where amd_fidelityfx_vk.dll loads, so not on Linux: one generated
# frame between every two the game draws, the HUD kept out of it) or dlss (NVIDIA DLSS Frame Generation on GeForce RTX
# 40 and 50 GPUs, paced with Reflex; it needs nx2navi_dlssg.dll and NVIDIA Streamline's files next to the executable,
# which the setup puts there). Both need DLSS or FSR 3.1 upscaling (dlaa = 1). While frame generation runs, fps_limit
# counts the frames the game draws and the screen shows that times the multiplier (2 for fsr3); fps_limit = display
# alone holds the screen to the monitor's rate and the game to that rate divided by the multiplier.
# fsr3 takes effect at once, dlss (on or off) at the next start.
# framegen = off
# On Linux dlss runs through NVIDIA's NGX itself, without Streamline: it needs dlaa = 1 with DLSS when the game starts
# and NVIDIA's libnvidia-ngx-dlssg.so.<version> (from NVIDIA's DLSS SDK repository) beside the game, then switches at
# once, and its multiplier is fg_mult: 2 to 6 (above 2 needs an RTX 50 GPU) or dynamic, which generates only the frames
# needed to reach the display's refresh rate at the rate the game draws. framegen_multiplier is not used there.
# fg_mult = 2
# DLSS frame generation's multiplier: the frames on screen for each frame the game draws, 2, 3 or 4 (one, two or three
# generated). 2 needs an RTX 40 GPU or newer, 3 and 4 an RTX 50 and a current driver. A value above what the GPU and
# driver allow (Streamline reports it) is lowered to that, with a line in the log; the menu offers only what they allow.
# Each step adds a little latency, and a display shows no more than its refresh rate. Changes while the game runs
# take effect at once.
# framegen_multiplier = 2

# HDR output, two kinds (both need HDR on in Windows, "Use HDR", or a Wayland compositor with colour management;
# without it the picture stays SDR, the game's own): hdr = 1 sends the game's scene tone mapped to your display, one
# fixed look, and applies at once; hdr = native shows the game's own HDR mode, which the game takes up at the next
# start. hdr_peak_nits is your display's peak, hdr_game_nits the picture's white (paper white, 203 is the HDR
# reference) and hdr_ui_nits the HUD's and the menu's. The settings of the look of earlier versions
# (hdr_tonemap, hdr_exposure, hdr_gamma and the rest) are read by nothing now and do no harm.
# hdr = 0
# hdr_live = 0 leaves the HDR path out of a run that starts with hdr = 0, as before hdr followed the menu: switching
# HDR on then takes a restart. Only worth setting if the picture differs with it at 1.
# hdr_live = 1
# hdr_peak_nits = 1000
# hdr_game_nits = 203
# hdr_ui_nits = 203
[language]
# Language: a Windows locale such as de-DE, or a Switch code (en-US, ja, zh-Hant, ...). Default: the Windows locale.
# locale = en-US
# lang = en-US
# System mode reported to the game: docked or handheld.
# operation_mode = docked
# Language of the settings menu and the overlay: en, ru, pt-BR or zh-Hans. Default: the Windows display language, English if
# it is none of them.
# menu_lang = en

[discord]
# Rich Presence: "Playing NX2navi" with the play time in your Discord status while the game runs, through the
# local IPC of a running Discord client; without one it does nothing. 0 = off. The Game tab of the settings menu (F1)
# and the launcher's Settings switch it too, and a change in the menu reaches the presence thread at once.
# discord_rpc = 1
# The Discord application whose name ("Playing <name>") and artwork the status shows: its Application ID from
# discord.com/developers/applications (General Information), digits only. Default: the NX2navi application, built
# in from runtime/discord_app.h (NX2NAVI_DISCORD_APP_ID_DEFAULT). Register your own application to choose another
# name and art.
# discord_app_id = 1568862362969309275
# The key of the application's uploaded art asset (Rich Presence > Art Assets) shown as the status's large image.
# discord_image = nx2navi_512x512

[coop]
# Co-op: your own game and save, with the other players' progress shared through a relay (docs/coop.md). Positions are
# not shared yet: you do not see the other players in your world.
# The menu's Co-op tab sets the name, the relay and the rules below, and makes, joins, accepts and leaves a room.
# 0 = off: no socket is opened, no name is looked up and nothing is sent. Takes effect at once.
# coop = 0
# Co-op is a beta: the menu shows its Co-op tab only with coop_beta = 1, which no menu row sets. Takes effect at once.
# coop_beta = 0
# Your name for the other players, at most 24 bytes. Default: Player- and four random hex digits.
# coop_name =
# The relay, host or host:port (default port 24680): run one with `python tools/coop_relay.py` and forward its UDP port, or
# use one a friend runs. No default: with none set the session stays idle and the log says why.
# coop_server =
# The local UDP port, 0 = any. For a firewall rule or a port forward when testing on a LAN. Takes effect at the next start.
# coop_port = 0
# For tests and LAN parties, no menu row: coop_auto = create or join starts a session at launch (a join needs coop_room, the
# room's secret of 18 characters from 23456789BCDFGHJKMNPRSTVWXYZ, XXXXXX-XXXXXX-XXXXXX; a create without it makes a new
# one). While the session runs, a room the relay lost (a restart) is made again or waited for (asked for every 10 s; a
# refusal for too many joins from your network waits a minute), and a create that finds the room open joins it. Joining
# is not accepting: nothing from the room is applied before you accept it (coop_accept).
# off = no session at launch.
# coop_auto = off
# coop_room =
# Test: percent of received co-op datagrams thrown away, 0 to 100.
# coop_drop = 0
# The rules of a room you make (a room you join uses its maker's; the log names them). Take effect at once.
# Story, world and dungeon progress: shared or own.
# coop_progress = shared
# Equipment, key items, upgrades; songs; medallions, stones and the other quest items: shared or own. Own only with
# coop_progress = own: in a shared world a chest one player opened is empty for the others.
# coop_items = shared
# coop_songs = shared
# coop_quest = shared
# Small keys: pool = every key anyone picks up goes into the dungeon's pool and every door anyone opens takes one out of it;
# own = keys stay with the player who picked them up (doors still open for everyone). Own with coop_progress = own.
# coop_keys = pool
# Rupees and ammunition: own, or pool = one count every player adds to and spends from.
# coop_rupees = own
# coop_ammo = own
# Nothing from a room is applied to your game before you accept who is in it (by their keys) and its rules, and before
# your save is copied to coop-backup in the profile folder (docs/coop.md: going back). ask = the log shows each room's
# offer with a number (eight hex digits, new at each start of the game); accept it on the menu's Co-op tab, or put that
# number here while the game runs to accept that offer (this line as the file is now counts, before the one read at
# start). A new member, other rules or another save loaded make a new offer with a new number (another save joins the
# pools only once you accept it). A member who never says who it is keeps the room from being accepted. Your key pools
# count only you and the players you accepted, among those the room's maker lets use them. Every message that can
# change your save is signed by the player who sent it, so someone who knows the room's secret and runs the relay (or
# sends from its address on your network) can read and disturb the room but not set flags in a player's name.
# Accepting every room (always) is taken from the environment only (NX2NAVI_COOP_ACCEPT=always), never from this file,
# so an ini someone hands you cannot accept a room for you.
# coop_accept = ask
# The room's maker (the Co-op tab's Remove and Reset the pools do the same): coop_remove = a player's key fingerprint
# (xxxx-xxxx-xxxx-xxxx, as the log and the tab show it beside each member; without dashes it may be all digits), in the
# room or after it left; every game drops that key and what it took from the pools, and keys others brought while it
# was in a pool stay theirs. A member number here is not taken: numbers pass to other players as members come and go.
# coop_reset_pools = a new value (1, then 2, or 0 and then 1 again): every game joins the pools again with what they
# held for the players in the room, and what each held counts from the reset on. A player who drained a pool and is
# still in the room when you reset keeps what it took: remove it first, then reset (both in one save of this file act
# in that order).
# Both are read from this file once a second while you are in a room, and each value acts once. The key your game signs
# the rules of a room you made with is kept in the profile folder (coop-keys, readable by your account only), so a
# restart keeps you the room's maker; never hand out a copy of the profile folder.
# coop_remove =
# coop_reset_pools =

[audio]
# none mixes without opening an audio device.
# audio = default
# Audio the output queues before it starts, in ms (0 to 500; above about 170, half the output ring, it is held there).
# audio_prefill_ms = 20
# Log where gaps in the sound come from (1 = on): output underruns, late passes of the audio render thread, late
# audio updates from the game and voices that ran out of queued sound, with the game frame of each.
# audio_stats = 0
# Host priority of the game's audio thread: 0 leaves it alone, 1 raises it to the highest while its audio update is
# read (the audio render thread waits for that), 2 also keeps it above normal. Takes effect at the next start.
# audio_game_prio = 0
# Volume in percent (0 to 100), on top of the game's own. Changes while the game runs take effect at once.
# volume = 100
# Speakers: stereo, 5.1 or auto (5.1 when the Windows speaker setup has six or more channels, stereo otherwise). The
# game is told the same channel count and mixes for it. A change while the game runs takes effect at once: the game is
# told its output device changed, as on a console, and mixes for the new count.
# audio_channels = stereo
# Which output device the sound plays on. default (the default): the system default device, which the game follows when
# the Windows default changes. Or a number, counting from 1 in the order `nx2navi.exe --list-audio-devices` prints; or part
# of a device's name, ignoring case (`headphones`). When nothing matches, or the device will not open, the log lists the
# devices and the game plays on the system default. With audio_channels = auto, the channel count is the chosen device's.
# The command line `--audio-device <value>` overrides this for one run. Takes effect at the next start.
# audio_device = default
# The game's microphone input (the Ocarina can be played into one): 1 lets it read the default recording device,
# 0 refuses it, as earlier versions did.
# microphone = 1

[input]
# Gamepad motion controls (0 = off) and vibration (0 = off).
# motion = 1
# rumble = 1
# Rumble strength in percent (0 to 300; 0 = off) and the exponent on the amplitude (0.2 to 1.5; lower = weak pulses
# stronger, 1 = linear). rumble_triggers = 1 also drives the impulse triggers of an Xbox pad.
# rumble_strength = 130
# rumble_curve = 0.6
# rumble_triggers = 0
# PlayStation 3 controller (DualShock 3), Windows: 1 reads its buttons, sticks and rumble directly from the controller
# when no driver presents it to the game as a gamepad; 0 (default) leaves it to SDL and XInput. Changes while the game
# runs take effect at once.
# ds3 = 0
# Which gamepad button is Switch A: nintendo (the default: the Switch's buttons by position on any pad, east A,
# south B, west Y, north X), xbox (south A, east B, west X, north Y, as an Xbox pad labels them; cross A on a
# PlayStation pad), playstation (the same as xbox: cross A, circle B, square X, triangle Y) or auto (nintendo for a
# pad with Nintendo labels, xbox for the others). The bind_ lines below apply
# on top. Changes while the game runs take effect at once. 0.1.9's xbox_prompts is read as nintendo with prompts = xbox.
# pad_layout = nintendo
# The buttons the game's prompts show: nintendo (the game's own), xbox (LB RB LT RT, Menu, View), playstation (L1 R1
# L2 R2, Options, Create) or steamdeck (the Deck's A B X Y, L1 R1 L2 R2, Menu, View), with A B X Y as the button
# pad_layout presses is labelled: with nintendo, A shows as B or circle and X as Y or triangle. pc: the key each
# button has (the bind_ lines below when the files are made), WASD
# and the mouse. Read at start; their files come from tools/pad_prompts.py (docs/controls.md, "Button prompts").
# prompts = nintendo
# Gamepad stick dead zone in percent of full deflection (0 to 50). Raise it for a stick that drifts: inside the circle
# the stick reads centred, and the rest of its travel is scaled to start there.
# stick_deadzone = 8
# Right-stick camera speed (0.1 to 3): multiplies the game's own camera turn rates, top speed included, at each of the
# five camera sensitivities in the game's Options. Keys bound to the right stick follow it; the mouse keeps mouse_sens
# for its turn per movement, and in stick mode the fastest turn it can reach rises and falls with this.
# A change while the game runs takes effect at once (the first one of a run within a few seconds).
# stick_sens = 1
# The dead zone of one stick, in place of stick_deadzone for that stick (0 to 50; unset = stick_deadzone). The Controls
# tab shows each stick's deflection now, so a drifting stick's resting value can be read off.
# lstick_deadzone = 8
# rstick_deadzone = 8
# Mouse camera mode: direct (default) or stick. Direct turns the camera by the raw mouse motion, the same on both axes,
# with no stick dead zone or response curve, at any frame rate and however fast the mouse moves. Stick makes the
# mouse a right stick. Neither changes the gamepad. The Controls tab changes this while the game runs.
# mouse_mode = direct
# Mouse camera sensitivity (0 to 10; 0 = no mouse camera and no cursor capture). In direct mode a count of the mouse
# turns the camera about 0.08 degrees at 1; the game's own camera sensitivity (Options) scales it as it scales the
# stick. The Controls tab changes this while the game runs.
# mouse_sens = 1
# The keyboard's sticks are all or nothing, so the game only sees a full tilt from them; while this key is held the
# keyboard-driven stick values shrink and the keyboard can walk instead of run. Key names as in the bind_ lines below,
# none turns it off. 0.2 to 0.9, the default holds for anything else. Only the keyboard is scaled: the mouse camera and
# the gamepad are unchanged. The Controls tab sets both. The default stops just above half because a keyboard sample of
# 32767 times 0.5 is one short of the 16384 the game's stick-direction bits need, and below that the keyboard stick
# reads as a partial tilt only.
# walk_key = lalt
# walk_scale = 0.51
# Write the input of this run to a file that replays as a script (docs/controls.md, "Scripted input"). Default: off.
# input_record = input.txt
# Remap buttons: the Controls tab of the settings menu (F1) writes these lines. One line per Switch button or stick
# direction: bind_<A B X Y L R ZL ZR PLUS MINUS LEFT UP RIGHT DOWN LCLICK RCLICK> or bind_<LSTICK RSTICK>_<LEFT UP RIGHT
# DOWN>, or bind_SHAKE (shakes the right stick; no default key, Ocarina has no use for it) = comma separated
# pad:<name> and key:<name> entries, key:default / pad:default for that part's defaults, or none.
# A line replaces that button's defaults (pad and keyboard); buttons without a line keep them. Stick directions and
# SHAKE take keys only. Invalid entries are logged and skipped. Changes while the game runs take effect at once.
#   pad names: south east west north l r zl zr start back lstick rstick dpad_left dpad_up dpad_right dpad_down
#   lpaddle1 rpaddle1 lpaddle2 rpaddle2 (the back buttons of an Xbox Elite, DualSense Edge or Steam Controller)
#   key names: a-z 0-9 f2-f7 f9 f12 num0-num9 space tab enter backspace lshift rshift lctrl rctrl lalt capslock insert
#   delete home end pageup pagedown up down left right num_add num_sub num_mul num_div num_decimal minus equals
#   lbracket rbracket semicolon apostrophe comma period slash backslash grave mouse_l mouse_r mouse_m mouse_x1 mouse_x2
# Defaults (pad_layout = nintendo): A = pad:east, key:E, key:lshift (held: dash). B = pad:south, key:lctrl,
# key:backspace. X = pad:north, key:space. Y = pad:west, key:mouse_l. A pad that reports A/B and X/Y swapped
# (some GuliKit models): set pad_layout = xbox, or swap them back here (keyboard keys stay):
# bind_A = pad:south, key:E, key:lshift
# bind_B = pad:east, key:lctrl, key:backspace
# bind_X = pad:west, key:space
# bind_Y = pad:north, key:mouse_l

[mods]
# Folder with the player's mods, one subfolder per mod (<name>/romfs, <name>/exefs, <name>/cheats). A relative folder
# is under the game's folder. Default: mods next to the game's exe. The build takes the ExeFS patches and code cheats
# of the mods folder next to the exe only. The launcher's Mods screen adds mods and puts downloads in order.
# mods_dir =
# Load order, top first, names separated by |; a mod not listed goes below the listed ones, by name. The top mod wins a
# file two mods replace. The launcher's Mods screen sets it (Load Order); the in-game menu has no Mods tab.
# mods_order =
# Mods switched off, names separated by |. The launcher's Mods screen sets it.
# mods_disabled =
# auto: when an enabled mod has cheats that change the game's code and the game was built without them, the game starts
# the setup's rebuild at start (a few minutes) and starts again after it. off: the game starts with the code it has.
# A rebuild that failed or was cancelled is not started again for the same cheats. Default: auto.
# mods_rebuild = auto

[files]
# Developer overlay: a folder with RomFS files and a manifest.tsv (tools/make_romfs_patches.py format). Players put
# their mods into the mods folder instead ([mods]). Default: none.
# romfs_mods =
# Folder with the Switch shared fonts (BFTTF files, tools/extract_fonts.py output), for Chinese and Korean text.
# Default: shared-fonts next to nx2navi.exe. NX2NAVI_SHARED_FONTS overrides it.
# shared_fonts =

[cheats]
# One line per cheat of the mods' cheat files, mods\<mod>\cheats\<build id>.txt (docs/cheats.md); the Cheats tab of
# the menu writes them. cheat_<hash> = 1 switches a cheat on, the hash being FNV-1a 32 of "<mod folder>/<cheat name>".
# Default: every cheat off.
# cheat_0123abcd = 1

[diagnostics]
# Extra lines in the log, comma separated: svc, host, ipc, fs, gfx, perf, sched, audio, present, coop or all. The menu's
# Performance log sets perf (frame timings). Default: none.
# trace = perf
# A game thread that waits this many seconds for a lock, or for another thread to end, is logged once with the thread
# it waits for and what that one is doing (at most 24 lines a run): a game that stops then leaves the chain in its log.
# Off by default: its thread takes the game's thread lock every 2 seconds. 10 is a good value for a hang report.
# wait_watch_s = 0
# 1: with trace = perf, walks of the process's address space count its private and write-combined memory every 600
# frames (the "mem: process" line) and where its committed memory lies every 3600 frames (the "commit census" lines).
# Each walk holds the address-space lock for 100-240 ms and delays the next frame by as much, so they are off unless
# asked for.
# mem_census = 0
# The memory the game is taken to need, in MB of commit (RAM plus page file not yet promised to a program). When less
# is free at the start, the log gets a warning line: short of memory the game stutters, loses textures or stops with
# an out-of-memory or lost-device error. The default is what the game took in play at 3840x2160 on an RTX 5090; the
# video card's memory counts on Windows. 0: no warning. Default: 24576.
# mem_need_mb = 24576
# The size in MiB (16 to 256) of the blocks the renderer takes video memory in. A block goes back to the driver only once
# everything in it is freed, so smaller blocks hold less free space after area changes; larger ones need fewer
# allocations while an area loads. Default: 64.
# vma_block_mb = 64
# Flight recorder: keeps the last frames and saves them to a flight folder next to the folder of nx2navi.exe when a frame
# takes longer than flight_spike_ms (0 to 60000).
# flight = 0
# flight_spike_ms = 100
# GPU pass marks: after a GPU crash (device lost) the log names the render pass or dispatch the GPU had started and the
# last one it finished. Unset: on AMD GPUs (VK_AMD_buffer_marker). 1: also on NVIDIA GPUs
# (VK_NV_device_diagnostic_checkpoints), which NVIDIA says costs a little. 0: off.
# pass_crumbs = 1
# A test of how the game ends when the video card stops answering (device lost): the hand-over to the GPU queue with
# this number is not made and is taken as one the driver refused with a lost device. The log then holds one report and
# one FATAL line, and the game ends with exit code 72. 0: never. Default: 0.
# device_lost_at = 0
# Once a second, puts the crash report's exception filter back if a GPU driver or an overlay loaded after the start set
# its own in its place, and says so in the log. Without it, a crash in such a DLL can end the game with no FATAL line in
# the log. 0: off.
# crash_filter_guard = 1
# Render targets up to this size (bytes, 0 to 1073741824) go back to their game memory whenever the game waits for a
# GPU fence. A larger target goes back too unless the GPU drew it in each of the last 16 frames. A game that reads a
# larger target drawn every frame without invalidating it sees stale bytes at the default. With trace = gfx the log
# names each dirty target that stays on the GPU ("fence write-back skipped-drawn-every-frame") and each one over 65536
# bytes that went back ("over-default"). Raising it costs a copy per such target and wait.
# wb_fence_limit = 65536
# Log kernel services that block longer than this many ms, with the service number (0: off). A stalled frame shows up as
# time in the replay; this says whether the guest was waiting in a service, and which one.
# svc_slow_ms = 0
# 1: every signalled event and every thread that ends wakes all the game threads that wait for any event, as before;
# each takes one of the game's virtual cores to find that its own event is not the one, often from a thread that was
# working. 0 (default): only the threads that wait for that event or thread are woken. With trace = sched the
# "sched:" line counts both ("signals woke ..., left ... asleep").
# sync_wake_all = 0
# 1 (default): a thread of its own writes the log lines, to the log file and the console, so a write that stalls (a
# busy disk, a virus scanner, a console that scrolls) no longer holds the thread that logged, which is often the one
# that draws the frame. A crash, a fatal error and the end of the game write out what is still queued first. 0: each
# line is written by the thread that logs it, as before. Either way a "log: ... writes of the log took over 2 ms" line
# counts the slow writes and says whose time they took.
# log_async = 1
# 1: the frame-time summary also splits the replay by op kind (draw, bind, transfer), which says where a stalled frame
# went. The clock reads cost a fraction of a millisecond a frame, so it is off by default.
# replay_time = 0
# A frame whose texture work on the render thread (new textures, uploads, upload submits) takes at least this many ms
# gets a "texup:" line in the log that splits it: creation, staging, storage copy, ASTC recording, scratch buffers and
# new device memory (up to 400 lines; 0: none; 16 catches the area-change hitches).
# texup_log_ms = 0
# 1: the frame-time summary also lists the locks the replay waited for, by the code that asked (addresses; the line's
# "ref" is perf_lock_timing_here, which the link map names), the recorder's waits, and the texture lock's holds over
# 1 ms. Off by default.
# lock_time = 0
# Texture entries the bindless texture pool may create in one refresh (images and their uploads); the rest wait for
# the next refresh and show their stand-in until then. Lower spreads streaming over more frames (smaller hitches, a
# frame or two of pop-in); higher loads sooner.
# pool_create_budget = 16
# The refresh runs before every draw that samples the pool, so the budget above holds per draw. Once a frame's
# refreshes have taken this many ms, they create nothing more until the next frame (the first refresh of each frame
# keeps its budget): an area load then spreads over several frames with stand-ins instead of holding one frame for
# 200-400 ms. 0: no limit.
# pool_create_ms = 6
# An area load is told from streaming by the number of textures still to make: at this many or more, the two limits
# above give way (a frame may take pool_load_ms for them) and the picture before the load stays on screen until the
# area's textures are there, instead of showing the ground and the walls black for the first seconds. 0: never.
# pool_load_backlog = 512
# pool_load_ms = 50
# Written storage-buffer ranges up to this size in MiB (1 to 256) go back to game memory as a merge, in 4 MiB pieces:
# copied within video memory and only the words that changed written over PCIe. Larger ones are copied back whole at
# the end of every submission. Raising it from 4 to 16 took a 5.5 MiB range off the plain copies (about 46 MiB a frame
# in the field), and 32 the 4-15 MiB ranges of Kokiri Forest.
# wb_merge_max = 32
# 1: pages a VRAM mirror wrote back to game memory are not copied up into it again, as it already holds them (a write by
# the game still is). Outdoors this removed about 57 MiB of copies a frame. 0: they are copied up again.
# wb_no_reup = 1
# 1: finds out who reads what the GPU wrote back into guest memory. The 2 MiB regions inside a range just written back
# are made inaccessible until their first read or write, which is logged with its instruction ("mem: write-back
# probe"). The bytes are there all along; the probe only watches, at the cost of the faults it takes.
# wb_probe = 0
# The wait after each frame's first GPU submission. 1 (default): it ends when the video card has run the submission
# and the results the game reads next are in its memory (the last frame's visibility counts above all); what the card
# wrote besides goes back to the game's memory in a submission of its own, behind the wait. 0: the game waits until
# everything before it has reached its memory, about 0.7 ms more of every frame in the open. 3: the wait ends as soon
# as the card has run the submission, without the visibility counts: the game then reads counts that may be a frame
# old, and lit things near the camera can lose their light for single frames. With shadow storage (Linux, Wine) 1
# and 3 change nothing; 2 forces 1 there as well, for tests.
# wb_split = 1
# 1: what the game's shaders write into its memory is copied back from the video card once, and after that only when
# the game's CPU code touches it (most of it never is: 66 MiB a frame in Hyrule Field), for about a third more frames
# a second in open areas. 0: copied back at the end of every GPU submission, as in 0.7.0.
# Default: 1 on Windows, except 0 on AMD graphics cards (AMD's own driver), where two reports had the game end in play
# with it on and the cause is not found: set 1 there to try it, and back to 0 if the game ends in a fight (the log's
# "mem: lazy write-back" line says which is in effect and why). 0 elsewhere. With shadow storage (Linux, Wine) it
# stays off whatever is set.
# wb_lazy = 1
# With wb_lazy, off by default: with it on the graphics device was lost shortly after some save loads (about one load
# in three of a save in Hyrule Field at night on one card), and not once in 15 loads with it off; the cause is open.
# 1: when the game's CPU code touches such memory, only the 64 KiB around the place it touched are copied
# back and kept up to date from then on; the rest of that buffer stays on the video card. The game reads a few words
# at the start of buffers of several MiB about twice a second. 0: the whole buffer comes back and everything near it
# is copied back every frame for up to 32 seconds, as before (a tenth fewer frames a second in a village).
# wb_lazy_chunk = 0
# Barriers between the game's draw passes and compute dispatches. 1: only where the game itself asks for one, or where
# the work changes between drawing and compute; this takes about a tenth off the graphics card's time for a frame.
# 0: a full barrier before every pass and every dispatch. Default: 1 on NVIDIA graphics cards, 0 on all others, where it
# has not been tried: set 1 there to try it, and back to 0 if the picture shows faults.
# barrier_game = 1
# Mesh draws whose count the graphics card computes: 1 (default) prepares the counts of a run of such draws at its
# first one, so their draw pass is not ended and begun again at each of them (about 40 passes fewer a frame in a
# village). 0: each such draw ends its pass and prepares its own count, as before.
# mesh_ahead = 1
# Graphics cards without mesh shaders (GTX 10 series, RX 500 and 5000 series, many integrated ones): the game's mesh
# draws, most of the world, are computed in compute passes and drawn as ordinary triangles. auto (default) does that
# only on such a card; 1 does it on every card (for testing: it is slower than mesh shaders); 0 never, and such a
# card then draws no world.
# mesh_fallback = auto
# Video memory in MB for what those compute passes hand to the draws. A draw that needs more loses the geometry that
# does not fit, and the log says so.
# mesh_fallback_mb = 128
# Threads that prepare those compute passes for the mesh programs a place uses, the first time each is drawn (until
# then its draws are left out). auto (default): one for each 4 logical processors, 1 to 4.
# mesh_fallback_workers = auto
# Windows, 1 (experimental): the game's heap is committed a 64 KiB run at a time where it is first used, instead of
# all of it when the game sizes it. Windows counts committed memory against RAM plus the page file whether it is used
# or not, and in Castle Town after Hyrule Field about 3.4 of the heap's 6.8 GiB had never been used. 0: committed up
# front. Linux commits pages on first use anyway. With 1, a PC whose commit is used up ends the game at the moment a
# heap page is first used, in the middle of play, where with 0 it would not have started.
# heap_lazy = 0
# Linux, shadow storage: 1 clears the write watch's protection of a range in one call before the GPU's bytes are copied
# back into the game's memory, instead of taking a fault at the first store into each page.
# shadow_bulk_write = 0
# 1: the game's flushes of mirrored memory are copied into the mirrors even where the write watch already sees every
# page the game wrote; 0 (default) copies only the written pages there, which saved 47-57 MiB of copies a frame in
# Kokiri Forest.
# flush_upload = 0
# For measuring: only mirrors of the game's memory pools up to this many KiB send what shaders write through the
# device address table to game memory as well; larger mirrors keep it in video memory and are not evicted. On PCIe 3.0
# those second writes cost about 3 ms of GPU time a frame. 0 (default): every mirror sends them. A pool that needs them
# and is left without shows as foliage stretched into columns or flickering.
# gpa_twin_max_kb = 0