perf(vk): the recorder takes a whole part and submits it, so the render thread never waits at a part's end #1866

Merged
HumanBean merged 1 commit from HumanBean/Longshot:perf/recorder-parts into main 2026-11-20 17:12:48 +00:00
Owner

Summary

With the recorder thread on (Threaded rendering Auto from 6 cores, merged in #1763), the render thread packed every Vulkan call into a ring and woke the recorder per call; at each part's end it then waited until the recorder had made every call, and submitted the part itself. So the two threads overlapped only inside a part, stopped to sync several times a frame, and the recorder woke ~1,100 times a frame (9 packets each) and spun 4 s in every 10 between them: the churn that once let an i7-9700K's clocks drop and cost a Ryzen 9 7900X 3-11% when spinning.

Now the render thread hands the recorder a whole part, and the recorder ends and submits it: the render thread decides the part's signal values up front and goes straight on to the next part. Anything that needs a part to be on the queue (slot reuse, fence and timeline waits, readbacks, the queue lock's holders, DLSS/FSR/XeSS/menu recording of their own) waits for a "submitted up to part N" counter, bounded; everything else waits for nothing. Threads waiting for the recorder sleep instead of spinning. RX_VK_RECORDER_PARTS=0 is the old hand-off, in the same build.

Read first: docs/dev/reference/threads.md "The recorder / The hand-off by part"; post_part, part_submit and queue_flush_for (gpu/vk/vk_replay.cpp); Recorder::loop, wait_until, settle_parts and QueueMutex (gpu/vk/vk_recorder.h, gpu/vk/vk_internal.h).

Two independent code reviews found and this fixes: a permanent hang with batched submits on RADV (a held last part with a waiter), unbounded new waits, a too-wide lock escape, a poll that defeated RADV batching, and spinning waits. Their last pass found no hang, deadlock or out-of-order submit left.

Testing

Tested: Linux, RTX 3090 (NVIDIA 620.x), Ryzen with 24 threads, 2560x1440: Castle Town Market standing, walking and turning, two game starts with the hand-off on and two with RX_VK_RECORDER_PARTS=0, plus a warm start of each; no Vulkan errors, no hang. The recorder's stats line: 28.5 hand-offs a frame (335 packets each) against ~1,100 (9 each); spun idle 0.16 s against 4.15 s in 10 s; blocked on the queue's lock 0.03 s in 10 s (longest 1.3 ms); the render thread waited for it 0.02 ms a frame. Renderer builds on Linux; Windows syntax check of the changed files with llvm-mingw. New game-free test tests/recorder_parts_test.cpp with 8 controls (compiled, not run here).

Test Farm on f08644e3 against main, three areas, standing (first results; Intel, the Steam Deck and the integrated GPUs pending): no hang and no crash on any of them. Market: RTX 3080 Ti (Ryzen 9 5900X) +3.2%, RX 9060 XT (Ryzen 5 5600, Windows) +4%, RX 9070 XT (Ryzen 7 5800X3D, Windows) +0.7%, RTX 4070 (5800X3D) +0.5%, GTX 1070 (i7-8750H laptop) -2.8%, Steam Machine (Navi 33, AMD custom 6-core) -2.7% (Hyrule Field -7.4%). The two slower ones are power-limited 6-core CPUs: if they hold up, the recorder's Auto rule should leave such CPUs on the render thread alone (by measured render-thread load rather than core count), a follow-up.

Still to test:

  • linux+amd, windows+amd, windows+nvidia, linux+nvidia, windows+intel, linux+intel: three areas against main, expecting equal or better fps and no hang (@Beanstalk)
  • The Steam Deck and 4-core PCs: unchanged (the recorder is off below 6 cores), expecting even (@Beanstalk)
  • Linux and Windows: mise run test, test_recorder_parts.py, test_recorder_sync.py, test_recorder_own_idle.py and the fixtures it touched, expecting them to pass (by hand: CI on this pull request)

Measurements

Verdict: a small, consistent gain on an RTX 3090 (+1.8% Market standing, better p99), and the recorder's wake-ups and spinning nearly gone; AMD numbers come from the Test Farm.

Baseline: this PR's own build with RX_VK_RECORDER_PARTS=0, which is main's hand-off, so the runs differ only by this change.

Setup: Linux, RTX 3090, 2560x1440, Threaded rendering Auto (render thread + recorder), Castle Town Market; two interleaved game starts each, after a warm start of each.

Market Before: main's hand-off, this PR @ f08644e3 with RX_VK_RECORDER_PARTS=0 After: this PR @ f08644e3
standing fps (runs) 87.2 (86, 88) 88.8 (88, 90)
standing p99 ms 14.4 13.4
walking fps 82.6 83.5
turning fps 99.6 100.2
recorder wake-ups a frame ~1,100 28.5
recorder spinning idle, s in 10 s 4.15 0.16

RX_VK_RECORDER_PARTS ships on: never slower here, and it removes the recorder's churn on every CPU that runs it.

Risks

A wrong wait would show as a hang or a read of a part not yet on the GPU: every new wait is bounded and the reviews checked the lock order; RX_VK_RECORDER_PARTS=0 restores the old hand-off. Threads that take the queue lock (texture uploads) now first wait for the parts handed over before them. Nothing changes for the player's files, and nothing changes where the recorder is off (below 6 cores, the Deck).

Checklist

Ready for review

  • One topic, on current main; the title is a Conventional Commit.
  • Tested on my own PC, in the game for a change the game runs, as Testing says.
  • mise run test passes (any failure that also fails on main is named).
  • Measured against main, or Measurements says why there is nothing to measure.
  • No game files, build output or hand-edited generated files; third_party/ untouched; nothing GPL-incompatible.

When it applies

  • A switch added or changed: it ships at the default the measurements chose, with its row in gpu/vk/switches.def.
  • A check under tests/ that reads changed source is updated, not weakened.
  • Host code builds on Windows (llvm-mingw) and Linux, with UTF-8 paths.
  • Docs updated.
  • AI assistance recorded with Assisted-by: / Co-Authored-By: trailers.

Ready to merge

  • Everything in Still to test has run, its results in Tested.
  • Tier 1 hardware it can affect is tested (docs/dev/standards.md, "Hardware tiers").
  • Approved in review, every comment answered.
  • CI passes on the head; on current main, measured again or a comment says why the numbers still hold.

🤖 Generated with Claude Code

https://claude.ai/code/session_017vw8e3aEPBsd7QfUkaatjP

## Summary With the recorder thread on (Threaded rendering Auto from 6 cores, merged in #1763), the render thread packed every Vulkan call into a ring and woke the recorder per call; at each part's end it then waited until the recorder had made every call, and submitted the part itself. So the two threads overlapped only inside a part, stopped to sync several times a frame, and the recorder woke ~1,100 times a frame (9 packets each) and spun 4 s in every 10 between them: the churn that once let an i7-9700K's clocks drop and cost a Ryzen 9 7900X 3-11% when spinning. Now the render thread hands the recorder a whole part, and the recorder ends and submits it: the render thread decides the part's signal values up front and goes straight on to the next part. Anything that needs a part to be on the queue (slot reuse, fence and timeline waits, readbacks, the queue lock's holders, DLSS/FSR/XeSS/menu recording of their own) waits for a "submitted up to part N" counter, bounded; everything else waits for nothing. Threads waiting for the recorder sleep instead of spinning. `RX_VK_RECORDER_PARTS=0` is the old hand-off, in the same build. Read first: `docs/dev/reference/threads.md` "The recorder / The hand-off by part"; `post_part`, `part_submit` and `queue_flush_for` (`gpu/vk/vk_replay.cpp`); `Recorder::loop`, `wait_until`, `settle_parts` and `QueueMutex` (`gpu/vk/vk_recorder.h`, `gpu/vk/vk_internal.h`). Two independent code reviews found and this fixes: a permanent hang with batched submits on RADV (a held last part with a waiter), unbounded new waits, a too-wide lock escape, a poll that defeated RADV batching, and spinning waits. Their last pass found no hang, deadlock or out-of-order submit left. ## Testing **Tested:** Linux, RTX 3090 (NVIDIA 620.x), Ryzen with 24 threads, 2560x1440: Castle Town Market standing, walking and turning, two game starts with the hand-off on and two with `RX_VK_RECORDER_PARTS=0`, plus a warm start of each; no Vulkan errors, no hang. The recorder's stats line: 28.5 hand-offs a frame (335 packets each) against ~1,100 (9 each); spun idle 0.16 s against 4.15 s in 10 s; blocked on the queue's lock 0.03 s in 10 s (longest 1.3 ms); the render thread waited for it 0.02 ms a frame. Renderer builds on Linux; Windows syntax check of the changed files with llvm-mingw. New game-free test `tests/recorder_parts_test.cpp` with 8 controls (compiled, not run here). Test Farm on f08644e3 against main, three areas, standing (first results; Intel, the Steam Deck and the integrated GPUs pending): no hang and no crash on any of them. Market: RTX 3080 Ti (Ryzen 9 5900X) +3.2%, RX 9060 XT (Ryzen 5 5600, Windows) +4%, RX 9070 XT (Ryzen 7 5800X3D, Windows) +0.7%, RTX 4070 (5800X3D) +0.5%, GTX 1070 (i7-8750H laptop) -2.8%, Steam Machine (Navi 33, AMD custom 6-core) -2.7% (Hyrule Field -7.4%). The two slower ones are power-limited 6-core CPUs: if they hold up, the recorder's Auto rule should leave such CPUs on the render thread alone (by measured render-thread load rather than core count), a follow-up. **Still to test:** - linux+amd, windows+amd, windows+nvidia, linux+nvidia, windows+intel, linux+intel: three areas against main, expecting equal or better fps and no hang (@Beanstalk) - The Steam Deck and 4-core PCs: unchanged (the recorder is off below 6 cores), expecting even (@Beanstalk) - Linux and Windows: `mise run test`, `test_recorder_parts.py`, `test_recorder_sync.py`, `test_recorder_own_idle.py` and the fixtures it touched, expecting them to pass (by hand: CI on this pull request) ## Measurements **Verdict:** a small, consistent gain on an RTX 3090 (+1.8% Market standing, better p99), and the recorder's wake-ups and spinning nearly gone; AMD numbers come from the Test Farm. Baseline: this PR's own build with `RX_VK_RECORDER_PARTS=0`, which is main's hand-off, so the runs differ only by this change. Setup: Linux, RTX 3090, 2560x1440, Threaded rendering Auto (render thread + recorder), Castle Town Market; two interleaved game starts each, after a warm start of each. | Market | Before: main's hand-off, this PR @ f08644e3 with `RX_VK_RECORDER_PARTS=0` | After: this PR @ f08644e3 | |---|---|---| | standing fps (runs) | 87.2 (86, 88) | 88.8 (88, 90) | | standing p99 ms | 14.4 | 13.4 | | walking fps | 82.6 | 83.5 | | turning fps | 99.6 | 100.2 | | recorder wake-ups a frame | ~1,100 | 28.5 | | recorder spinning idle, s in 10 s | 4.15 | 0.16 | `RX_VK_RECORDER_PARTS` ships on: never slower here, and it removes the recorder's churn on every CPU that runs it. ## Risks A wrong wait would show as a hang or a read of a part not yet on the GPU: every new wait is bounded and the reviews checked the lock order; `RX_VK_RECORDER_PARTS=0` restores the old hand-off. Threads that take the queue lock (texture uploads) now first wait for the parts handed over before them. Nothing changes for the player's files, and nothing changes where the recorder is off (below 6 cores, the Deck). ## Checklist ### Ready for review - [x] One topic, on current `main`; the title is a Conventional Commit. - [x] Tested on my own PC, in the game for a change the game runs, as Testing says. - [ ] `mise run test` passes (any failure that also fails on `main` is named). - [x] Measured against `main`, or Measurements says why there is nothing to measure. - [x] No game files, build output or hand-edited generated files; `third_party/` untouched; nothing GPL-incompatible. ### When it applies - [x] A switch added or changed: it ships at the default the measurements chose, with its row in `gpu/vk/switches.def`. - [x] A check under `tests/` that reads changed source is updated, not weakened. - [x] Host code builds on Windows (llvm-mingw) and Linux, with UTF-8 paths. - [x] Docs updated. - [x] AI assistance recorded with `Assisted-by:` / `Co-Authored-By:` trailers. ### Ready to merge - [ ] Everything in Still to test has run, its results in Tested. - [ ] Tier 1 hardware it can affect is tested (docs/dev/standards.md, "Hardware tiers"). - [ ] Approved in review, every comment answered. - [ ] CI passes on the head; on current `main`, measured again or a comment says why the numbers still hold. 🤖 Generated with [Claude Code](https://claude.com/claude-code) https://claude.ai/code/session_017vw8e3aEPBsd7QfUkaatjP
perf(vk): the recorder takes a whole part and submits it, so the render thread never waits at a part's end
Some checks reported errors
CI / Python source checks (ubuntu-24.04, part 1/2) (pull_request) Has been skipped
CI / Python source checks (ubuntu-24.04, part 2/2) (pull_request) Has been skipped
CI / SPIR-V fixtures (global narrow load/store, direct surface atomics, F2I immediate) (pull_request) Has been skipped
Third-party parts / The same as upstream (pull_request) Has been skipped
Website / Website on Cloudflare (pull_request) Has been skipped
CI / Workflow lint (actionlint) (pull_request) Successful in 13s
CI / Conventional Commit title (pull_request) Successful in 8s
Website / Website on Cloudflare (pull_request_target) Has been skipped
Third-party parts / As their notes say (pull_request) Successful in 14s
Guide screenshots / Guide images and local UI source review (pull_request) Failing after 12s
CI / Website build and checks (pull_request) Failing after 2m15s
PR description / Description follows the template (pull_request_target) Failing after 17s
beanstalk/linux+amd 0/1 pass, 0 running, 0 queued
beanstalk/approval approved by HumanBean (reviewed)
beanstalk/linux+nvidia 1/1 fail
beanstalk/windows+nvidia 0/1 pass, 1 error
beanstalk/windows+intel 0/1 pass, 0 running, 0 queued
beanstalk/steamdeck 0/1 pass, 0 running, 0 queued
beanstalk/windows+amd 1/1 fail
beanstalk/linux+intel 0/1 pass, 0 running, 0 queued
beanstalk/amd+igpu 0/1 pass, 0 running, 0 queued
f08644e37c
RX_VK_RECORDER_PARTS (on wherever the recorder thread runs; =0: the old drains, an A/B in one build).

- The render thread (replay_mu's holder) publishes a part's packets at the part's end, or in
  256 KB chunks of a long one, instead of packet by packet: one wake-up of the recorder a part.
  The recorder sleeps on the ring's counter between batches (WaitOnAddress/futex, else the
  condition variable) after a 10 us spin.
- submit_recording decides the part's submit on the render thread (timeline value, its number on
  the recordings' timeline, the Reflex frame) and maps the recording for the waiters, then hands
  its two vkEndCommandBuffer and its vkQueueSubmit2 (or its hold with RX_VK_BATCH) to the ring as
  one packet (rec::post_part -> part_submit). The render thread goes on into the next part at
  once; replay_mu's unlock publishes and no longer waits.
- What needs a part submitted waits for that, not for the ring to empty: a slot's reuse waits
  for its own part (recording_wait), then resets its pools at once (OwnSlot); a wait on the
  recordings' timeline waits for that recording's submit (wait_rec_seq); other fence and
  timeline waits and fence resets wait for every part handed over (parts_settle); query pool
  reads/resets and allocation invalidations wait for nothing. Threads without replay_mu wait at
  their boundary calls for what was published (foreign_wait).
- dev.present_mu and the presents' queue locks are a QueueMutex (vk_internal.h): taking
  present_mu first waits for the parts handed over (the queue's order), and a thread holding a
  queue lock never waits for a recorder that waits for it at a part's submit (counted, logged).
  The presenter thread, present_commit and Reflex's latency sleep are Settled: they need no part.
- Every wait for the parts is bounded and says whether it was met (wait_until's deadline): a
  recordings' wait not submitted in its time is a wait that timed out; a flush before a wait
  settles the parts (within the wait's time, polls not at all) before it looks at what is held,
  and the driver's wait gets only what is left of the time. A blocking wait that ends unmet
  sends what is held if the queue's lock is free and owes a flush to the part it waited for
  (flush_owed_upto: that part, once submitted, sends what is held; polls owe none), as
  RX_VK_BATCH would hang otherwise. Boundary settles are at most 250 ms; QueueMutex::try_lock
  does not settle.
- The escape is only for a thread the recorder waits for: it says it is blocked on present_mu
  only after a failed try, and a present_mu holder blocked on another queue lock says so; a
  slot's part unmet, or the recorder's stop under the lock, ends the process.
- A failed end or submit on the recorder goes to device_lost there. The GPU timeline diag stamps
  the submit where it is made.
- The stats line: batches a frame, packets a batch, parts submitted, the render thread's waits
  for the recorder (count, ms, a frame) apart from every thread's.
  A second line splits the recorder's busy time: parts' ends and submits, of that blocked on
  the queue's lock (count, longest), and the other packets; and packets a frame. In parts mode
  the waits for the recorder (drains, a thread off the replay lock, ring room) sleep on part_cv
  after the pauses instead of yielding.

Docs: threads.md, The hand-off by part. Tests: tests/test_recorder_parts.py (new, game-free:
the hand-off's order, the settle, the queue lock both ways, a slot's reset, with controls); the
drain checks keep RX_VK_RECORDER_PARTS off; fixtures and markers that cut the changed functions
updated. Not run here (no tests on this PC); not measured in the game.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_017vw8e3aEPBsd7QfUkaatjP
Author
Owner

Tested in game: Linux, RTX 3090 (NVIDIA 620.x), 2560x1440: Castle Town Market standing/walking/turning, two starts each with RX_VK_RECORDER_PARTS on and =0 plus warm starts; no Vulkan errors or hangs; recorder 28.5 hand-offs a frame vs ~1,100, spun idle 0.16 s vs 4.15 s per 10 s.

Tested in game: Linux, RTX 3090 (NVIDIA 620.x), 2560x1440: Castle Town Market standing/walking/turning, two starts each with RX_VK_RECORDER_PARTS on and =0 plus warm starts; no Vulkan errors or hangs; recorder 28.5 hand-offs a frame vs ~1,100, spun idle 0.16 s vs 4.15 s per 10 s.
Author
Owner

@Beanstalk Please test this PR against main on every platform: windows+amd, windows+nvidia, windows+intel, linux+amd, linux+nvidia, linux+intel, the steamdeck, amd+igpu. Settings vsync=2 (Off) and frame_rate_cap=6 (Uncapped), Threaded rendering Auto. From the kokiri-start save: Kokiri Forest, Hyrule Field and Castle Town Market, in each 20 s standing then 15 s turning the camera, two game starts a side; make sure each side has reached the area before measuring it. Measure fps, p99 frame time, stutters, crashes and hangs (a hang is the main risk: this changes which thread submits the GPU's work), and compare pictures with main. This only acts where the recorder thread runs (6+ cores); on the Steam Deck and 4-core PCs it should be even.

@Beanstalk Please test this PR against main on every platform: windows+amd, windows+nvidia, windows+intel, linux+amd, linux+nvidia, linux+intel, the steamdeck, amd+igpu. Settings vsync=2 (Off) and frame_rate_cap=6 (Uncapped), Threaded rendering Auto. From the `kokiri-start` save: Kokiri Forest, Hyrule Field and Castle Town Market, in each 20 s standing then 15 s turning the camera, two game starts a side; make sure each side has reached the area before measuring it. Measure fps, p99 frame time, stutters, crashes and hangs (a hang is the main risk: this changes which thread submits the GPU's work), and compare pictures with main. This only acts where the recorder thread runs (6+ cores); on the Steam Deck and 4-core PCs it should be even.
Member
Beanstalk test farm · an earlier run: commit f08644e (the newer results are in a comment below)

Commit f08644e · approved by HumanBean (reviewed) · dashboard

Started at 11:50 UTC.

Estimate: done in about 2 h 15 min, from the queue and the testers online now; some jobs can't start yet.

⚠ No tester has a PC for linux+intel (none seen in 14 days): it waits until one signs in.

Target Result Reports Frame time p99 (ms), main → PR VRAM peak (MB) Crashes
windows+amd ❌ fail (an outlier) 2 1 2 12.0 → 11.4 (−5%) — none
windows+nvidia ❌ fail 2 1 2 11.3 → 11.3 (+0%) — none
windows+intel ⏳ queued 0 of 1 — — —
linux+amd ❌ fail 1 1 22.2 → 48.0 (+117%) — none
linux+nvidia ❌ fail (an outlier) 2 1 2 17.6 → 17.6 (−0%) — none
linux+intel ⏳ queued 0 of 1 — — —
steamdeck ❌ fail 1 1 44.8 → 45.4 (+1%) — none
amd+igpu ❌ fail 1 1 43.9 → 43.9 (+0%) — none

Why it failed

  • windows+amd: three-areas kokiri-after-turn.fps_mean.min_delta_pct: 115.4 → 91.8 (−20.5%); limit 10.0
  • windows+amd: three-areas screenshot.max_diff: 0.7 → 0.7; limit 0.1
  • windows+amd: three-areas field-after-turn.stutter_count.max_delta_pct: 0.0 → 0.5; limit 20.0; the baseline is 0
  • windows+amd: three-areas market-standing.stutter_count.max_delta_pct: 0.0 → 1.5; limit 20.0; the baseline is 0
  • windows+amd: three-areas kokiri-after-turn.frametime_p99_ms.max_delta_pct: 11.1 → 13.7 (+23.4%); limit 15.0
  • windows+amd: three-areas market-after-turn.stutter_count.max_delta_pct: 0.0 → 0.5; limit 20.0; the baseline is 0
  • windows+nvidia: three-areas kokiri-after-turn.fps_mean.min_delta_pct: 109.7 → 98.6 (−10.1%); limit 10.0
  • windows+nvidia: three-areas market-after-turn.stutter_count.max_delta_pct: 0.0 → 1.0; limit 20.0; the baseline is 0
  • and 20 more, in the reports

Couldn't be judged: the tester's PC could not finish the test. The farm tries another PC, or the same one after a fix; it says nothing about the PR on its own.

  • windows+nvidia, report 1: the main base failed (exit 1): build.py stopped (exit status 1): fatal error: too many errors emitted, stopping now [-ferror-limit=]
What was tested

What will be tested · AI plan (gpt-6.1-terra)

Two starts per side compare the PR with main across the requested Windows and Linux GPU targets, Steam Deck, and AMD integrated graphics. Each start visits Kokiri Forest, Hyrule Field, and Market with V-Sync off and the frame rate uncapped; standing performance and pictures are compared, followed by camera turns and a second measurement. The farm cannot turn the camera concurrently with a measurement, so the 15-second measurements follow the turns rather than measuring the turns themselves. Crashes, failure to reach an area or present a frame, substantial FPS or p99 regressions, increased stutters, and picture differences fail the plan. Each target takes roughly 25 minutes; targets without a matching online PC will wait for one. No tester online now matches linux+intel: that target waits for one.

  • Where: windows+amd (1-2 reports), windows+nvidia (1-2 reports), windows+intel (1-2 reports), linux+amd (1-2 reports), linux+nvidia (1-2 reports), linux+intel (1-2 reports), steamdeck (1-2 reports), amd+igpu (1-2 reports).
  • three-areas: from the repository's save kokiri-start (settings vsync=2, frame_rate_cap=6), then teleport to Kokiri Forest, measure kokiri-standing for 20 s (after 5 s warm-up), take the picture kokiri to compare with main, move the stick, measure kokiri-after-turn for 15 s (after 0 s warm-up), teleport to Hyrule Field, measure field-standing for 20 s (after 5 s warm-up), take the picture field to compare with main, measure field-after-turn for 15 s (after 0 s warm-up), teleport to Market, measure market-standing for 20 s (after 5 s warm-up), take the picture market to compare with main, measure market-after-turn for 15 s (after 0 s warm-up); 2 runs on each side.
  • Measured: crashes, frame times, stutters and hitches, draws and frame rate, the game's log.
  • The PR fails if: it crashes; or frame rate is more than 10% below main; or a picture differs from main by more than 5%; or stutters is more than 20% above main; or frame time (p99) is more than 15% above main; or forbidden log lines is above 0.
  • Time: about 28 min of game runs per tester (4 game starts), plus the two builds.
Measurements on `windows+amd`, for the PR description

Verdict: fail on windows+amd (2/1 report): three-areas kokiri-after-turn.fps_mean.min_delta_pct (-20.5%); three-areas screenshot.max_diff; three-areas field-after-turn.stutter_count.max_delta_pct failed.

Hardware: 1: windows+amd+amd-proprietary+dgpu AMD Radeon RX 9070 XT, 2.0.416 AMD Ryzen 7 5800X3D 8-Core Processor, 32 GB RAM; 2: windows+amd+amd-proprietary+dgpu AMD Radeon RX 9060 XT, 2.0.416 AMD Ryzen 5 5600 6-Core Processor, 32 GB RAM
Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them
Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 2 game starts a side, kokiri-standing 20 s after 5 s warm-up, kokiri-after-turn 15 s after 0 s warm-up, field-standing 20 s after 5 s warm-up, field-after-turn 15 s after 0 s warm-up, market-standing 20 s after 5 s warm-up, market-after-turn 15 s after 0 s warm-up; each number the median of the runs, then of 2 reports
How to reproduce: the farm's plan above (@Beanstalk on this PR runs it again)

Before: main @ 89c3d21 After: this PR @ f08644e
three-areas · field-standing · frame rate 108.2 109.0 (+1%)
three-areas · kokiri-standing · frame rate 107.4 105.9 (−1%)
three-areas · market-standing · frame rate 66.4 68.0 (+2%)
three-areas · field-after-turn · frame rate 102.7 102.3 (−0%)
three-areas · kokiri-after-turn · frame rate 105.7 95.9 (−9%)
three-areas · market-after-turn · frame rate 73.4 81.9 (+12%)
three-areas · field-standing · stutters 0.2 0.0 (−100%)
three-areas · kokiri-standing · stutters 2.8 2.0 (−27%)
three-areas · market-standing · stutters 0.2 1.0 (+300%)
three-areas · field-after-turn · stutters 0.0 0.2
three-areas · kokiri-after-turn · stutters 0.2 0.0 (−100%)
three-areas · market-after-turn · stutters 0.2 0.2 (+0%)
three-areas · field-standing · frame time (p99) (ms) 12.0 11.4 (−5%)
three-areas · kokiri-standing · frame time (p99) (ms) 12.7 11.7 (−8%)
three-areas · market-standing · frame time (p99) (ms) 20.1 19.2 (−5%)
three-areas · field-after-turn · frame time (p99) (ms) 12.6 12.4 (−1%)
three-areas · kokiri-after-turn · frame time (p99) (ms) 12.0 13.1 (+9%)
three-areas · market-after-turn · frame time (p99) (ms) 18.2 16.7 (−8%)
three-areas · forbidden log lines 0.0 0.0

Flags and settings

None: the plan varied no setting, switch or build option.

As Markdown for the PR description:

## Measurements

**Verdict:** fail on windows+amd (2/1 report): `three-areas kokiri-after-turn.fps_mean.min_delta_pct (-20.5%)`; `three-areas screenshot.max_diff`; `three-areas field-after-turn.stutter_count.max_delta_pct` failed.

Hardware: 1: windows+amd+amd-proprietary+dgpu `AMD Radeon RX 9070 XT, 2.0.416` `AMD Ryzen 7 5800X3D 8-Core Processor, 32 GB RAM`; 2: windows+amd+amd-proprietary+dgpu `AMD Radeon RX 9060 XT, 2.0.416` `AMD Ryzen 5 5600 6-Core Processor, 32 GB RAM`  
Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them  
Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 2 game starts a side, kokiri-standing 20 s after 5 s warm-up, kokiri-after-turn 15 s after 0 s warm-up, field-standing 20 s after 5 s warm-up, field-after-turn 15 s after 0 s warm-up, market-standing 20 s after 5 s warm-up, market-after-turn 15 s after 0 s warm-up; each number the median of the runs, then of 2 reports  
How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again)

| | Before: main @ 89c3d21 | After: this PR @ f08644e |
|---|---|---|
| three-areas · field-standing · frame rate | 108.2 | 109.0 (+1%) |
| three-areas · kokiri-standing · frame rate | 107.4 | 105.9 (−1%) |
| three-areas · market-standing · frame rate | 66.4 | 68.0 (+2%) |
| three-areas · field-after-turn · frame rate | 102.7 | 102.3 (−0%) |
| three-areas · kokiri-after-turn · frame rate | 105.7 | 95.9 (−9%) |
| three-areas · market-after-turn · frame rate | 73.4 | 81.9 (+12%) |
| three-areas · field-standing · stutters | 0.2 | 0.0 (−100%) |
| three-areas · kokiri-standing · stutters | 2.8 | 2.0 (−27%) |
| three-areas · market-standing · stutters | 0.2 | 1.0 (+300%) |
| three-areas · field-after-turn · stutters | 0.0 | 0.2 |
| three-areas · kokiri-after-turn · stutters | 0.2 | 0.0 (−100%) |
| three-areas · market-after-turn · stutters | 0.2 | 0.2 (+0%) |
| three-areas · field-standing · frame time (p99) (ms) | 12.0 | 11.4 (−5%) |
| three-areas · kokiri-standing · frame time (p99) (ms) | 12.7 | 11.7 (−8%) |
| three-areas · market-standing · frame time (p99) (ms) | 20.1 | 19.2 (−5%) |
| three-areas · field-after-turn · frame time (p99) (ms) | 12.6 | 12.4 (−1%) |
| three-areas · kokiri-after-turn · frame time (p99) (ms) | 12.0 | 13.1 (+9%) |
| three-areas · market-after-turn · frame time (p99) (ms) | 18.2 | 16.7 (−8%) |
| three-areas · forbidden log lines | 0.0 | 0.0 |

**Flags and settings**

None: the plan varied no setting, switch or build option.
Measurements on `windows+nvidia`, for the PR description

Verdict: fail on windows+nvidia (2/1 report): three-areas kokiri-after-turn.fps_mean.min_delta_pct (-10.1%); three-areas market-after-turn.stutter_count.max_delta_pct failed.

Hardware: 1: windows+nvidia+nvidia-proprietary+dgpu NVIDIA GeForce RTX 4080, 621.1.0 AMD Ryzen 7 9800X3D 8-Core Processor, 32 GB RAM; 2: windows+nvidia+nvidia-proprietary+dgpu NVIDIA GeForce RTX 3080 Ti, 621.24.0 AMD Ryzen 9 5900X 12-Core Processor, 64 GB RAM
Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them
Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 2 game starts a side, kokiri-standing 20 s after 5 s warm-up, kokiri-after-turn 15 s after 0 s warm-up, field-standing 20 s after 5 s warm-up, field-after-turn 15 s after 0 s warm-up, market-standing 20 s after 5 s warm-up, market-after-turn 15 s after 0 s warm-up; each number the median of the runs, then of 2 reports
How to reproduce: the farm's plan above (@Beanstalk on this PR runs it again)

Before: main @ 89c3d21 After: this PR @ f08644e
three-areas · field-standing · frame rate 108.1 108.4 (+0%)
three-areas · kokiri-standing · frame rate 108.9 101.2 (−7%)
three-areas · market-standing · frame rate 80.0 82.5 (+3%)
three-areas · field-after-turn · frame rate 117.2 115.5 (−1%)
three-areas · kokiri-after-turn · frame rate 109.7 98.6 (−10%)
three-areas · market-after-turn · frame rate 81.1 83.5 (+3%)
three-areas · field-standing · stutters 0.0 0.0
three-areas · kokiri-standing · stutters 0.0 0.0
three-areas · market-standing · stutters 0.5 0.0 (−100%)
three-areas · field-after-turn · stutters 0.0 0.0
three-areas · kokiri-after-turn · stutters 0.5 0.0 (−100%)
three-areas · market-after-turn · stutters 0.0 1.0
three-areas · field-standing · frame time (p99) (ms) 11.3 11.3 (+0%)
three-areas · kokiri-standing · frame time (p99) (ms) 11.1 11.4 (+3%)
three-areas · market-standing · frame time (p99) (ms) 15.4 14.8 (−4%)
three-areas · field-after-turn · frame time (p99) (ms) 10.3 10.9 (+6%)
three-areas · kokiri-after-turn · frame time (p99) (ms) 11.9 12.2 (+2%)
three-areas · market-after-turn · frame time (p99) (ms) 15.2 15.7 (+3%)
three-areas · forbidden log lines 0.0 0.0

Flags and settings

None: the plan varied no setting, switch or build option.

As Markdown for the PR description:

## Measurements

**Verdict:** fail on windows+nvidia (2/1 report): `three-areas kokiri-after-turn.fps_mean.min_delta_pct (-10.1%)`; `three-areas market-after-turn.stutter_count.max_delta_pct` failed.

Hardware: 1: windows+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce RTX 4080, 621.1.0` `AMD Ryzen 7 9800X3D 8-Core Processor, 32 GB RAM`; 2: windows+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce RTX 3080 Ti, 621.24.0` `AMD Ryzen 9 5900X 12-Core Processor, 64 GB RAM`  
Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them  
Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 2 game starts a side, kokiri-standing 20 s after 5 s warm-up, kokiri-after-turn 15 s after 0 s warm-up, field-standing 20 s after 5 s warm-up, field-after-turn 15 s after 0 s warm-up, market-standing 20 s after 5 s warm-up, market-after-turn 15 s after 0 s warm-up; each number the median of the runs, then of 2 reports  
How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again)

| | Before: main @ 89c3d21 | After: this PR @ f08644e |
|---|---|---|
| three-areas · field-standing · frame rate | 108.1 | 108.4 (+0%) |
| three-areas · kokiri-standing · frame rate | 108.9 | 101.2 (−7%) |
| three-areas · market-standing · frame rate | 80.0 | 82.5 (+3%) |
| three-areas · field-after-turn · frame rate | 117.2 | 115.5 (−1%) |
| three-areas · kokiri-after-turn · frame rate | 109.7 | 98.6 (−10%) |
| three-areas · market-after-turn · frame rate | 81.1 | 83.5 (+3%) |
| three-areas · field-standing · stutters | 0.0 | 0.0 |
| three-areas · kokiri-standing · stutters | 0.0 | 0.0 |
| three-areas · market-standing · stutters | 0.5 | 0.0 (−100%) |
| three-areas · field-after-turn · stutters | 0.0 | 0.0 |
| three-areas · kokiri-after-turn · stutters | 0.5 | 0.0 (−100%) |
| three-areas · market-after-turn · stutters | 0.0 | 1.0 |
| three-areas · field-standing · frame time (p99) (ms) | 11.3 | 11.3 (+0%) |
| three-areas · kokiri-standing · frame time (p99) (ms) | 11.1 | 11.4 (+3%) |
| three-areas · market-standing · frame time (p99) (ms) | 15.4 | 14.8 (−4%) |
| three-areas · field-after-turn · frame time (p99) (ms) | 10.3 | 10.9 (+6%) |
| three-areas · kokiri-after-turn · frame time (p99) (ms) | 11.9 | 12.2 (+2%) |
| three-areas · market-after-turn · frame time (p99) (ms) | 15.2 | 15.7 (+3%) |
| three-areas · forbidden log lines | 0.0 | 0.0 |

**Flags and settings**

None: the plan varied no setting, switch or build option.
Measurements on `linux+amd`, for the PR description

Verdict: fail on linux+amd (1/1 report): three-areas kokiri-after-turn.fps_mean.min_delta_pct (-10.2%); three-areas field-after-turn.fps_mean.min_delta_pct (-13.8%); three-areas market-after-turn.fps_mean.min_delta_pct (-36.0%) failed.

Hardware: 1: linux+amd+radv+dgpu AMD Steam Machine (RADV NAVI33), 26.1.99 AMD Custom CPU 1772, 16 GB RAM
Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them
Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 2 game starts a side, kokiri-standing 20 s after 5 s warm-up, kokiri-after-turn 15 s after 0 s warm-up, field-standing 20 s after 5 s warm-up, field-after-turn 15 s after 0 s warm-up, market-standing 20 s after 5 s warm-up, market-after-turn 15 s after 0 s warm-up; each number the median of the runs, then of 1 report
How to reproduce: the farm's plan above (@Beanstalk on this PR runs it again)

Before: main @ 89c3d21 After: this PR @ f08644e
three-areas · field-standing · frame rate 50.0 46.3 (−7%)
three-areas · kokiri-standing · frame rate 51.7 53.4 (+3%)
three-areas · market-standing · frame rate 19.8 19.2 (−3%)
three-areas · field-after-turn · frame rate 54.3 46.8 (−14%)
three-areas · kokiri-after-turn · frame rate 58.8 52.9 (−10%)
three-areas · market-after-turn · frame rate 64.8 41.5 (−36%)
three-areas · field-standing · stutters 5.0 24.5 (+390%)
three-areas · kokiri-standing · stutters 0.5 1.0 (+100%)
three-areas · market-standing · stutters 179.5 86.0 (−52%)
three-areas · field-after-turn · stutters 14.5 47.5 (+228%)
three-areas · kokiri-after-turn · stutters 1.0 0.0 (−100%)
three-areas · market-after-turn · stutters 1.5 60.0 (+3900%)
three-areas · field-standing · frame time (p99) (ms) 22.2 48.0 (+117%)
three-areas · kokiri-standing · frame time (p99) (ms) 21.2 21.4 (+1%)
three-areas · market-standing · frame time (p99) (ms) 84.6 87.6 (+4%)
three-areas · field-after-turn · frame time (p99) (ms) 63.8 70.5 (+11%)
three-areas · kokiri-after-turn · frame time (p99) (ms) 19.7 21.4 (+9%)
three-areas · market-after-turn · frame time (p99) (ms) 21.2 53.0 (+150%)
three-areas · forbidden log lines 0.0 0.0

Flags and settings

None: the plan varied no setting, switch or build option.

As Markdown for the PR description:

## Measurements

**Verdict:** fail on linux+amd (1/1 report): `three-areas kokiri-after-turn.fps_mean.min_delta_pct (-10.2%)`; `three-areas field-after-turn.fps_mean.min_delta_pct (-13.8%)`; `three-areas market-after-turn.fps_mean.min_delta_pct (-36.0%)` failed.

Hardware: 1: linux+amd+radv+dgpu `AMD Steam Machine (RADV NAVI33), 26.1.99` `AMD Custom CPU 1772, 16 GB RAM`  
Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them  
Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 2 game starts a side, kokiri-standing 20 s after 5 s warm-up, kokiri-after-turn 15 s after 0 s warm-up, field-standing 20 s after 5 s warm-up, field-after-turn 15 s after 0 s warm-up, market-standing 20 s after 5 s warm-up, market-after-turn 15 s after 0 s warm-up; each number the median of the runs, then of 1 report  
How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again)

| | Before: main @ 89c3d21 | After: this PR @ f08644e |
|---|---|---|
| three-areas · field-standing · frame rate | 50.0 | 46.3 (−7%) |
| three-areas · kokiri-standing · frame rate | 51.7 | 53.4 (+3%) |
| three-areas · market-standing · frame rate | 19.8 | 19.2 (−3%) |
| three-areas · field-after-turn · frame rate | 54.3 | 46.8 (−14%) |
| three-areas · kokiri-after-turn · frame rate | 58.8 | 52.9 (−10%) |
| three-areas · market-after-turn · frame rate | 64.8 | 41.5 (−36%) |
| three-areas · field-standing · stutters | 5.0 | 24.5 (+390%) |
| three-areas · kokiri-standing · stutters | 0.5 | 1.0 (+100%) |
| three-areas · market-standing · stutters | 179.5 | 86.0 (−52%) |
| three-areas · field-after-turn · stutters | 14.5 | 47.5 (+228%) |
| three-areas · kokiri-after-turn · stutters | 1.0 | 0.0 (−100%) |
| three-areas · market-after-turn · stutters | 1.5 | 60.0 (+3900%) |
| three-areas · field-standing · frame time (p99) (ms) | 22.2 | 48.0 (+117%) |
| three-areas · kokiri-standing · frame time (p99) (ms) | 21.2 | 21.4 (+1%) |
| three-areas · market-standing · frame time (p99) (ms) | 84.6 | 87.6 (+4%) |
| three-areas · field-after-turn · frame time (p99) (ms) | 63.8 | 70.5 (+11%) |
| three-areas · kokiri-after-turn · frame time (p99) (ms) | 19.7 | 21.4 (+9%) |
| three-areas · market-after-turn · frame time (p99) (ms) | 21.2 | 53.0 (+150%) |
| three-areas · forbidden log lines | 0.0 | 0.0 |

**Flags and settings**

None: the plan varied no setting, switch or build option.
Measurements on `linux+nvidia`, for the PR description

Verdict: fail on linux+nvidia (2/1 report): three-areas kokiri-standing.stutter_count.max_delta_pct (+325.0%); three-areas field-standing.stutter_count.max_delta_pct (+400.0%); three-areas field-after-turn.stutter_count.max_delta_pct (+200.0%) failed.

Hardware: 1: linux+nvidia+nvidia-proprietary+dgpu NVIDIA GeForce RTX 4070, 615.71.9 AMD Ryzen 7 5800X3D 8-Core Processor, 64 GB RAM; 2: linux+nvidia+nvidia-proprietary+dgpu NVIDIA GeForce GTX 1070, 580.196.7 Intel(R) Core(TM) i7-8750H CPU @ 2.20GHz, 32 GB RAM
Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them
Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 2 game starts a side, kokiri-standing 20 s after 5 s warm-up, kokiri-after-turn 15 s after 0 s warm-up, field-standing 20 s after 5 s warm-up, field-after-turn 15 s after 0 s warm-up, market-standing 20 s after 5 s warm-up, market-after-turn 15 s after 0 s warm-up; each number the median of the runs, then of 2 reports
How to reproduce: the farm's plan above (@Beanstalk on this PR runs it again)

Before: main @ 89c3d21 After: this PR @ f08644e
three-areas · field-standing · frame rate 81.2 81.3 (+0%)
three-areas · kokiri-standing · frame rate 84.3 84.3 (−0%)
three-areas · market-standing · frame rate 63.5 63.2 (−1%)
three-areas · field-after-turn · frame rate 90.0 91.2 (+1%)
three-areas · kokiri-after-turn · frame rate 86.6 83.9 (−3%)
three-areas · market-after-turn · frame rate 80.4 89.5 (+11%)
three-areas · field-standing · stutters 0.2 1.2 (+400%)
three-areas · kokiri-standing · stutters 1.0 4.5 (+350%)
three-areas · market-standing · stutters 1.8 1.5 (−14%)
three-areas · field-after-turn · stutters 0.5 1.5 (+200%)
three-areas · kokiri-after-turn · stutters 1.0 0.0 (−100%)
three-areas · market-after-turn · stutters 1.8 2.0 (+14%)
three-areas · field-standing · frame time (p99) (ms) 17.6 17.6 (−0%)
three-areas · kokiri-standing · frame time (p99) (ms) 18.0 17.3 (−4%)
three-areas · market-standing · frame time (p99) (ms) 23.6 23.4 (−1%)
three-areas · field-after-turn · frame time (p99) (ms) 16.0 16.5 (+3%)
three-areas · kokiri-after-turn · frame time (p99) (ms) 16.3 17.0 (+4%)
three-areas · market-after-turn · frame time (p99) (ms) 19.4 19.3 (−1%)
three-areas · forbidden log lines 0.0 0.0

Flags and settings

None: the plan varied no setting, switch or build option.

As Markdown for the PR description:

## Measurements

**Verdict:** fail on linux+nvidia (2/1 report): `three-areas kokiri-standing.stutter_count.max_delta_pct (+325.0%)`; `three-areas field-standing.stutter_count.max_delta_pct (+400.0%)`; `three-areas field-after-turn.stutter_count.max_delta_pct (+200.0%)` failed.

Hardware: 1: linux+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce RTX 4070, 615.71.9` `AMD Ryzen 7 5800X3D 8-Core Processor, 64 GB RAM`; 2: linux+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce GTX 1070, 580.196.7` `Intel(R) Core(TM) i7-8750H CPU @ 2.20GHz, 32 GB RAM`  
Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them  
Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 2 game starts a side, kokiri-standing 20 s after 5 s warm-up, kokiri-after-turn 15 s after 0 s warm-up, field-standing 20 s after 5 s warm-up, field-after-turn 15 s after 0 s warm-up, market-standing 20 s after 5 s warm-up, market-after-turn 15 s after 0 s warm-up; each number the median of the runs, then of 2 reports  
How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again)

| | Before: main @ 89c3d21 | After: this PR @ f08644e |
|---|---|---|
| three-areas · field-standing · frame rate | 81.2 | 81.3 (+0%) |
| three-areas · kokiri-standing · frame rate | 84.3 | 84.3 (−0%) |
| three-areas · market-standing · frame rate | 63.5 | 63.2 (−1%) |
| three-areas · field-after-turn · frame rate | 90.0 | 91.2 (+1%) |
| three-areas · kokiri-after-turn · frame rate | 86.6 | 83.9 (−3%) |
| three-areas · market-after-turn · frame rate | 80.4 | 89.5 (+11%) |
| three-areas · field-standing · stutters | 0.2 | 1.2 (+400%) |
| three-areas · kokiri-standing · stutters | 1.0 | 4.5 (+350%) |
| three-areas · market-standing · stutters | 1.8 | 1.5 (−14%) |
| three-areas · field-after-turn · stutters | 0.5 | 1.5 (+200%) |
| three-areas · kokiri-after-turn · stutters | 1.0 | 0.0 (−100%) |
| three-areas · market-after-turn · stutters | 1.8 | 2.0 (+14%) |
| three-areas · field-standing · frame time (p99) (ms) | 17.6 | 17.6 (−0%) |
| three-areas · kokiri-standing · frame time (p99) (ms) | 18.0 | 17.3 (−4%) |
| three-areas · market-standing · frame time (p99) (ms) | 23.6 | 23.4 (−1%) |
| three-areas · field-after-turn · frame time (p99) (ms) | 16.0 | 16.5 (+3%) |
| three-areas · kokiri-after-turn · frame time (p99) (ms) | 16.3 | 17.0 (+4%) |
| three-areas · market-after-turn · frame time (p99) (ms) | 19.4 | 19.3 (−1%) |
| three-areas · forbidden log lines | 0.0 | 0.0 |

**Flags and settings**

None: the plan varied no setting, switch or build option.
Measurements on `steamdeck`, for the PR description

Verdict: fail on steamdeck (1/1 report): three-areas market-standing.stutter_count.max_delta_pct (+50.0%); three-areas market-after-turn.stutter_count.max_delta_pct (+88.9%); three-areas field-after-turn.frametime_p99_ms.max_delta_pct (+19.4%) failed.

Hardware: 1: linux+amd+radv+igpu+handheld AMD Custom GPU 0405 (RADV VANGOGH), 26.1.99 AMD Custom APU 0405, 12 GB RAM
Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them
Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 2 game starts a side, kokiri-standing 20 s after 5 s warm-up, kokiri-after-turn 15 s after 0 s warm-up, field-standing 20 s after 5 s warm-up, field-after-turn 15 s after 0 s warm-up, market-standing 20 s after 5 s warm-up, market-after-turn 15 s after 0 s warm-up; each number the median of the runs, then of 1 report
How to reproduce: the farm's plan above (@Beanstalk on this PR runs it again)

Before: main @ 89c3d21 After: this PR @ f08644e
three-areas · field-standing · frame rate 23.9 23.6 (−1%)
three-areas · kokiri-standing · frame rate 24.2 23.8 (−2%)
three-areas · market-standing · frame rate 16.5 16.0 (−3%)
three-areas · field-after-turn · frame rate 27.7 25.2 (−9%)
three-areas · kokiri-after-turn · frame rate 27.0 26.4 (−2%)
three-areas · market-after-turn · frame rate 17.0 19.7 (+15%)
three-areas · field-standing · stutters 0.0 0.0
three-areas · kokiri-standing · stutters 0.5 0.5 (+0%)
three-areas · market-standing · stutters 8.0 12.0 (+50%)
three-areas · field-after-turn · stutters 0.0 0.0
three-areas · kokiri-after-turn · stutters 0.0 0.0
three-areas · market-after-turn · stutters 4.5 8.5 (+89%)
three-areas · field-standing · frame time (p99) (ms) 44.8 45.4 (+1%)
three-areas · kokiri-standing · frame time (p99) (ms) 54.1 55.6 (+3%)
three-areas · market-standing · frame time (p99) (ms) 143.7 150.7 (+5%)
three-areas · field-after-turn · frame time (p99) (ms) 39.3 47.0 (+19%)
three-areas · kokiri-after-turn · frame time (p99) (ms) 41.5 43.0 (+4%)
three-areas · market-after-turn · frame time (p99) (ms) 122.8 119.1 (−3%)
three-areas · forbidden log lines 0.0 0.0

Flags and settings

None: the plan varied no setting, switch or build option.

As Markdown for the PR description:

## Measurements

**Verdict:** fail on steamdeck (1/1 report): `three-areas market-standing.stutter_count.max_delta_pct (+50.0%)`; `three-areas market-after-turn.stutter_count.max_delta_pct (+88.9%)`; `three-areas field-after-turn.frametime_p99_ms.max_delta_pct (+19.4%)` failed.

Hardware: 1: linux+amd+radv+igpu+handheld `AMD Custom GPU 0405 (RADV VANGOGH), 26.1.99` `AMD Custom APU 0405, 12 GB RAM`  
Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them  
Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 2 game starts a side, kokiri-standing 20 s after 5 s warm-up, kokiri-after-turn 15 s after 0 s warm-up, field-standing 20 s after 5 s warm-up, field-after-turn 15 s after 0 s warm-up, market-standing 20 s after 5 s warm-up, market-after-turn 15 s after 0 s warm-up; each number the median of the runs, then of 1 report  
How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again)

| | Before: main @ 89c3d21 | After: this PR @ f08644e |
|---|---|---|
| three-areas · field-standing · frame rate | 23.9 | 23.6 (−1%) |
| three-areas · kokiri-standing · frame rate | 24.2 | 23.8 (−2%) |
| three-areas · market-standing · frame rate | 16.5 | 16.0 (−3%) |
| three-areas · field-after-turn · frame rate | 27.7 | 25.2 (−9%) |
| three-areas · kokiri-after-turn · frame rate | 27.0 | 26.4 (−2%) |
| three-areas · market-after-turn · frame rate | 17.0 | 19.7 (+15%) |
| three-areas · field-standing · stutters | 0.0 | 0.0 |
| three-areas · kokiri-standing · stutters | 0.5 | 0.5 (+0%) |
| three-areas · market-standing · stutters | 8.0 | 12.0 (+50%) |
| three-areas · field-after-turn · stutters | 0.0 | 0.0 |
| three-areas · kokiri-after-turn · stutters | 0.0 | 0.0 |
| three-areas · market-after-turn · stutters | 4.5 | 8.5 (+89%) |
| three-areas · field-standing · frame time (p99) (ms) | 44.8 | 45.4 (+1%) |
| three-areas · kokiri-standing · frame time (p99) (ms) | 54.1 | 55.6 (+3%) |
| three-areas · market-standing · frame time (p99) (ms) | 143.7 | 150.7 (+5%) |
| three-areas · field-after-turn · frame time (p99) (ms) | 39.3 | 47.0 (+19%) |
| three-areas · kokiri-after-turn · frame time (p99) (ms) | 41.5 | 43.0 (+4%) |
| three-areas · market-after-turn · frame time (p99) (ms) | 122.8 | 119.1 (−3%) |
| three-areas · forbidden log lines | 0.0 | 0.0 |

**Flags and settings**

None: the plan varied no setting, switch or build option.
Measurements on `amd+igpu`, for the PR description

Verdict: fail on amd+igpu (1/1 report): three-areas screenshot.max_diff; three-areas kokiri-standing.stutter_count.max_delta_pct failed.

Hardware: 1: linux+amd+radv+igpu+handheld AMD Custom GPU 0405 (RADV VANGOGH), 26.1.99 AMD Custom APU 0405, 12 GB RAM
Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them
Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 2 game starts a side, kokiri-standing 20 s after 5 s warm-up, kokiri-after-turn 15 s after 0 s warm-up, field-standing 20 s after 5 s warm-up, field-after-turn 15 s after 0 s warm-up, market-standing 20 s after 5 s warm-up, market-after-turn 15 s after 0 s warm-up; each number the median of the runs, then of 1 report
How to reproduce: the farm's plan above (@Beanstalk on this PR runs it again)

Before: main @ 89c3d21 After: this PR @ f08644e
three-areas · field-standing · frame rate 24.3 24.4 (+0%)
three-areas · kokiri-standing · frame rate 25.5 24.5 (−4%)
three-areas · market-standing · frame rate 17.6 18.1 (+3%)
three-areas · field-after-turn · frame rate 27.5 27.3 (−1%)
three-areas · kokiri-after-turn · frame rate 24.6 26.4 (+7%)
three-areas · market-after-turn · frame rate 19.4 20.7 (+6%)
three-areas · field-standing · stutters 0.0 0.0
three-areas · kokiri-standing · stutters 0.0 0.5
three-areas · market-standing · stutters 5.5 5.0 (−9%)
three-areas · field-after-turn · stutters 0.0 0.0
three-areas · kokiri-after-turn · stutters 0.0 0.0
three-areas · market-after-turn · stutters 8.5 7.0 (−18%)
three-areas · field-standing · frame time (p99) (ms) 43.9 43.9 (+0%)
three-areas · kokiri-standing · frame time (p99) (ms) 47.9 50.6 (+6%)
three-areas · market-standing · frame time (p99) (ms) 117.2 108.1 (−8%)
three-areas · field-after-turn · frame time (p99) (ms) 39.3 39.7 (+1%)
three-areas · kokiri-after-turn · frame time (p99) (ms) 46.5 42.1 (−10%)
three-areas · market-after-turn · frame time (p99) (ms) 129.8 109.3 (−16%)
three-areas · forbidden log lines 0.0 0.0

Flags and settings

None: the plan varied no setting, switch or build option.

As Markdown for the PR description:

## Measurements

**Verdict:** fail on amd+igpu (1/1 report): `three-areas screenshot.max_diff`; `three-areas kokiri-standing.stutter_count.max_delta_pct` failed.

Hardware: 1: linux+amd+radv+igpu+handheld `AMD Custom GPU 0405 (RADV VANGOGH), 26.1.99` `AMD Custom APU 0405, 12 GB RAM`  
Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them  
Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 2 game starts a side, kokiri-standing 20 s after 5 s warm-up, kokiri-after-turn 15 s after 0 s warm-up, field-standing 20 s after 5 s warm-up, field-after-turn 15 s after 0 s warm-up, market-standing 20 s after 5 s warm-up, market-after-turn 15 s after 0 s warm-up; each number the median of the runs, then of 1 report  
How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again)

| | Before: main @ 89c3d21 | After: this PR @ f08644e |
|---|---|---|
| three-areas · field-standing · frame rate | 24.3 | 24.4 (+0%) |
| three-areas · kokiri-standing · frame rate | 25.5 | 24.5 (−4%) |
| three-areas · market-standing · frame rate | 17.6 | 18.1 (+3%) |
| three-areas · field-after-turn · frame rate | 27.5 | 27.3 (−1%) |
| three-areas · kokiri-after-turn · frame rate | 24.6 | 26.4 (+7%) |
| three-areas · market-after-turn · frame rate | 19.4 | 20.7 (+6%) |
| three-areas · field-standing · stutters | 0.0 | 0.0 |
| three-areas · kokiri-standing · stutters | 0.0 | 0.5 |
| three-areas · market-standing · stutters | 5.5 | 5.0 (−9%) |
| three-areas · field-after-turn · stutters | 0.0 | 0.0 |
| three-areas · kokiri-after-turn · stutters | 0.0 | 0.0 |
| three-areas · market-after-turn · stutters | 8.5 | 7.0 (−18%) |
| three-areas · field-standing · frame time (p99) (ms) | 43.9 | 43.9 (+0%) |
| three-areas · kokiri-standing · frame time (p99) (ms) | 47.9 | 50.6 (+6%) |
| three-areas · market-standing · frame time (p99) (ms) | 117.2 | 108.1 (−8%) |
| three-areas · field-after-turn · frame time (p99) (ms) | 39.3 | 39.7 (+1%) |
| three-areas · kokiri-after-turn · frame time (p99) (ms) | 46.5 | 42.1 (−10%) |
| three-areas · market-after-turn · frame time (p99) (ms) | 129.8 | 109.3 (−16%) |
| three-areas · forbidden log lines | 0.0 | 0.0 |

**Flags and settings**

None: the plan varied no setting, switch or build option.
The testers' hardware
Target Report Ran on
windows+amd 1 windows+amd+amd-proprietary+dgpu AMD Radeon RX 9070 XT, 2.0.416 AMD Ryzen 7 5800X3D 8-Core Processor, 32 GB RAM
windows+amd 2 windows+amd+amd-proprietary+dgpu AMD Radeon RX 9060 XT, 2.0.416 AMD Ryzen 5 5600 6-Core Processor, 32 GB RAM
windows+nvidia 1 windows+nvidia+nvidia-proprietary+dgpu NVIDIA GeForce RTX 4080, 621.1.0 AMD Ryzen 7 9800X3D 8-Core Processor, 32 GB RAM
windows+nvidia 2 windows+nvidia+nvidia-proprietary+dgpu NVIDIA GeForce RTX 3080 Ti, 621.24.0 AMD Ryzen 9 5900X 12-Core Processor, 64 GB RAM
linux+amd 1 linux+amd+radv+dgpu AMD Steam Machine (RADV NAVI33), 26.1.99 AMD Custom CPU 1772, 16 GB RAM
linux+nvidia 1 linux+nvidia+nvidia-proprietary+dgpu NVIDIA GeForce RTX 4070, 615.71.9 AMD Ryzen 7 5800X3D 8-Core Processor, 64 GB RAM
linux+nvidia 2 linux+nvidia+nvidia-proprietary+dgpu NVIDIA GeForce GTX 1070, 580.196.7 Intel(R) Core(TM) i7-8750H CPU @ 2.20GHz, 32 GB RAM
steamdeck 1 linux+amd+radv+igpu+handheld AMD Custom GPU 0405 (RADV VANGOGH), 26.1.99 AMD Custom APU 0405, 12 GB RAM
amd+igpu 1 linux+amd+radv+igpu+handheld AMD Custom GPU 0405 (RADV VANGOGH), 26.1.99 AMD Custom APU 0405, 12 GB RAM
<!-- beanstalk:previous --> <details><summary>Beanstalk test farm · an earlier run: commit <code>f08644e</code> (the newer results are in a comment below)</summary> Commit `f08644e` · approved by HumanBean (reviewed) · [dashboard](https://beanstalk.magicbean.dev/prs/1866) Started at 11:50 UTC. **Estimate:** done in about 2 h 15 min, from the queue and the testers online now; some jobs can't start yet. ⚠ No tester has a PC for `linux+intel` (none seen in 14 days): it waits until one signs in. | Target | Result | Reports | Frame time p99 (ms), main → PR | VRAM peak (MB) | Crashes | | --- | --- | --- | --- | --- | --- | | `windows+amd` | ❌ fail (an outlier) | 2 [1](https://beanstalk.magicbean.dev/reports/rp_18b65d3e2212c214) [2](https://beanstalk.magicbean.dev/reports/rp_28cd9c2a5d5a15aa) | 12.0 → 11.4 (−5%) | — | none | | `windows+nvidia` | ❌ fail | 2 [1](https://beanstalk.magicbean.dev/reports/rp_ff5f883c9bb9d420) [2](https://beanstalk.magicbean.dev/reports/rp_24abd5b090f1f8b3) | 11.3 → 11.3 (+0%) | — | none | | `windows+intel` | ⏳ queued | 0 of 1 | — | — | — | | `linux+amd` | ❌ fail | 1 [1](https://beanstalk.magicbean.dev/reports/rp_69e22fd96ba9a4f5) | 22.2 → 48.0 (+117%) | — | none | | `linux+nvidia` | ❌ fail (an outlier) | 2 [1](https://beanstalk.magicbean.dev/reports/rp_d549463ccce0ea2f) [2](https://beanstalk.magicbean.dev/reports/rp_a469be1b399ffdbd) | 17.6 → 17.6 (−0%) | — | none | | `linux+intel` | ⏳ queued | 0 of 1 | — | — | — | | `steamdeck` | ❌ fail | 1 [1](https://beanstalk.magicbean.dev/reports/rp_b5cf890ddf8fdfea) | 44.8 → 45.4 (+1%) | — | none | | `amd+igpu` | ❌ fail | 1 [1](https://beanstalk.magicbean.dev/reports/rp_f47d73c8897e8c71) | 43.9 → 43.9 (+0%) | — | none | **Why it failed** - `windows+amd`: `three-areas` `kokiri-after-turn.fps_mean.min_delta_pct`: `115.4 → 91.8 (−20.5%); limit 10.0` - `windows+amd`: `three-areas` `screenshot.max_diff`: `0.7 → 0.7; limit 0.1` - `windows+amd`: `three-areas` `field-after-turn.stutter_count.max_delta_pct`: `0.0 → 0.5; limit 20.0; the baseline is 0` - `windows+amd`: `three-areas` `market-standing.stutter_count.max_delta_pct`: `0.0 → 1.5; limit 20.0; the baseline is 0` - `windows+amd`: `three-areas` `kokiri-after-turn.frametime_p99_ms.max_delta_pct`: `11.1 → 13.7 (+23.4%); limit 15.0` - `windows+amd`: `three-areas` `market-after-turn.stutter_count.max_delta_pct`: `0.0 → 0.5; limit 20.0; the baseline is 0` - `windows+nvidia`: `three-areas` `kokiri-after-turn.fps_mean.min_delta_pct`: `109.7 → 98.6 (−10.1%); limit 10.0` - `windows+nvidia`: `three-areas` `market-after-turn.stutter_count.max_delta_pct`: `0.0 → 1.0; limit 20.0; the baseline is 0` - and 20 more, in the reports **Couldn't be judged**: the tester's PC could not finish the test. The farm tries another PC, or the same one after a fix; it says nothing about the PR on its own. - `windows+nvidia`, [report 1](https://beanstalk.magicbean.dev/reports/rp_ff5f883c9bb9d420): `the main base failed (exit 1): build.py stopped (exit status 1): fatal error: too many errors emitted, stopping now [-ferror-limit=]` <details><summary>What was tested</summary> **What will be tested** · AI plan (`gpt-6.1-terra`) Two starts per side compare the PR with main across the requested Windows and Linux GPU targets, Steam Deck, and AMD integrated graphics. Each start visits Kokiri Forest, Hyrule Field, and Market with V-Sync off and the frame rate uncapped; standing performance and pictures are compared, followed by camera turns and a second measurement. The farm cannot turn the camera concurrently with a measurement, so the 15-second measurements follow the turns rather than measuring the turns themselves. Crashes, failure to reach an area or present a frame, substantial FPS or p99 regressions, increased stutters, and picture differences fail the plan. Each target takes roughly 25 minutes; targets without a matching online PC will wait for one. No tester online now matches linux+intel: that target waits for one. - **Where:** windows+amd (1-2 reports), windows+nvidia (1-2 reports), windows+intel (1-2 reports), linux+amd (1-2 reports), linux+nvidia (1-2 reports), linux+intel (1-2 reports), steamdeck (1-2 reports), amd+igpu (1-2 reports). - **three-areas:** from the repository's save `kokiri-start` (settings vsync=2, frame_rate_cap=6), then teleport to Kokiri Forest, measure kokiri-standing for 20 s (after 5 s warm-up), take the picture kokiri to compare with main, move the stick, measure kokiri-after-turn for 15 s (after 0 s warm-up), teleport to Hyrule Field, measure field-standing for 20 s (after 5 s warm-up), take the picture field to compare with main, measure field-after-turn for 15 s (after 0 s warm-up), teleport to Market, measure market-standing for 20 s (after 5 s warm-up), take the picture market to compare with main, measure market-after-turn for 15 s (after 0 s warm-up); 2 runs on each side. - **Measured:** crashes, frame times, stutters and hitches, draws and frame rate, the game's log. - **The PR fails if:** it crashes; or frame rate is more than 10% below main; or a picture differs from main by more than 5%; or stutters is more than 20% above main; or frame time (p99) is more than 15% above main; or forbidden log lines is above 0. - **Time:** about 28 min of game runs per tester (4 game starts), plus the two builds. </details> <details><summary>Measurements on `windows+amd`, for the PR description</summary> **Verdict:** fail on windows+amd (2/1 report): `three-areas kokiri-after-turn.fps_mean.min_delta_pct (-20.5%)`; `three-areas screenshot.max_diff`; `three-areas field-after-turn.stutter_count.max_delta_pct` failed. Hardware: 1: windows+amd+amd-proprietary+dgpu `AMD Radeon RX 9070 XT, 2.0.416` `AMD Ryzen 7 5800X3D 8-Core Processor, 32 GB RAM`; 2: windows+amd+amd-proprietary+dgpu `AMD Radeon RX 9060 XT, 2.0.416` `AMD Ryzen 5 5600 6-Core Processor, 32 GB RAM` Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 2 game starts a side, kokiri-standing 20 s after 5 s warm-up, kokiri-after-turn 15 s after 0 s warm-up, field-standing 20 s after 5 s warm-up, field-after-turn 15 s after 0 s warm-up, market-standing 20 s after 5 s warm-up, market-after-turn 15 s after 0 s warm-up; each number the median of the runs, then of 2 reports How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again) | | Before: main @ 89c3d21 | After: this PR @ f08644e | |---|---|---| | three-areas · field-standing · frame rate | 108.2 | 109.0 (+1%) | | three-areas · kokiri-standing · frame rate | 107.4 | 105.9 (−1%) | | three-areas · market-standing · frame rate | 66.4 | 68.0 (+2%) | | three-areas · field-after-turn · frame rate | 102.7 | 102.3 (−0%) | | three-areas · kokiri-after-turn · frame rate | 105.7 | 95.9 (−9%) | | three-areas · market-after-turn · frame rate | 73.4 | 81.9 (+12%) | | three-areas · field-standing · stutters | 0.2 | 0.0 (−100%) | | three-areas · kokiri-standing · stutters | 2.8 | 2.0 (−27%) | | three-areas · market-standing · stutters | 0.2 | 1.0 (+300%) | | three-areas · field-after-turn · stutters | 0.0 | 0.2 | | three-areas · kokiri-after-turn · stutters | 0.2 | 0.0 (−100%) | | three-areas · market-after-turn · stutters | 0.2 | 0.2 (+0%) | | three-areas · field-standing · frame time (p99) (ms) | 12.0 | 11.4 (−5%) | | three-areas · kokiri-standing · frame time (p99) (ms) | 12.7 | 11.7 (−8%) | | three-areas · market-standing · frame time (p99) (ms) | 20.1 | 19.2 (−5%) | | three-areas · field-after-turn · frame time (p99) (ms) | 12.6 | 12.4 (−1%) | | three-areas · kokiri-after-turn · frame time (p99) (ms) | 12.0 | 13.1 (+9%) | | three-areas · market-after-turn · frame time (p99) (ms) | 18.2 | 16.7 (−8%) | | three-areas · forbidden log lines | 0.0 | 0.0 | **Flags and settings** None: the plan varied no setting, switch or build option. As Markdown for the PR description: ```markdown ## Measurements **Verdict:** fail on windows+amd (2/1 report): `three-areas kokiri-after-turn.fps_mean.min_delta_pct (-20.5%)`; `three-areas screenshot.max_diff`; `three-areas field-after-turn.stutter_count.max_delta_pct` failed. Hardware: 1: windows+amd+amd-proprietary+dgpu `AMD Radeon RX 9070 XT, 2.0.416` `AMD Ryzen 7 5800X3D 8-Core Processor, 32 GB RAM`; 2: windows+amd+amd-proprietary+dgpu `AMD Radeon RX 9060 XT, 2.0.416` `AMD Ryzen 5 5600 6-Core Processor, 32 GB RAM` Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 2 game starts a side, kokiri-standing 20 s after 5 s warm-up, kokiri-after-turn 15 s after 0 s warm-up, field-standing 20 s after 5 s warm-up, field-after-turn 15 s after 0 s warm-up, market-standing 20 s after 5 s warm-up, market-after-turn 15 s after 0 s warm-up; each number the median of the runs, then of 2 reports How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again) | | Before: main @ 89c3d21 | After: this PR @ f08644e | |---|---|---| | three-areas · field-standing · frame rate | 108.2 | 109.0 (+1%) | | three-areas · kokiri-standing · frame rate | 107.4 | 105.9 (−1%) | | three-areas · market-standing · frame rate | 66.4 | 68.0 (+2%) | | three-areas · field-after-turn · frame rate | 102.7 | 102.3 (−0%) | | three-areas · kokiri-after-turn · frame rate | 105.7 | 95.9 (−9%) | | three-areas · market-after-turn · frame rate | 73.4 | 81.9 (+12%) | | three-areas · field-standing · stutters | 0.2 | 0.0 (−100%) | | three-areas · kokiri-standing · stutters | 2.8 | 2.0 (−27%) | | three-areas · market-standing · stutters | 0.2 | 1.0 (+300%) | | three-areas · field-after-turn · stutters | 0.0 | 0.2 | | three-areas · kokiri-after-turn · stutters | 0.2 | 0.0 (−100%) | | three-areas · market-after-turn · stutters | 0.2 | 0.2 (+0%) | | three-areas · field-standing · frame time (p99) (ms) | 12.0 | 11.4 (−5%) | | three-areas · kokiri-standing · frame time (p99) (ms) | 12.7 | 11.7 (−8%) | | three-areas · market-standing · frame time (p99) (ms) | 20.1 | 19.2 (−5%) | | three-areas · field-after-turn · frame time (p99) (ms) | 12.6 | 12.4 (−1%) | | three-areas · kokiri-after-turn · frame time (p99) (ms) | 12.0 | 13.1 (+9%) | | three-areas · market-after-turn · frame time (p99) (ms) | 18.2 | 16.7 (−8%) | | three-areas · forbidden log lines | 0.0 | 0.0 | **Flags and settings** None: the plan varied no setting, switch or build option. ``` </details> <details><summary>Measurements on `windows+nvidia`, for the PR description</summary> **Verdict:** fail on windows+nvidia (2/1 report): `three-areas kokiri-after-turn.fps_mean.min_delta_pct (-10.1%)`; `three-areas market-after-turn.stutter_count.max_delta_pct` failed. Hardware: 1: windows+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce RTX 4080, 621.1.0` `AMD Ryzen 7 9800X3D 8-Core Processor, 32 GB RAM`; 2: windows+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce RTX 3080 Ti, 621.24.0` `AMD Ryzen 9 5900X 12-Core Processor, 64 GB RAM` Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 2 game starts a side, kokiri-standing 20 s after 5 s warm-up, kokiri-after-turn 15 s after 0 s warm-up, field-standing 20 s after 5 s warm-up, field-after-turn 15 s after 0 s warm-up, market-standing 20 s after 5 s warm-up, market-after-turn 15 s after 0 s warm-up; each number the median of the runs, then of 2 reports How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again) | | Before: main @ 89c3d21 | After: this PR @ f08644e | |---|---|---| | three-areas · field-standing · frame rate | 108.1 | 108.4 (+0%) | | three-areas · kokiri-standing · frame rate | 108.9 | 101.2 (−7%) | | three-areas · market-standing · frame rate | 80.0 | 82.5 (+3%) | | three-areas · field-after-turn · frame rate | 117.2 | 115.5 (−1%) | | three-areas · kokiri-after-turn · frame rate | 109.7 | 98.6 (−10%) | | three-areas · market-after-turn · frame rate | 81.1 | 83.5 (+3%) | | three-areas · field-standing · stutters | 0.0 | 0.0 | | three-areas · kokiri-standing · stutters | 0.0 | 0.0 | | three-areas · market-standing · stutters | 0.5 | 0.0 (−100%) | | three-areas · field-after-turn · stutters | 0.0 | 0.0 | | three-areas · kokiri-after-turn · stutters | 0.5 | 0.0 (−100%) | | three-areas · market-after-turn · stutters | 0.0 | 1.0 | | three-areas · field-standing · frame time (p99) (ms) | 11.3 | 11.3 (+0%) | | three-areas · kokiri-standing · frame time (p99) (ms) | 11.1 | 11.4 (+3%) | | three-areas · market-standing · frame time (p99) (ms) | 15.4 | 14.8 (−4%) | | three-areas · field-after-turn · frame time (p99) (ms) | 10.3 | 10.9 (+6%) | | three-areas · kokiri-after-turn · frame time (p99) (ms) | 11.9 | 12.2 (+2%) | | three-areas · market-after-turn · frame time (p99) (ms) | 15.2 | 15.7 (+3%) | | three-areas · forbidden log lines | 0.0 | 0.0 | **Flags and settings** None: the plan varied no setting, switch or build option. As Markdown for the PR description: ```markdown ## Measurements **Verdict:** fail on windows+nvidia (2/1 report): `three-areas kokiri-after-turn.fps_mean.min_delta_pct (-10.1%)`; `three-areas market-after-turn.stutter_count.max_delta_pct` failed. Hardware: 1: windows+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce RTX 4080, 621.1.0` `AMD Ryzen 7 9800X3D 8-Core Processor, 32 GB RAM`; 2: windows+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce RTX 3080 Ti, 621.24.0` `AMD Ryzen 9 5900X 12-Core Processor, 64 GB RAM` Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 2 game starts a side, kokiri-standing 20 s after 5 s warm-up, kokiri-after-turn 15 s after 0 s warm-up, field-standing 20 s after 5 s warm-up, field-after-turn 15 s after 0 s warm-up, market-standing 20 s after 5 s warm-up, market-after-turn 15 s after 0 s warm-up; each number the median of the runs, then of 2 reports How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again) | | Before: main @ 89c3d21 | After: this PR @ f08644e | |---|---|---| | three-areas · field-standing · frame rate | 108.1 | 108.4 (+0%) | | three-areas · kokiri-standing · frame rate | 108.9 | 101.2 (−7%) | | three-areas · market-standing · frame rate | 80.0 | 82.5 (+3%) | | three-areas · field-after-turn · frame rate | 117.2 | 115.5 (−1%) | | three-areas · kokiri-after-turn · frame rate | 109.7 | 98.6 (−10%) | | three-areas · market-after-turn · frame rate | 81.1 | 83.5 (+3%) | | three-areas · field-standing · stutters | 0.0 | 0.0 | | three-areas · kokiri-standing · stutters | 0.0 | 0.0 | | three-areas · market-standing · stutters | 0.5 | 0.0 (−100%) | | three-areas · field-after-turn · stutters | 0.0 | 0.0 | | three-areas · kokiri-after-turn · stutters | 0.5 | 0.0 (−100%) | | three-areas · market-after-turn · stutters | 0.0 | 1.0 | | three-areas · field-standing · frame time (p99) (ms) | 11.3 | 11.3 (+0%) | | three-areas · kokiri-standing · frame time (p99) (ms) | 11.1 | 11.4 (+3%) | | three-areas · market-standing · frame time (p99) (ms) | 15.4 | 14.8 (−4%) | | three-areas · field-after-turn · frame time (p99) (ms) | 10.3 | 10.9 (+6%) | | three-areas · kokiri-after-turn · frame time (p99) (ms) | 11.9 | 12.2 (+2%) | | three-areas · market-after-turn · frame time (p99) (ms) | 15.2 | 15.7 (+3%) | | three-areas · forbidden log lines | 0.0 | 0.0 | **Flags and settings** None: the plan varied no setting, switch or build option. ``` </details> <details><summary>Measurements on `linux+amd`, for the PR description</summary> **Verdict:** fail on linux+amd (1/1 report): `three-areas kokiri-after-turn.fps_mean.min_delta_pct (-10.2%)`; `three-areas field-after-turn.fps_mean.min_delta_pct (-13.8%)`; `three-areas market-after-turn.fps_mean.min_delta_pct (-36.0%)` failed. Hardware: 1: linux+amd+radv+dgpu `AMD Steam Machine (RADV NAVI33), 26.1.99` `AMD Custom CPU 1772, 16 GB RAM` Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 2 game starts a side, kokiri-standing 20 s after 5 s warm-up, kokiri-after-turn 15 s after 0 s warm-up, field-standing 20 s after 5 s warm-up, field-after-turn 15 s after 0 s warm-up, market-standing 20 s after 5 s warm-up, market-after-turn 15 s after 0 s warm-up; each number the median of the runs, then of 1 report How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again) | | Before: main @ 89c3d21 | After: this PR @ f08644e | |---|---|---| | three-areas · field-standing · frame rate | 50.0 | 46.3 (−7%) | | three-areas · kokiri-standing · frame rate | 51.7 | 53.4 (+3%) | | three-areas · market-standing · frame rate | 19.8 | 19.2 (−3%) | | three-areas · field-after-turn · frame rate | 54.3 | 46.8 (−14%) | | three-areas · kokiri-after-turn · frame rate | 58.8 | 52.9 (−10%) | | three-areas · market-after-turn · frame rate | 64.8 | 41.5 (−36%) | | three-areas · field-standing · stutters | 5.0 | 24.5 (+390%) | | three-areas · kokiri-standing · stutters | 0.5 | 1.0 (+100%) | | three-areas · market-standing · stutters | 179.5 | 86.0 (−52%) | | three-areas · field-after-turn · stutters | 14.5 | 47.5 (+228%) | | three-areas · kokiri-after-turn · stutters | 1.0 | 0.0 (−100%) | | three-areas · market-after-turn · stutters | 1.5 | 60.0 (+3900%) | | three-areas · field-standing · frame time (p99) (ms) | 22.2 | 48.0 (+117%) | | three-areas · kokiri-standing · frame time (p99) (ms) | 21.2 | 21.4 (+1%) | | three-areas · market-standing · frame time (p99) (ms) | 84.6 | 87.6 (+4%) | | three-areas · field-after-turn · frame time (p99) (ms) | 63.8 | 70.5 (+11%) | | three-areas · kokiri-after-turn · frame time (p99) (ms) | 19.7 | 21.4 (+9%) | | three-areas · market-after-turn · frame time (p99) (ms) | 21.2 | 53.0 (+150%) | | three-areas · forbidden log lines | 0.0 | 0.0 | **Flags and settings** None: the plan varied no setting, switch or build option. As Markdown for the PR description: ```markdown ## Measurements **Verdict:** fail on linux+amd (1/1 report): `three-areas kokiri-after-turn.fps_mean.min_delta_pct (-10.2%)`; `three-areas field-after-turn.fps_mean.min_delta_pct (-13.8%)`; `three-areas market-after-turn.fps_mean.min_delta_pct (-36.0%)` failed. Hardware: 1: linux+amd+radv+dgpu `AMD Steam Machine (RADV NAVI33), 26.1.99` `AMD Custom CPU 1772, 16 GB RAM` Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 2 game starts a side, kokiri-standing 20 s after 5 s warm-up, kokiri-after-turn 15 s after 0 s warm-up, field-standing 20 s after 5 s warm-up, field-after-turn 15 s after 0 s warm-up, market-standing 20 s after 5 s warm-up, market-after-turn 15 s after 0 s warm-up; each number the median of the runs, then of 1 report How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again) | | Before: main @ 89c3d21 | After: this PR @ f08644e | |---|---|---| | three-areas · field-standing · frame rate | 50.0 | 46.3 (−7%) | | three-areas · kokiri-standing · frame rate | 51.7 | 53.4 (+3%) | | three-areas · market-standing · frame rate | 19.8 | 19.2 (−3%) | | three-areas · field-after-turn · frame rate | 54.3 | 46.8 (−14%) | | three-areas · kokiri-after-turn · frame rate | 58.8 | 52.9 (−10%) | | three-areas · market-after-turn · frame rate | 64.8 | 41.5 (−36%) | | three-areas · field-standing · stutters | 5.0 | 24.5 (+390%) | | three-areas · kokiri-standing · stutters | 0.5 | 1.0 (+100%) | | three-areas · market-standing · stutters | 179.5 | 86.0 (−52%) | | three-areas · field-after-turn · stutters | 14.5 | 47.5 (+228%) | | three-areas · kokiri-after-turn · stutters | 1.0 | 0.0 (−100%) | | three-areas · market-after-turn · stutters | 1.5 | 60.0 (+3900%) | | three-areas · field-standing · frame time (p99) (ms) | 22.2 | 48.0 (+117%) | | three-areas · kokiri-standing · frame time (p99) (ms) | 21.2 | 21.4 (+1%) | | three-areas · market-standing · frame time (p99) (ms) | 84.6 | 87.6 (+4%) | | three-areas · field-after-turn · frame time (p99) (ms) | 63.8 | 70.5 (+11%) | | three-areas · kokiri-after-turn · frame time (p99) (ms) | 19.7 | 21.4 (+9%) | | three-areas · market-after-turn · frame time (p99) (ms) | 21.2 | 53.0 (+150%) | | three-areas · forbidden log lines | 0.0 | 0.0 | **Flags and settings** None: the plan varied no setting, switch or build option. ``` </details> <details><summary>Measurements on `linux+nvidia`, for the PR description</summary> **Verdict:** fail on linux+nvidia (2/1 report): `three-areas kokiri-standing.stutter_count.max_delta_pct (+325.0%)`; `three-areas field-standing.stutter_count.max_delta_pct (+400.0%)`; `three-areas field-after-turn.stutter_count.max_delta_pct (+200.0%)` failed. Hardware: 1: linux+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce RTX 4070, 615.71.9` `AMD Ryzen 7 5800X3D 8-Core Processor, 64 GB RAM`; 2: linux+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce GTX 1070, 580.196.7` `Intel(R) Core(TM) i7-8750H CPU @ 2.20GHz, 32 GB RAM` Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 2 game starts a side, kokiri-standing 20 s after 5 s warm-up, kokiri-after-turn 15 s after 0 s warm-up, field-standing 20 s after 5 s warm-up, field-after-turn 15 s after 0 s warm-up, market-standing 20 s after 5 s warm-up, market-after-turn 15 s after 0 s warm-up; each number the median of the runs, then of 2 reports How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again) | | Before: main @ 89c3d21 | After: this PR @ f08644e | |---|---|---| | three-areas · field-standing · frame rate | 81.2 | 81.3 (+0%) | | three-areas · kokiri-standing · frame rate | 84.3 | 84.3 (−0%) | | three-areas · market-standing · frame rate | 63.5 | 63.2 (−1%) | | three-areas · field-after-turn · frame rate | 90.0 | 91.2 (+1%) | | three-areas · kokiri-after-turn · frame rate | 86.6 | 83.9 (−3%) | | three-areas · market-after-turn · frame rate | 80.4 | 89.5 (+11%) | | three-areas · field-standing · stutters | 0.2 | 1.2 (+400%) | | three-areas · kokiri-standing · stutters | 1.0 | 4.5 (+350%) | | three-areas · market-standing · stutters | 1.8 | 1.5 (−14%) | | three-areas · field-after-turn · stutters | 0.5 | 1.5 (+200%) | | three-areas · kokiri-after-turn · stutters | 1.0 | 0.0 (−100%) | | three-areas · market-after-turn · stutters | 1.8 | 2.0 (+14%) | | three-areas · field-standing · frame time (p99) (ms) | 17.6 | 17.6 (−0%) | | three-areas · kokiri-standing · frame time (p99) (ms) | 18.0 | 17.3 (−4%) | | three-areas · market-standing · frame time (p99) (ms) | 23.6 | 23.4 (−1%) | | three-areas · field-after-turn · frame time (p99) (ms) | 16.0 | 16.5 (+3%) | | three-areas · kokiri-after-turn · frame time (p99) (ms) | 16.3 | 17.0 (+4%) | | three-areas · market-after-turn · frame time (p99) (ms) | 19.4 | 19.3 (−1%) | | three-areas · forbidden log lines | 0.0 | 0.0 | **Flags and settings** None: the plan varied no setting, switch or build option. As Markdown for the PR description: ```markdown ## Measurements **Verdict:** fail on linux+nvidia (2/1 report): `three-areas kokiri-standing.stutter_count.max_delta_pct (+325.0%)`; `three-areas field-standing.stutter_count.max_delta_pct (+400.0%)`; `three-areas field-after-turn.stutter_count.max_delta_pct (+200.0%)` failed. Hardware: 1: linux+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce RTX 4070, 615.71.9` `AMD Ryzen 7 5800X3D 8-Core Processor, 64 GB RAM`; 2: linux+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce GTX 1070, 580.196.7` `Intel(R) Core(TM) i7-8750H CPU @ 2.20GHz, 32 GB RAM` Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 2 game starts a side, kokiri-standing 20 s after 5 s warm-up, kokiri-after-turn 15 s after 0 s warm-up, field-standing 20 s after 5 s warm-up, field-after-turn 15 s after 0 s warm-up, market-standing 20 s after 5 s warm-up, market-after-turn 15 s after 0 s warm-up; each number the median of the runs, then of 2 reports How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again) | | Before: main @ 89c3d21 | After: this PR @ f08644e | |---|---|---| | three-areas · field-standing · frame rate | 81.2 | 81.3 (+0%) | | three-areas · kokiri-standing · frame rate | 84.3 | 84.3 (−0%) | | three-areas · market-standing · frame rate | 63.5 | 63.2 (−1%) | | three-areas · field-after-turn · frame rate | 90.0 | 91.2 (+1%) | | three-areas · kokiri-after-turn · frame rate | 86.6 | 83.9 (−3%) | | three-areas · market-after-turn · frame rate | 80.4 | 89.5 (+11%) | | three-areas · field-standing · stutters | 0.2 | 1.2 (+400%) | | three-areas · kokiri-standing · stutters | 1.0 | 4.5 (+350%) | | three-areas · market-standing · stutters | 1.8 | 1.5 (−14%) | | three-areas · field-after-turn · stutters | 0.5 | 1.5 (+200%) | | three-areas · kokiri-after-turn · stutters | 1.0 | 0.0 (−100%) | | three-areas · market-after-turn · stutters | 1.8 | 2.0 (+14%) | | three-areas · field-standing · frame time (p99) (ms) | 17.6 | 17.6 (−0%) | | three-areas · kokiri-standing · frame time (p99) (ms) | 18.0 | 17.3 (−4%) | | three-areas · market-standing · frame time (p99) (ms) | 23.6 | 23.4 (−1%) | | three-areas · field-after-turn · frame time (p99) (ms) | 16.0 | 16.5 (+3%) | | three-areas · kokiri-after-turn · frame time (p99) (ms) | 16.3 | 17.0 (+4%) | | three-areas · market-after-turn · frame time (p99) (ms) | 19.4 | 19.3 (−1%) | | three-areas · forbidden log lines | 0.0 | 0.0 | **Flags and settings** None: the plan varied no setting, switch or build option. ``` </details> <details><summary>Measurements on `steamdeck`, for the PR description</summary> **Verdict:** fail on steamdeck (1/1 report): `three-areas market-standing.stutter_count.max_delta_pct (+50.0%)`; `three-areas market-after-turn.stutter_count.max_delta_pct (+88.9%)`; `three-areas field-after-turn.frametime_p99_ms.max_delta_pct (+19.4%)` failed. Hardware: 1: linux+amd+radv+igpu+handheld `AMD Custom GPU 0405 (RADV VANGOGH), 26.1.99` `AMD Custom APU 0405, 12 GB RAM` Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 2 game starts a side, kokiri-standing 20 s after 5 s warm-up, kokiri-after-turn 15 s after 0 s warm-up, field-standing 20 s after 5 s warm-up, field-after-turn 15 s after 0 s warm-up, market-standing 20 s after 5 s warm-up, market-after-turn 15 s after 0 s warm-up; each number the median of the runs, then of 1 report How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again) | | Before: main @ 89c3d21 | After: this PR @ f08644e | |---|---|---| | three-areas · field-standing · frame rate | 23.9 | 23.6 (−1%) | | three-areas · kokiri-standing · frame rate | 24.2 | 23.8 (−2%) | | three-areas · market-standing · frame rate | 16.5 | 16.0 (−3%) | | three-areas · field-after-turn · frame rate | 27.7 | 25.2 (−9%) | | three-areas · kokiri-after-turn · frame rate | 27.0 | 26.4 (−2%) | | three-areas · market-after-turn · frame rate | 17.0 | 19.7 (+15%) | | three-areas · field-standing · stutters | 0.0 | 0.0 | | three-areas · kokiri-standing · stutters | 0.5 | 0.5 (+0%) | | three-areas · market-standing · stutters | 8.0 | 12.0 (+50%) | | three-areas · field-after-turn · stutters | 0.0 | 0.0 | | three-areas · kokiri-after-turn · stutters | 0.0 | 0.0 | | three-areas · market-after-turn · stutters | 4.5 | 8.5 (+89%) | | three-areas · field-standing · frame time (p99) (ms) | 44.8 | 45.4 (+1%) | | three-areas · kokiri-standing · frame time (p99) (ms) | 54.1 | 55.6 (+3%) | | three-areas · market-standing · frame time (p99) (ms) | 143.7 | 150.7 (+5%) | | three-areas · field-after-turn · frame time (p99) (ms) | 39.3 | 47.0 (+19%) | | three-areas · kokiri-after-turn · frame time (p99) (ms) | 41.5 | 43.0 (+4%) | | three-areas · market-after-turn · frame time (p99) (ms) | 122.8 | 119.1 (−3%) | | three-areas · forbidden log lines | 0.0 | 0.0 | **Flags and settings** None: the plan varied no setting, switch or build option. As Markdown for the PR description: ```markdown ## Measurements **Verdict:** fail on steamdeck (1/1 report): `three-areas market-standing.stutter_count.max_delta_pct (+50.0%)`; `three-areas market-after-turn.stutter_count.max_delta_pct (+88.9%)`; `three-areas field-after-turn.frametime_p99_ms.max_delta_pct (+19.4%)` failed. Hardware: 1: linux+amd+radv+igpu+handheld `AMD Custom GPU 0405 (RADV VANGOGH), 26.1.99` `AMD Custom APU 0405, 12 GB RAM` Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 2 game starts a side, kokiri-standing 20 s after 5 s warm-up, kokiri-after-turn 15 s after 0 s warm-up, field-standing 20 s after 5 s warm-up, field-after-turn 15 s after 0 s warm-up, market-standing 20 s after 5 s warm-up, market-after-turn 15 s after 0 s warm-up; each number the median of the runs, then of 1 report How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again) | | Before: main @ 89c3d21 | After: this PR @ f08644e | |---|---|---| | three-areas · field-standing · frame rate | 23.9 | 23.6 (−1%) | | three-areas · kokiri-standing · frame rate | 24.2 | 23.8 (−2%) | | three-areas · market-standing · frame rate | 16.5 | 16.0 (−3%) | | three-areas · field-after-turn · frame rate | 27.7 | 25.2 (−9%) | | three-areas · kokiri-after-turn · frame rate | 27.0 | 26.4 (−2%) | | three-areas · market-after-turn · frame rate | 17.0 | 19.7 (+15%) | | three-areas · field-standing · stutters | 0.0 | 0.0 | | three-areas · kokiri-standing · stutters | 0.5 | 0.5 (+0%) | | three-areas · market-standing · stutters | 8.0 | 12.0 (+50%) | | three-areas · field-after-turn · stutters | 0.0 | 0.0 | | three-areas · kokiri-after-turn · stutters | 0.0 | 0.0 | | three-areas · market-after-turn · stutters | 4.5 | 8.5 (+89%) | | three-areas · field-standing · frame time (p99) (ms) | 44.8 | 45.4 (+1%) | | three-areas · kokiri-standing · frame time (p99) (ms) | 54.1 | 55.6 (+3%) | | three-areas · market-standing · frame time (p99) (ms) | 143.7 | 150.7 (+5%) | | three-areas · field-after-turn · frame time (p99) (ms) | 39.3 | 47.0 (+19%) | | three-areas · kokiri-after-turn · frame time (p99) (ms) | 41.5 | 43.0 (+4%) | | three-areas · market-after-turn · frame time (p99) (ms) | 122.8 | 119.1 (−3%) | | three-areas · forbidden log lines | 0.0 | 0.0 | **Flags and settings** None: the plan varied no setting, switch or build option. ``` </details> <details><summary>Measurements on `amd+igpu`, for the PR description</summary> **Verdict:** fail on amd+igpu (1/1 report): `three-areas screenshot.max_diff`; `three-areas kokiri-standing.stutter_count.max_delta_pct` failed. Hardware: 1: linux+amd+radv+igpu+handheld `AMD Custom GPU 0405 (RADV VANGOGH), 26.1.99` `AMD Custom APU 0405, 12 GB RAM` Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 2 game starts a side, kokiri-standing 20 s after 5 s warm-up, kokiri-after-turn 15 s after 0 s warm-up, field-standing 20 s after 5 s warm-up, field-after-turn 15 s after 0 s warm-up, market-standing 20 s after 5 s warm-up, market-after-turn 15 s after 0 s warm-up; each number the median of the runs, then of 1 report How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again) | | Before: main @ 89c3d21 | After: this PR @ f08644e | |---|---|---| | three-areas · field-standing · frame rate | 24.3 | 24.4 (+0%) | | three-areas · kokiri-standing · frame rate | 25.5 | 24.5 (−4%) | | three-areas · market-standing · frame rate | 17.6 | 18.1 (+3%) | | three-areas · field-after-turn · frame rate | 27.5 | 27.3 (−1%) | | three-areas · kokiri-after-turn · frame rate | 24.6 | 26.4 (+7%) | | three-areas · market-after-turn · frame rate | 19.4 | 20.7 (+6%) | | three-areas · field-standing · stutters | 0.0 | 0.0 | | three-areas · kokiri-standing · stutters | 0.0 | 0.5 | | three-areas · market-standing · stutters | 5.5 | 5.0 (−9%) | | three-areas · field-after-turn · stutters | 0.0 | 0.0 | | three-areas · kokiri-after-turn · stutters | 0.0 | 0.0 | | three-areas · market-after-turn · stutters | 8.5 | 7.0 (−18%) | | three-areas · field-standing · frame time (p99) (ms) | 43.9 | 43.9 (+0%) | | three-areas · kokiri-standing · frame time (p99) (ms) | 47.9 | 50.6 (+6%) | | three-areas · market-standing · frame time (p99) (ms) | 117.2 | 108.1 (−8%) | | three-areas · field-after-turn · frame time (p99) (ms) | 39.3 | 39.7 (+1%) | | three-areas · kokiri-after-turn · frame time (p99) (ms) | 46.5 | 42.1 (−10%) | | three-areas · market-after-turn · frame time (p99) (ms) | 129.8 | 109.3 (−16%) | | three-areas · forbidden log lines | 0.0 | 0.0 | **Flags and settings** None: the plan varied no setting, switch or build option. As Markdown for the PR description: ```markdown ## Measurements **Verdict:** fail on amd+igpu (1/1 report): `three-areas screenshot.max_diff`; `three-areas kokiri-standing.stutter_count.max_delta_pct` failed. Hardware: 1: linux+amd+radv+igpu+handheld `AMD Custom GPU 0405 (RADV VANGOGH), 26.1.99` `AMD Custom APU 0405, 12 GB RAM` Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 2 game starts a side, kokiri-standing 20 s after 5 s warm-up, kokiri-after-turn 15 s after 0 s warm-up, field-standing 20 s after 5 s warm-up, field-after-turn 15 s after 0 s warm-up, market-standing 20 s after 5 s warm-up, market-after-turn 15 s after 0 s warm-up; each number the median of the runs, then of 1 report How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again) | | Before: main @ 89c3d21 | After: this PR @ f08644e | |---|---|---| | three-areas · field-standing · frame rate | 24.3 | 24.4 (+0%) | | three-areas · kokiri-standing · frame rate | 25.5 | 24.5 (−4%) | | three-areas · market-standing · frame rate | 17.6 | 18.1 (+3%) | | three-areas · field-after-turn · frame rate | 27.5 | 27.3 (−1%) | | three-areas · kokiri-after-turn · frame rate | 24.6 | 26.4 (+7%) | | three-areas · market-after-turn · frame rate | 19.4 | 20.7 (+6%) | | three-areas · field-standing · stutters | 0.0 | 0.0 | | three-areas · kokiri-standing · stutters | 0.0 | 0.5 | | three-areas · market-standing · stutters | 5.5 | 5.0 (−9%) | | three-areas · field-after-turn · stutters | 0.0 | 0.0 | | three-areas · kokiri-after-turn · stutters | 0.0 | 0.0 | | three-areas · market-after-turn · stutters | 8.5 | 7.0 (−18%) | | three-areas · field-standing · frame time (p99) (ms) | 43.9 | 43.9 (+0%) | | three-areas · kokiri-standing · frame time (p99) (ms) | 47.9 | 50.6 (+6%) | | three-areas · market-standing · frame time (p99) (ms) | 117.2 | 108.1 (−8%) | | three-areas · field-after-turn · frame time (p99) (ms) | 39.3 | 39.7 (+1%) | | three-areas · kokiri-after-turn · frame time (p99) (ms) | 46.5 | 42.1 (−10%) | | three-areas · market-after-turn · frame time (p99) (ms) | 129.8 | 109.3 (−16%) | | three-areas · forbidden log lines | 0.0 | 0.0 | **Flags and settings** None: the plan varied no setting, switch or build option. ``` </details> <details><summary>The testers' hardware</summary> | Target | Report | Ran on | | --- | --- | --- | | `windows+amd` | [1](https://beanstalk.magicbean.dev/reports/rp_18b65d3e2212c214) | windows+amd+amd-proprietary+dgpu `AMD Radeon RX 9070 XT, 2.0.416` `AMD Ryzen 7 5800X3D 8-Core Processor, 32 GB RAM` | | `windows+amd` | [2](https://beanstalk.magicbean.dev/reports/rp_28cd9c2a5d5a15aa) | windows+amd+amd-proprietary+dgpu `AMD Radeon RX 9060 XT, 2.0.416` `AMD Ryzen 5 5600 6-Core Processor, 32 GB RAM` | | `windows+nvidia` | [1](https://beanstalk.magicbean.dev/reports/rp_ff5f883c9bb9d420) | windows+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce RTX 4080, 621.1.0` `AMD Ryzen 7 9800X3D 8-Core Processor, 32 GB RAM` | | `windows+nvidia` | [2](https://beanstalk.magicbean.dev/reports/rp_24abd5b090f1f8b3) | windows+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce RTX 3080 Ti, 621.24.0` `AMD Ryzen 9 5900X 12-Core Processor, 64 GB RAM` | | `linux+amd` | [1](https://beanstalk.magicbean.dev/reports/rp_69e22fd96ba9a4f5) | linux+amd+radv+dgpu `AMD Steam Machine (RADV NAVI33), 26.1.99` `AMD Custom CPU 1772, 16 GB RAM` | | `linux+nvidia` | [1](https://beanstalk.magicbean.dev/reports/rp_d549463ccce0ea2f) | linux+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce RTX 4070, 615.71.9` `AMD Ryzen 7 5800X3D 8-Core Processor, 64 GB RAM` | | `linux+nvidia` | [2](https://beanstalk.magicbean.dev/reports/rp_a469be1b399ffdbd) | linux+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce GTX 1070, 580.196.7` `Intel(R) Core(TM) i7-8750H CPU @ 2.20GHz, 32 GB RAM` | | `steamdeck` | [1](https://beanstalk.magicbean.dev/reports/rp_b5cf890ddf8fdfea) | linux+amd+radv+igpu+handheld `AMD Custom GPU 0405 (RADV VANGOGH), 26.1.99` `AMD Custom APU 0405, 12 GB RAM` | | `amd+igpu` | [1](https://beanstalk.magicbean.dev/reports/rp_f47d73c8897e8c71) | linux+amd+radv+igpu+handheld `AMD Custom GPU 0405 (RADV VANGOGH), 26.1.99` `AMD Custom APU 0405, 12 GB RAM` | </details> </details>
HumanBean added this to the WIP project 2026-11-20 15:15:37 +00:00
Author
Owner

@Beanstalk Please test this PR against main on every platform: windows+amd, windows+nvidia, windows+intel, linux+amd, linux+nvidia, linux+intel, the steamdeck, amd+igpu. Settings vsync=2 (Off) and frame_rate_cap=6 (Uncapped), Threaded rendering Auto. From the kokiri-start save: Kokiri Forest, Hyrule Field and Castle Town Market, in each 20 s standing then 15 s turning the camera, three game starts a side; make sure each side has reached the area before measuring it. Measure fps, p99 frame time, stutters, crashes and hangs (a hang is the main risk: this changes which thread submits the GPU's work), and compare pictures with main. This only acts where the recorder thread runs (6+ cores); on the Steam Deck and 4-core PCs it should be even.

@Beanstalk Please test this PR against main on every platform: windows+amd, windows+nvidia, windows+intel, linux+amd, linux+nvidia, linux+intel, the steamdeck, amd+igpu. Settings vsync=2 (Off) and frame_rate_cap=6 (Uncapped), Threaded rendering Auto. From the `kokiri-start` save: Kokiri Forest, Hyrule Field and Castle Town Market, in each 20 s standing then 15 s turning the camera, three game starts a side; make sure each side has reached the area before measuring it. Measure fps, p99 frame time, stutters, crashes and hangs (a hang is the main risk: this changes which thread submits the GPU's work), and compare pictures with main. This only acts where the recorder thread runs (6+ cores); on the Steam Deck and 4-core PCs it should be even.
Member

Beanstalk test farm · 🔄 testing (4 of 8 reports so far, a failure already)

Commit f08644e · approved by HumanBean (reviewed) · dashboard

Done (started 16:21 UTC, last report 17:11 UTC). Run 2: re-planned by HumanBean. Comment @Beanstalk to run it again.

Target Result Reports Frame time p99 (ms), main → PR VRAM peak (MB) Crashes
windows+amd ❌ fail (an outlier) 1 1 10.8 → 10.6 (−2%) — none
windows+nvidia ⚠️ couldn't be judged 1 1 — — none
windows+intel ⏳ queued 0 of 1 — — —
linux+amd ⏳ queued 0 of 1 — — —
linux+nvidia ❌ fail (an outlier) 2 1 2 21.8 → 21.9 (+0%) — none
linux+intel ⏳ queued 0 of 1 — — —
steamdeck ⏳ queued 0 of 1 — — —
amd+igpu ⏳ queued 0 of 1 — — —

Why it failed

  • windows+amd: three-areas forest-standing.fps_mean.min_delta_pct: 126.6 → 114.5 (−9.6%); limit 5.0
  • windows+amd: three-areas screenshot.max_diff: 0.7 → 0.9; limit 0.1
  • windows+amd: three-areas forest-standing.frametime_p99_ms.max_delta_pct: 9.9 → 11.3 (+14.9%); limit 10.0
  • linux+nvidia: three-areas forest-standing.frametime_p99_ms.max_delta_pct: 20.3 → 22.6 (+11.2%); limit 10.0

Couldn't be judged: the tester's PC could not finish the test. The farm tries another PC, or the same one after a fix; it says nothing about the PR on its own.

  • windows+nvidia, report 1: the main base failed (exit 1): build.py stopped (exit status 1): fatal error: too many errors emitted, stopping now [-ferror-limit=]
  • linux+nvidia, report 2: Insufficient frame-time evidence for 6 required comparison(s). See rule results. a step ran out of time: timeout: the save did not load into play within 900 s
What was tested

What will be tested · AI plan (gpt-6.1-terra)

Against main, three starts per side will visit Kokiri Forest, Hyrule Field and Market with V-Sync off and the frame-rate cap uncapped; threaded rendering remains at its default Auto. Each area gets a 20-second standing measurement and a camera-turn exercise with comparison pictures; the farm cannot measure while issuing the turn input. The plan measures FPS, p99 frame time and stutters, and fails on crashes, area-arrival timeouts, substantial performance regressions or picture differences. Each target takes roughly 35 minutes; the Windows and Linux Intel targets will wait for matching testers to come online. No tester online now matches windows+intel, linux+intel: those targets wait for one.

  • Where: windows+amd (1-2 reports), windows+nvidia (1-2 reports), windows+intel (1-2 reports), linux+amd (1-2 reports), linux+nvidia (1-2 reports), linux+intel (1-2 reports), steamdeck (1-2 reports), amd+igpu (1-2 reports).
  • three-areas: from the repository's save kokiri-start (settings vsync=2, frame_rate_cap=6), then teleport to Kokiri Forest, measure forest-standing for 20 s (after 5 s warm-up), take the picture forest-standing to compare with main, move the stick, take the picture forest-turned to compare with main, teleport to Hyrule Field, measure field-standing for 20 s (after 5 s warm-up), take the picture field-standing to compare with main, take the picture field-turned to compare with main, teleport to Market, measure market-standing for 20 s (after 5 s warm-up), take the picture market-standing to compare with main, take the picture market-turned to compare with main; 3 runs on each side.
  • Measured: crashes, frame times, stutters and hitches, draws and frame rate, the game's log.
  • The PR fails if: it crashes; or frame rate is more than 5% below main; or a picture differs from main by more than 5%; or stutters is more than 20% above main; or frame time (p99) is more than 10% above main; or forbidden log lines is above 0.
  • Time: about 38 min of game runs per tester (6 game starts), plus the two builds.
Measurements on `windows+amd`, for the PR description

Verdict: fail on windows+amd (1/1 report): three-areas forest-standing.fps_mean.min_delta_pct (-9.6%); three-areas screenshot.max_diff; three-areas forest-standing.frametime_p99_ms.max_delta_pct (+14.9%) failed.

Hardware: 1: windows+amd+amd-proprietary+dgpu AMD Radeon RX 9070 XT, 2.0.416 AMD Ryzen 7 5800X3D 8-Core Processor, 32 GB RAM
Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them
Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 3 game starts a side, forest-standing 20 s after 5 s warm-up, field-standing 20 s after 5 s warm-up, market-standing 20 s after 5 s warm-up; each number the median of the runs, then of 1 report
How to reproduce: the farm's plan above (@Beanstalk on this PR runs it again)

Before: main @ 89c3d21 After: this PR @ f08644e
three-areas · field-standing · frame rate 119.9 123.0 (+3%)
three-areas · forest-standing · frame rate 126.6 114.5 (−10%)
three-areas · market-standing · frame rate 68.3 71.0 (+4%)
three-areas · field-standing · stutters 0.0 0.0
three-areas · forest-standing · stutters 0.0 0.0
three-areas · market-standing · stutters 0.0 0.0
three-areas · field-standing · frame time (p99) (ms) 10.8 10.6 (−2%)
three-areas · forest-standing · frame time (p99) (ms) 9.9 11.3 (+15%)
three-areas · market-standing · frame time (p99) (ms) 18.4 18.1 (−1%)
three-areas · forbidden log lines 0.0 0.0

Flags and settings

None: the plan varied no setting, switch or build option.

As Markdown for the PR description:

## Measurements

**Verdict:** fail on windows+amd (1/1 report): `three-areas forest-standing.fps_mean.min_delta_pct (-9.6%)`; `three-areas screenshot.max_diff`; `three-areas forest-standing.frametime_p99_ms.max_delta_pct (+14.9%)` failed.

Hardware: 1: windows+amd+amd-proprietary+dgpu `AMD Radeon RX 9070 XT, 2.0.416` `AMD Ryzen 7 5800X3D 8-Core Processor, 32 GB RAM`  
Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them  
Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 3 game starts a side, forest-standing 20 s after 5 s warm-up, field-standing 20 s after 5 s warm-up, market-standing 20 s after 5 s warm-up; each number the median of the runs, then of 1 report  
How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again)

| | Before: main @ 89c3d21 | After: this PR @ f08644e |
|---|---|---|
| three-areas · field-standing · frame rate | 119.9 | 123.0 (+3%) |
| three-areas · forest-standing · frame rate | 126.6 | 114.5 (−10%) |
| three-areas · market-standing · frame rate | 68.3 | 71.0 (+4%) |
| three-areas · field-standing · stutters | 0.0 | 0.0 |
| three-areas · forest-standing · stutters | 0.0 | 0.0 |
| three-areas · market-standing · stutters | 0.0 | 0.0 |
| three-areas · field-standing · frame time (p99) (ms) | 10.8 | 10.6 (−2%) |
| three-areas · forest-standing · frame time (p99) (ms) | 9.9 | 11.3 (+15%) |
| three-areas · market-standing · frame time (p99) (ms) | 18.4 | 18.1 (−1%) |
| three-areas · forbidden log lines | 0.0 | 0.0 |

**Flags and settings**

None: the plan varied no setting, switch or build option.
Measurements on `windows+nvidia`, for the PR description

Verdict: could not be judged on windows+nvidia (1/1 report).

Hardware: 1: windows+nvidia+nvidia-proprietary+dgpu NVIDIA GeForce RTX 4080, 621.1.0 AMD Ryzen 7 9800X3D 8-Core Processor, 32 GB RAM
Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them
Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 3 game starts a side, forest-standing 20 s after 5 s warm-up, field-standing 20 s after 5 s warm-up, market-standing 20 s after 5 s warm-up; each number the median of the runs, then of 1 report
How to reproduce: the farm's plan above (@Beanstalk on this PR runs it again)

Nothing was measured: no metric came back.

Flags and settings

None: the plan varied no setting, switch or build option.

As Markdown for the PR description:

## Measurements

**Verdict:** could not be judged on windows+nvidia (1/1 report).

Hardware: 1: windows+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce RTX 4080, 621.1.0` `AMD Ryzen 7 9800X3D 8-Core Processor, 32 GB RAM`  
Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them  
Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 3 game starts a side, forest-standing 20 s after 5 s warm-up, field-standing 20 s after 5 s warm-up, market-standing 20 s after 5 s warm-up; each number the median of the runs, then of 1 report  
How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again)

Nothing was measured: no metric came back.

**Flags and settings**

None: the plan varied no setting, switch or build option.
Measurements on `linux+nvidia`, for the PR description

Verdict: fail on linux+nvidia (2/1 report): three-areas forest-standing.frametime_p99_ms.max_delta_pct (+11.2%) failed.

Hardware: 1: linux+nvidia+nvidia-proprietary+dgpu NVIDIA GeForce GTX 1070, 580.196.7 Intel(R) Core(TM) i7-8750H CPU @ 2.20GHz, 32 GB RAM; 2: linux+nvidia+nvidia-proprietary+dgpu NVIDIA GeForce RTX 5070, 620.70.6 AMD Ryzen 5 9600X 6-Core Processor, 32 GB RAM
Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them
Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 3 game starts a side, forest-standing 20 s after 5 s warm-up, field-standing 20 s after 5 s warm-up, market-standing 20 s after 5 s warm-up; each number the median of the runs, then of 2 reports
How to reproduce: the farm's plan above (@Beanstalk on this PR runs it again)

Before: main @ 89c3d21 After: this PR @ f08644e
three-areas · field-standing · frame rate 50.7 50.8 (+0%)
three-areas · forest-standing · frame rate 53.9 53.7 (−0%)
three-areas · market-standing · frame rate 40.4 39.5 (−2%)
three-areas · field-standing · stutters 0.0 0.0
three-areas · forest-standing · stutters 0.0 0.0
three-areas · market-standing · stutters 0.0 0.0
three-areas · field-standing · frame time (p99) (ms) 21.8 21.9 (+0%)
three-areas · forest-standing · frame time (p99) (ms) 20.3 22.6 (+11%)
three-areas · market-standing · frame time (p99) (ms) 28.2 30.1 (+7%)
three-areas · forbidden log lines 0.0 0.0

Flags and settings

None: the plan varied no setting, switch or build option.

As Markdown for the PR description:

## Measurements

**Verdict:** fail on linux+nvidia (2/1 report): `three-areas forest-standing.frametime_p99_ms.max_delta_pct (+11.2%)` failed.

Hardware: 1: linux+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce GTX 1070, 580.196.7` `Intel(R) Core(TM) i7-8750H CPU @ 2.20GHz, 32 GB RAM`; 2: linux+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce RTX 5070, 620.70.6` `AMD Ryzen 5 9600X 6-Core Processor, 32 GB RAM`  
Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them  
Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 3 game starts a side, forest-standing 20 s after 5 s warm-up, field-standing 20 s after 5 s warm-up, market-standing 20 s after 5 s warm-up; each number the median of the runs, then of 2 reports  
How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again)

| | Before: main @ 89c3d21 | After: this PR @ f08644e |
|---|---|---|
| three-areas · field-standing · frame rate | 50.7 | 50.8 (+0%) |
| three-areas · forest-standing · frame rate | 53.9 | 53.7 (−0%) |
| three-areas · market-standing · frame rate | 40.4 | 39.5 (−2%) |
| three-areas · field-standing · stutters | 0.0 | 0.0 |
| three-areas · forest-standing · stutters | 0.0 | 0.0 |
| three-areas · market-standing · stutters | 0.0 | 0.0 |
| three-areas · field-standing · frame time (p99) (ms) | 21.8 | 21.9 (+0%) |
| three-areas · forest-standing · frame time (p99) (ms) | 20.3 | 22.6 (+11%) |
| three-areas · market-standing · frame time (p99) (ms) | 28.2 | 30.1 (+7%) |
| three-areas · forbidden log lines | 0.0 | 0.0 |

**Flags and settings**

None: the plan varied no setting, switch or build option.
The testers' hardware
Target Report Ran on
windows+amd 1 windows+amd+amd-proprietary+dgpu AMD Radeon RX 9070 XT, 2.0.416 AMD Ryzen 7 5800X3D 8-Core Processor, 32 GB RAM
windows+nvidia 1 windows+nvidia+nvidia-proprietary+dgpu NVIDIA GeForce RTX 4080, 621.1.0 AMD Ryzen 7 9800X3D 8-Core Processor, 32 GB RAM
linux+nvidia 1 linux+nvidia+nvidia-proprietary+dgpu NVIDIA GeForce GTX 1070, 580.196.7 Intel(R) Core(TM) i7-8750H CPU @ 2.20GHz, 32 GB RAM
linux+nvidia 2 linux+nvidia+nvidia-proprietary+dgpu NVIDIA GeForce RTX 5070, 620.70.6 AMD Ryzen 5 9600X 6-Core Processor, 32 GB RAM
<!-- beanstalk:sticky --> ### Beanstalk test farm · 🔄 testing (4 of 8 reports so far, a failure already) Commit `f08644e` · approved by HumanBean (reviewed) · [dashboard](https://beanstalk.magicbean.dev/prs/1866) Done (started 16:21 UTC, last report 17:11 UTC). Run 2: re-planned by HumanBean. Comment `@Beanstalk` to run it again. | Target | Result | Reports | Frame time p99 (ms), main → PR | VRAM peak (MB) | Crashes | | --- | --- | --- | --- | --- | --- | | `windows+amd` | ❌ fail (an outlier) | 1 [1](https://beanstalk.magicbean.dev/reports/rp_c89210d0846b746f) | 10.8 → 10.6 (−2%) | — | none | | `windows+nvidia` | ⚠️ couldn't be judged | 1 [1](https://beanstalk.magicbean.dev/reports/rp_9993b02c2c799c3b) | — | — | none | | `windows+intel` | ⏳ queued | 0 of 1 | — | — | — | | `linux+amd` | ⏳ queued | 0 of 1 | — | — | — | | `linux+nvidia` | ❌ fail (an outlier) | 2 [1](https://beanstalk.magicbean.dev/reports/rp_58e348c5ebac64ff) [2](https://beanstalk.magicbean.dev/reports/rp_40dabb02ecc42fd9) | 21.8 → 21.9 (+0%) | — | none | | `linux+intel` | ⏳ queued | 0 of 1 | — | — | — | | `steamdeck` | ⏳ queued | 0 of 1 | — | — | — | | `amd+igpu` | ⏳ queued | 0 of 1 | — | — | — | **Why it failed** - `windows+amd`: `three-areas` `forest-standing.fps_mean.min_delta_pct`: `126.6 → 114.5 (−9.6%); limit 5.0` - `windows+amd`: `three-areas` `screenshot.max_diff`: `0.7 → 0.9; limit 0.1` - `windows+amd`: `three-areas` `forest-standing.frametime_p99_ms.max_delta_pct`: `9.9 → 11.3 (+14.9%); limit 10.0` - `linux+nvidia`: `three-areas` `forest-standing.frametime_p99_ms.max_delta_pct`: `20.3 → 22.6 (+11.2%); limit 10.0` **Couldn't be judged**: the tester's PC could not finish the test. The farm tries another PC, or the same one after a fix; it says nothing about the PR on its own. - `windows+nvidia`, [report 1](https://beanstalk.magicbean.dev/reports/rp_9993b02c2c799c3b): `the main base failed (exit 1): build.py stopped (exit status 1): fatal error: too many errors emitted, stopping now [-ferror-limit=]` - `linux+nvidia`, [report 2](https://beanstalk.magicbean.dev/reports/rp_40dabb02ecc42fd9): `Insufficient frame-time evidence for 6 required comparison(s). See rule results. a step ran out of time: timeout: the save did not load into play within 900 s` <details><summary>What was tested</summary> **What will be tested** · AI plan (`gpt-6.1-terra`) Against main, three starts per side will visit Kokiri Forest, Hyrule Field and Market with V-Sync off and the frame-rate cap uncapped; threaded rendering remains at its default Auto. Each area gets a 20-second standing measurement and a camera-turn exercise with comparison pictures; the farm cannot measure while issuing the turn input. The plan measures FPS, p99 frame time and stutters, and fails on crashes, area-arrival timeouts, substantial performance regressions or picture differences. Each target takes roughly 35 minutes; the Windows and Linux Intel targets will wait for matching testers to come online. No tester online now matches windows+intel, linux+intel: those targets wait for one. - **Where:** windows+amd (1-2 reports), windows+nvidia (1-2 reports), windows+intel (1-2 reports), linux+amd (1-2 reports), linux+nvidia (1-2 reports), linux+intel (1-2 reports), steamdeck (1-2 reports), amd+igpu (1-2 reports). - **three-areas:** from the repository's save `kokiri-start` (settings vsync=2, frame_rate_cap=6), then teleport to Kokiri Forest, measure forest-standing for 20 s (after 5 s warm-up), take the picture forest-standing to compare with main, move the stick, take the picture forest-turned to compare with main, teleport to Hyrule Field, measure field-standing for 20 s (after 5 s warm-up), take the picture field-standing to compare with main, take the picture field-turned to compare with main, teleport to Market, measure market-standing for 20 s (after 5 s warm-up), take the picture market-standing to compare with main, take the picture market-turned to compare with main; 3 runs on each side. - **Measured:** crashes, frame times, stutters and hitches, draws and frame rate, the game's log. - **The PR fails if:** it crashes; or frame rate is more than 5% below main; or a picture differs from main by more than 5%; or stutters is more than 20% above main; or frame time (p99) is more than 10% above main; or forbidden log lines is above 0. - **Time:** about 38 min of game runs per tester (6 game starts), plus the two builds. </details> <details><summary>Measurements on `windows+amd`, for the PR description</summary> **Verdict:** fail on windows+amd (1/1 report): `three-areas forest-standing.fps_mean.min_delta_pct (-9.6%)`; `three-areas screenshot.max_diff`; `three-areas forest-standing.frametime_p99_ms.max_delta_pct (+14.9%)` failed. Hardware: 1: windows+amd+amd-proprietary+dgpu `AMD Radeon RX 9070 XT, 2.0.416` `AMD Ryzen 7 5800X3D 8-Core Processor, 32 GB RAM` Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 3 game starts a side, forest-standing 20 s after 5 s warm-up, field-standing 20 s after 5 s warm-up, market-standing 20 s after 5 s warm-up; each number the median of the runs, then of 1 report How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again) | | Before: main @ 89c3d21 | After: this PR @ f08644e | |---|---|---| | three-areas · field-standing · frame rate | 119.9 | 123.0 (+3%) | | three-areas · forest-standing · frame rate | 126.6 | 114.5 (−10%) | | three-areas · market-standing · frame rate | 68.3 | 71.0 (+4%) | | three-areas · field-standing · stutters | 0.0 | 0.0 | | three-areas · forest-standing · stutters | 0.0 | 0.0 | | three-areas · market-standing · stutters | 0.0 | 0.0 | | three-areas · field-standing · frame time (p99) (ms) | 10.8 | 10.6 (−2%) | | three-areas · forest-standing · frame time (p99) (ms) | 9.9 | 11.3 (+15%) | | three-areas · market-standing · frame time (p99) (ms) | 18.4 | 18.1 (−1%) | | three-areas · forbidden log lines | 0.0 | 0.0 | **Flags and settings** None: the plan varied no setting, switch or build option. As Markdown for the PR description: ```markdown ## Measurements **Verdict:** fail on windows+amd (1/1 report): `three-areas forest-standing.fps_mean.min_delta_pct (-9.6%)`; `three-areas screenshot.max_diff`; `three-areas forest-standing.frametime_p99_ms.max_delta_pct (+14.9%)` failed. Hardware: 1: windows+amd+amd-proprietary+dgpu `AMD Radeon RX 9070 XT, 2.0.416` `AMD Ryzen 7 5800X3D 8-Core Processor, 32 GB RAM` Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 3 game starts a side, forest-standing 20 s after 5 s warm-up, field-standing 20 s after 5 s warm-up, market-standing 20 s after 5 s warm-up; each number the median of the runs, then of 1 report How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again) | | Before: main @ 89c3d21 | After: this PR @ f08644e | |---|---|---| | three-areas · field-standing · frame rate | 119.9 | 123.0 (+3%) | | three-areas · forest-standing · frame rate | 126.6 | 114.5 (−10%) | | three-areas · market-standing · frame rate | 68.3 | 71.0 (+4%) | | three-areas · field-standing · stutters | 0.0 | 0.0 | | three-areas · forest-standing · stutters | 0.0 | 0.0 | | three-areas · market-standing · stutters | 0.0 | 0.0 | | three-areas · field-standing · frame time (p99) (ms) | 10.8 | 10.6 (−2%) | | three-areas · forest-standing · frame time (p99) (ms) | 9.9 | 11.3 (+15%) | | three-areas · market-standing · frame time (p99) (ms) | 18.4 | 18.1 (−1%) | | three-areas · forbidden log lines | 0.0 | 0.0 | **Flags and settings** None: the plan varied no setting, switch or build option. ``` </details> <details><summary>Measurements on `windows+nvidia`, for the PR description</summary> **Verdict:** could not be judged on windows+nvidia (1/1 report). Hardware: 1: windows+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce RTX 4080, 621.1.0` `AMD Ryzen 7 9800X3D 8-Core Processor, 32 GB RAM` Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 3 game starts a side, forest-standing 20 s after 5 s warm-up, field-standing 20 s after 5 s warm-up, market-standing 20 s after 5 s warm-up; each number the median of the runs, then of 1 report How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again) Nothing was measured: no metric came back. **Flags and settings** None: the plan varied no setting, switch or build option. As Markdown for the PR description: ```markdown ## Measurements **Verdict:** could not be judged on windows+nvidia (1/1 report). Hardware: 1: windows+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce RTX 4080, 621.1.0` `AMD Ryzen 7 9800X3D 8-Core Processor, 32 GB RAM` Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 3 game starts a side, forest-standing 20 s after 5 s warm-up, field-standing 20 s after 5 s warm-up, market-standing 20 s after 5 s warm-up; each number the median of the runs, then of 1 report How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again) Nothing was measured: no metric came back. **Flags and settings** None: the plan varied no setting, switch or build option. ``` </details> <details><summary>Measurements on `linux+nvidia`, for the PR description</summary> **Verdict:** fail on linux+nvidia (2/1 report): `three-areas forest-standing.frametime_p99_ms.max_delta_pct (+11.2%)` failed. Hardware: 1: linux+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce GTX 1070, 580.196.7` `Intel(R) Core(TM) i7-8750H CPU @ 2.20GHz, 32 GB RAM`; 2: linux+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce RTX 5070, 620.70.6` `AMD Ryzen 5 9600X 6-Core Processor, 32 GB RAM` Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 3 game starts a side, forest-standing 20 s after 5 s warm-up, field-standing 20 s after 5 s warm-up, market-standing 20 s after 5 s warm-up; each number the median of the runs, then of 2 reports How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again) | | Before: main @ 89c3d21 | After: this PR @ f08644e | |---|---|---| | three-areas · field-standing · frame rate | 50.7 | 50.8 (+0%) | | three-areas · forest-standing · frame rate | 53.9 | 53.7 (−0%) | | three-areas · market-standing · frame rate | 40.4 | 39.5 (−2%) | | three-areas · field-standing · stutters | 0.0 | 0.0 | | three-areas · forest-standing · stutters | 0.0 | 0.0 | | three-areas · market-standing · stutters | 0.0 | 0.0 | | three-areas · field-standing · frame time (p99) (ms) | 21.8 | 21.9 (+0%) | | three-areas · forest-standing · frame time (p99) (ms) | 20.3 | 22.6 (+11%) | | three-areas · market-standing · frame time (p99) (ms) | 28.2 | 30.1 (+7%) | | three-areas · forbidden log lines | 0.0 | 0.0 | **Flags and settings** None: the plan varied no setting, switch or build option. As Markdown for the PR description: ```markdown ## Measurements **Verdict:** fail on linux+nvidia (2/1 report): `three-areas forest-standing.frametime_p99_ms.max_delta_pct (+11.2%)` failed. Hardware: 1: linux+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce GTX 1070, 580.196.7` `Intel(R) Core(TM) i7-8750H CPU @ 2.20GHz, 32 GB RAM`; 2: linux+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce RTX 5070, 620.70.6` `AMD Ryzen 5 9600X 6-Core Processor, 32 GB RAM` Scene: three-areas, from the repository's save kokiri-start (vsync=2, frame_rate_cap=6); each cell's settings and switches as the table names them Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); three-areas: 3 game starts a side, forest-standing 20 s after 5 s warm-up, field-standing 20 s after 5 s warm-up, market-standing 20 s after 5 s warm-up; each number the median of the runs, then of 2 reports How to reproduce: the farm's plan above (`@Beanstalk` on this PR runs it again) | | Before: main @ 89c3d21 | After: this PR @ f08644e | |---|---|---| | three-areas · field-standing · frame rate | 50.7 | 50.8 (+0%) | | three-areas · forest-standing · frame rate | 53.9 | 53.7 (−0%) | | three-areas · market-standing · frame rate | 40.4 | 39.5 (−2%) | | three-areas · field-standing · stutters | 0.0 | 0.0 | | three-areas · forest-standing · stutters | 0.0 | 0.0 | | three-areas · market-standing · stutters | 0.0 | 0.0 | | three-areas · field-standing · frame time (p99) (ms) | 21.8 | 21.9 (+0%) | | three-areas · forest-standing · frame time (p99) (ms) | 20.3 | 22.6 (+11%) | | three-areas · market-standing · frame time (p99) (ms) | 28.2 | 30.1 (+7%) | | three-areas · forbidden log lines | 0.0 | 0.0 | **Flags and settings** None: the plan varied no setting, switch or build option. ``` </details> <details><summary>The testers' hardware</summary> | Target | Report | Ran on | | --- | --- | --- | | `windows+amd` | [1](https://beanstalk.magicbean.dev/reports/rp_c89210d0846b746f) | windows+amd+amd-proprietary+dgpu `AMD Radeon RX 9070 XT, 2.0.416` `AMD Ryzen 7 5800X3D 8-Core Processor, 32 GB RAM` | | `windows+nvidia` | [1](https://beanstalk.magicbean.dev/reports/rp_9993b02c2c799c3b) | windows+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce RTX 4080, 621.1.0` `AMD Ryzen 7 9800X3D 8-Core Processor, 32 GB RAM` | | `linux+nvidia` | [1](https://beanstalk.magicbean.dev/reports/rp_58e348c5ebac64ff) | linux+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce GTX 1070, 580.196.7` `Intel(R) Core(TM) i7-8750H CPU @ 2.20GHz, 32 GB RAM` | | `linux+nvidia` | [2](https://beanstalk.magicbean.dev/reports/rp_40dabb02ecc42fd9) | linux+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce RTX 5070, 620.70.6` `AMD Ryzen 5 9600X 6-Core Processor, 32 GB RAM` | </details>
Author
Owner

Lab run (Linux, RX 580 RADV 26.2.6, i7-9700K 8 cores, 1080p, GPU-bound; this PR @ f08644e3, two interleaved game starts each, RX_VK_RECORDER_PARTS=0 as the baseline in the same build):

fps RX_VK_RECORDER_PARTS=0 this PR
Market standing 35.7 (36, 36) 36.6 (36, 37)
Market turning 53.8 (53, 54) 54.4 (54, 55)
Kokiri standing 51.9 (52, 52) 51.3 (51, 51)
Kokiri turning 41.3 (42, 41) 42.2 (44, 41)
Hyrule Field standing 46.4 46.3
Hyrule Field turning 51.1 50.3
CPU cores in use, Market standing 2.66 2.32
recorder hand-offs a frame 814 27
recorder spinning idle, s in 10 s 3.1 0.09

Even to slightly better in fps on a GPU-bound AMD desktop, with 10-16% less CPU in use across the scenes; no hang, no Vulkan error.

On the Test Farm's two slower results: the Steam Machine's Hyrule Field -7.4% is one run with a burst of 44 hitches (p99 69 ms, 44.0 fps) while its second run matched main (49.7 against 50.3 and 48.6); its render thread is idle 22 of 35 ms a frame there. The GTX 1070's -2.8% is a single run of the same size. A second farm run with three starts a side is requested.

Lab run (Linux, RX 580 RADV 26.2.6, i7-9700K 8 cores, 1080p, GPU-bound; this PR @ f08644e3, two interleaved game starts each, `RX_VK_RECORDER_PARTS=0` as the baseline in the same build): | fps | `RX_VK_RECORDER_PARTS=0` | this PR | |---|---|---| | Market standing | 35.7 (36, 36) | 36.6 (36, 37) | | Market turning | 53.8 (53, 54) | 54.4 (54, 55) | | Kokiri standing | 51.9 (52, 52) | 51.3 (51, 51) | | Kokiri turning | 41.3 (42, 41) | 42.2 (44, 41) | | Hyrule Field standing | 46.4 | 46.3 | | Hyrule Field turning | 51.1 | 50.3 | | CPU cores in use, Market standing | 2.66 | 2.32 | | recorder hand-offs a frame | 814 | 27 | | recorder spinning idle, s in 10 s | 3.1 | 0.09 | Even to slightly better in fps on a GPU-bound AMD desktop, with 10-16% less CPU in use across the scenes; no hang, no Vulkan error. On the Test Farm's two slower results: the Steam Machine's Hyrule Field -7.4% is one run with a burst of 44 hitches (p99 69 ms, 44.0 fps) while its second run matched main (49.7 against 50.3 and 48.6); its render thread is idle 22 of 35 ms a frame there. The GTX 1070's -2.8% is a single run of the same size. A second farm run with three starts a side is requested.
HumanBean changed title from WIP: perf(vk): the recorder takes a whole part and submits it, so the render thread never waits at a part's end to perf(vk): the recorder takes a whole part and submits it, so the render thread never waits at a part's end 2026-11-20 17:12:27 +00:00
HumanBean deleted branch perf/recorder-parts 2026-11-20 17:12:49 +00:00
Sign in to join this conversation.
No reviewers
No milestone
No project
No assignees
2 participants
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference
MagicBeanDev/Longshot!1866
No description provided.