WIP: perf(vk): over the hold's line the render target pool is given back before any texture is evicted #1104

Draft
GummiNaomi wants to merge 1 commit from GummiNaomi/Longshot:perf/pool-first-eviction into main
Member

Type: perf · Risk: low · Tested on: Windows 11, RTX 4070 SUPER (NVIDIA 617.42) · Needs: an 8 GB card; a 6 GB card; Linux; AMD; a Steam Deck

Why

When VRAM runs short, Longshot throws out textures that the game is still using, while hundreds of MB of idle render target images sit in the render target pool. At a 9.5 GB allowance (a 12 GB card with a browser and Discord open), on Castle Town's gate route the compaction's hold evicted 2,538 textures in 140 s, and 1,876 of them (74 %) were needed again and made again, most within 30 s. Each one is read, decoded and uploaded again; a pooled render target image costs a fraction of a millisecond to make again.

The pool's bound grew from 256 MB to up to 768 MB in #994, so this matters more now: the hold (RX_VK_COMPACT_PCT, 85 % of the allowance) counts its room from the free space in the texture blocks only, and the pool is only given back above 90 % (pool_pressed), after textures have already gone.

Follows #994 (merged) and its issue #995.

What changed

When VRAM is short, the render target pool's idle images are now given back before any texture is evicted, and the room they hold counts toward what the hold needs. Textures are evicted only for what the pool can't cover.

  • vk: in texture_frame's compaction hold (gpu/vk/vk_texture.cpp), the pool's idle images count as room the hold need not take from the textures (reclaim += idle), and when the pressure is over the hold's line the pool is given back first (pool_pressed = true, as above 90 % today). Textures are evicted only for what the pool can't cover. RX_VK_POOL_FIRST=0 turns it off (as before).
  • tests: tests/test_target_image_pool.py checks the order (the pool counted before vram_compact::held_budget).
  • docs: RX_VK_POOL_FIRST in docs/development.md, in the pool's row.

Measurements

Verdict: mixed, so a draft. Textures evicted fall 83-86 % in both rounds, and the ones made again from 1,876 to 1 (round 1). Slow frames fall in round 1 and stay about the same in round 2. The mean fps is 8-9 % lower with the change in both rounds. Part of that is the route's timing on a wall clock (the mean weighs the 200+ fps field against the 70 fps town), but the field right after the gate (40-50 s) is lower in both rounds, where giving the pool back makes more images again. So on a 12 GB card the pool's churn may cost about what the evictions saved. The open question is whether that holds on 6-8 GB cards, where evictions are far more common. Giving back only the hold's excess (see Overlap) should keep the eviction cut with less churn.

Hardware: Windows 11 Pro 26200, RTX 4070 SUPER 12 GB (NVIDIA 617.42), Ryzen 9 9950X3D (16 cores, SMT off), 32 GB
Scene: save set "First Step into Hyrule Town", Castle Town's gate into Hyrule Field and back (gate-bridge-wall.txt, wall clock, 140 s), 2560x1440, DLAA, uncapped, the render thread on (Auto), RX_VK_TEST_ALLOWED_MB=9500 (the allowance this PC's driver gives with the user's usual programs open)
Method: headless runs, the two sides in turn in one build; slow frames, 1 % low and mean from the frame trace's last three quarters; textures evicted from the log's textures retired … budget N lines (summed); render target images from the log's render target images: line
How to reproduce: the save set and route above with RX_VK_TEST_ALLOWED_MB set a little above what the game uses in town (9500 on a 12 GB card); RX_VK_POOL_FIRST=0 against the default

Before, round 1, RX_VK_POOL_FIRST=0: this PR @ a2cdf4e5 After, round 1: this PR @ a2cdf4e5 Before, round 2, RX_VK_POOL_FIRST=0: this PR @ 1909c292 After, round 2: this PR @ 1909c292
Textures evicted for the budget (140 s) 2,538 347 4,076 685
Of them made again (TEMP eviction log, round 1 only) 1,876 1
Frames over 25 / 50 / 100 ms 123 / 49 / 17 88 / 36 / 13 100 / 43 / 12 107 / 43 / 12
1 % low 47.7 fps 49.6 fps 52.6 fps 48.9 fps
Mean fps 150.1 138.0 159.8 146.0
Field after the gate, 40-50 s 164 fps 153 fps 204 fps 162 fps
Render target images made new 3,357 3,128 2,963 3,257

Baseline: the same build with RX_VK_POOL_FIRST=0, which is main's behaviour, so both sides ran the same binary and only the order of giving back changed; a separate main build would add build-to-build noise to a 140 s route.

Builds: round 1 on #994's branch a2cdf4e5 with this change (v3.0.0-rc.4-140-ga2cdf4e5-dirty, the eviction counts from a TEMP log not in this PR); round 2 on this PR's change on main 02903b04 (v3.0.0-rc.5-6-g1909c292, the same code as this PR's commit before a comment's numbers and the docs/tested files were added; since rebased onto main d76feb2d, which doesn't touch gpu/vk/vk_texture.cpp). Both sides of a round are the same build, run in turn.

fps by 10 s slice (wall clock), to show where the means differ:

pf-old-1   0:477 1:331 2:70 3:71 4:164 5:196 6:145 7:71 8:246 9:71 10:259 11:111 12:212 13:69
pf-new-1   0:483 1:313 2:70 3:70 4:153 5:190 6:128 7:71 8:72 9:259 10:70 11:72 12:255 13:117
pfpr-old-1 0:539 1:610 2:68 3:72 4:204 5:203 6:142 7:70 8:72 9:200 10:137 11:72 12:138 13:229
pfpr-new-1 0:507 1:381 2:71 3:72 4:162 5:202 6:140 7:70 8:72 9:193 10:144 11:174 12:144 13:71

Flags and settings

One new switch, on by default.

Flag What it is Before this PR After this PR Effect Helps / doesn't help / untested Recommendation
RX_VK_POOL_FIRST the render target pool given back before textures at the hold's line (no switch: textures first) 1 83-86 % fewer textures evicted at a 9.5 GB allowance; mean fps 8-9 % lower on this 12 GB card helps evictions on the RTX 4070 SUPER; frame rate mixed; untested on 6-8 GB cards 1 for testing; 0 if a card gets slower

Overlap with other PRs

  • #1035 (flix331, WIP), 3-tier render target pool pressure: changes the same pool_pressed rule in the other direction. Today the pool is emptied above 90 %; #1035 keeps half of it up to 97 % so render targets stay reusable on 6 GB cards. This PR gives it back earlier, from the hold's 85 % line, so no texture goes while idle pool images stay. They don't have to conflict: a version that gives back only as much of the pool as the hold is over (and halves rather than empties it when that is enough) would keep both aims. Happy to rework this one that way, or to fold it into #1035, whichever the maintainers and flix331 prefer. #1035 also predates #994, so it still assumes the fixed 256 MB bound.
  • #1057 (flix331, WIP), dynamic texture floor for cards under 7 GB: changes the hold's budget arithmetic in the same function (texture_frame), a different part of it. Likely a textual merge conflict, not a conflict of aims; whichever lands second rebases.
  • #943 (Amateurwizrd, WIP), render targets let go as newer ones cover them and pooled at once: more images reach the pool sooner, so more for this PR to give back before textures; complementary.
  • #875 (HumanBean, WIP), the eviction's waits in seconds below 60 fps: when the hold evicts, not what; complementary.
  • #994 (merged): the pool's larger bound and per-present dead-target sweep that this builds on.

Testing

Tested: Windows 11, RTX 4070 SUPER, NVIDIA 617.42: the runs above; tests/test_target_image_pool.py passes (10 tests); python -I -B tests/run_all.py: 223 passed, 62 failed, 379 skipped, and main 02903b04 on the same PC gives the same counts and the same failing tests (this PC's toolchain: Windows line endings, fixtures needing gcc or clang++, which weren't on the path for this run). The same change was in the build played in Castle Town and Hyrule Field on this PC (with #994's settings) without a visible change in the picture.

Not tested: a real 6 or 8 GB card (only a lowered allowance on a 12 GB one), AMD and Intel GPUs, Linux, a Steam Deck.

Must be tested:

  • An 8 GB card, Windows or Linux: Castle Town's market and the gate into Hyrule Field and back for a few minutes, the FPS counter on, then the same with RX_VK_POOL_FIRST=0, expecting fewer textures retired … budget lines in the log and no more dips (by hand)
  • A 6 GB card: the same, expecting the same; and with #1035 if it lands first (by hand)
  • Linux, any GPU: the same, expecting the same (Test Farm: @Beanstalk linux)
  • A Steam Deck: the same, expecting the same (by hand)

Risks

What could break: on a card at its limit the pool now empties earlier, from 85 % instead of 90 %, so some render targets that would have been reused are made new and the pool's images are destroyed (pool_trim, up to 64 a present). On this 12 GB card that looks to cost about what the evictions saved (the mean fps above). If it costs more than the evictions on a given card, frames are slower there. Nothing changes below the hold's line. The picture can't change: an evicted texture or a destroyed pool image is made again the same way as before.

Turning it off: RX_VK_POOL_FIRST=0.

Compatibility: Nothing changes.

Issues

Related: #995 (render target churn, fixed by #994). Overlaps #1035, #1057.

Checklist

Everyone

  • One topic.
  • The title is a Conventional Commit (type(scope): description) of the status line's type.
  • Why gives the reason plainly and links the issues that fit (the tracker searched).
  • Measurements has a verdict, or one line saying why there is nothing to measure.
  • Testing names what was tested, what wasn't, and what must still be tested, where and how.
  • Rebased on current main.
  • python -I -B tests/run_all.py passes (any failure that also fails on main is named).
  • No game files, dumps, build output or generated files.
  • third_party/ untouched (a new folder with a PROVENANCE.md instead).
  • The license is OK: GPL-3.0-or-later, nothing incompatible and nothing decompiled.

When it applies

  • AI assistance recorded with Assisted-by: / Co-Authored-By: trailers.
  • Numbers from the same PC, scene and settings, at least 2 interleaved runs each, the median reported.
  • Switches documented (docs/development.md; the README for a menu row) and shipped at their recommended value.
  • A file in docs/tested/ for each new measured run (format in docs/tested-configurations.md).
  • A check under tests/ that reads changed source updated in this PR, not weakened.
  • Docs updated.
  • Host code builds on both Windows (llvm-mingw) and Linux, with UTF-8 paths. (Built on Windows only.)

🤖 Generated with Claude Code

**Type:** perf · **Risk:** low · **Tested on:** Windows 11, RTX 4070 SUPER (NVIDIA 617.42) · **Needs:** an 8 GB card; a 6 GB card; Linux; AMD; a Steam Deck ## Why When VRAM runs short, Longshot throws out textures that the game is still using, while hundreds of MB of idle render target images sit in the render target pool. At a 9.5 GB allowance (a 12 GB card with a browser and Discord open), on Castle Town's gate route the compaction's hold evicted 2,538 textures in 140 s, and 1,876 of them (74 %) were needed again and made again, most within 30 s. Each one is read, decoded and uploaded again; a pooled render target image costs a fraction of a millisecond to make again. The pool's bound grew from 256 MB to up to 768 MB in #994, so this matters more now: the hold (`RX_VK_COMPACT_PCT`, 85 % of the allowance) counts its room from the free space in the texture blocks only, and the pool is only given back above 90 % (`pool_pressed`), after textures have already gone. Follows #994 (merged) and its issue #995. ## What changed When VRAM is short, the render target pool's idle images are now given back before any texture is evicted, and the room they hold counts toward what the hold needs. Textures are evicted only for what the pool can't cover. - vk: in `texture_frame`'s compaction hold (`gpu/vk/vk_texture.cpp`), the pool's idle images count as room the hold need not take from the textures (`reclaim += idle`), and when the pressure is over the hold's line the pool is given back first (`pool_pressed = true`, as above 90 % today). Textures are evicted only for what the pool can't cover. `RX_VK_POOL_FIRST=0` turns it off (as before). - tests: `tests/test_target_image_pool.py` checks the order (the pool counted before `vram_compact::held_budget`). - docs: `RX_VK_POOL_FIRST` in `docs/development.md`, in the pool's row. ## Measurements **Verdict:** mixed, so a draft. Textures evicted fall 83-86 % in both rounds, and the ones made again from 1,876 to 1 (round 1). Slow frames fall in round 1 and stay about the same in round 2. The mean fps is 8-9 % lower with the change in both rounds. Part of that is the route's timing on a wall clock (the mean weighs the 200+ fps field against the 70 fps town), but the field right after the gate (40-50 s) is lower in both rounds, where giving the pool back makes more images again. So on a 12 GB card the pool's churn may cost about what the evictions saved. The open question is whether that holds on 6-8 GB cards, where evictions are far more common. Giving back only the hold's excess (see Overlap) should keep the eviction cut with less churn. Hardware: Windows 11 Pro 26200, RTX 4070 SUPER 12 GB (NVIDIA 617.42), Ryzen 9 9950X3D (16 cores, SMT off), 32 GB Scene: save set "First Step into Hyrule Town", Castle Town's gate into Hyrule Field and back (gate-bridge-wall.txt, wall clock, 140 s), 2560x1440, DLAA, uncapped, the render thread on (Auto), `RX_VK_TEST_ALLOWED_MB=9500` (the allowance this PC's driver gives with the user's usual programs open) Method: headless runs, the two sides in turn in one build; slow frames, 1 % low and mean from the frame trace's last three quarters; textures evicted from the log's `textures retired … budget N` lines (summed); render target images from the log's `render target images:` line How to reproduce: the save set and route above with `RX_VK_TEST_ALLOWED_MB` set a little above what the game uses in town (9500 on a 12 GB card); `RX_VK_POOL_FIRST=0` against the default | | Before, round 1, RX_VK_POOL_FIRST=0: this PR @ a2cdf4e5 | After, round 1: this PR @ a2cdf4e5 | Before, round 2, RX_VK_POOL_FIRST=0: this PR @ 1909c292 | After, round 2: this PR @ 1909c292 | |---|---|---|---|---| | Textures evicted for the budget (140 s) | 2,538 | 347 | 4,076 | 685 | | Of them made again (TEMP eviction log, round 1 only) | 1,876 | 1 | | | | Frames over 25 / 50 / 100 ms | 123 / 49 / 17 | 88 / 36 / 13 | 100 / 43 / 12 | 107 / 43 / 12 | | 1 % low | 47.7 fps | 49.6 fps | 52.6 fps | 48.9 fps | | Mean fps | 150.1 | 138.0 | 159.8 | 146.0 | | Field after the gate, 40-50 s | 164 fps | 153 fps | 204 fps | 162 fps | | Render target images made new | 3,357 | 3,128 | 2,963 | 3,257 | Baseline: the same build with `RX_VK_POOL_FIRST=0`, which is main's behaviour, so both sides ran the same binary and only the order of giving back changed; a separate main build would add build-to-build noise to a 140 s route. Builds: round 1 on #994's branch a2cdf4e5 with this change (`v3.0.0-rc.4-140-ga2cdf4e5-dirty`, the eviction counts from a TEMP log not in this PR); round 2 on this PR's change on main 02903b04 (`v3.0.0-rc.5-6-g1909c292`, the same code as this PR's commit before a comment's numbers and the `docs/tested` files were added; since rebased onto main d76feb2d, which doesn't touch `gpu/vk/vk_texture.cpp`). Both sides of a round are the same build, run in turn. fps by 10 s slice (wall clock), to show where the means differ: ``` pf-old-1 0:477 1:331 2:70 3:71 4:164 5:196 6:145 7:71 8:246 9:71 10:259 11:111 12:212 13:69 pf-new-1 0:483 1:313 2:70 3:70 4:153 5:190 6:128 7:71 8:72 9:259 10:70 11:72 12:255 13:117 pfpr-old-1 0:539 1:610 2:68 3:72 4:204 5:203 6:142 7:70 8:72 9:200 10:137 11:72 12:138 13:229 pfpr-new-1 0:507 1:381 2:71 3:72 4:162 5:202 6:140 7:70 8:72 9:193 10:144 11:174 12:144 13:71 ``` ### Flags and settings One new switch, on by default. | Flag | What it is | Before this PR | After this PR | Effect | Helps / doesn't help / untested | Recommendation | |---|---|---|---|---|---|---| | `RX_VK_POOL_FIRST` | the render target pool given back before textures at the hold's line | (no switch: textures first) | 1 | 83-86 % fewer textures evicted at a 9.5 GB allowance; mean fps 8-9 % lower on this 12 GB card | helps evictions on the RTX 4070 SUPER; frame rate mixed; untested on 6-8 GB cards | 1 for testing; `0` if a card gets slower | ## Overlap with other PRs - **#1035** (flix331, WIP), 3-tier render target pool pressure: changes the same `pool_pressed` rule in the other direction. Today the pool is emptied above 90 %; #1035 keeps half of it up to 97 % so render targets stay reusable on 6 GB cards. This PR gives it back earlier, from the hold's 85 % line, so no texture goes while idle pool images stay. They don't have to conflict: a version that gives back only as much of the pool as the hold is over (and halves rather than empties it when that is enough) would keep both aims. Happy to rework this one that way, or to fold it into #1035, whichever the maintainers and flix331 prefer. #1035 also predates #994, so it still assumes the fixed 256 MB bound. - **#1057** (flix331, WIP), dynamic texture floor for cards under 7 GB: changes the hold's budget arithmetic in the same function (`texture_frame`), a different part of it. Likely a textual merge conflict, not a conflict of aims; whichever lands second rebases. - **#943** (Amateurwizrd, WIP), render targets let go as newer ones cover them and pooled at once: more images reach the pool sooner, so more for this PR to give back before textures; complementary. - **#875** (HumanBean, WIP), the eviction's waits in seconds below 60 fps: when the hold evicts, not what; complementary. - **#994** (merged): the pool's larger bound and per-present dead-target sweep that this builds on. ## Testing **Tested:** Windows 11, RTX 4070 SUPER, NVIDIA 617.42: the runs above; `tests/test_target_image_pool.py` passes (10 tests); `python -I -B tests/run_all.py`: 223 passed, 62 failed, 379 skipped, and main 02903b04 on the same PC gives the same counts and the same failing tests (this PC's toolchain: Windows line endings, fixtures needing gcc or clang++, which weren't on the path for this run). The same change was in the build played in Castle Town and Hyrule Field on this PC (with #994's settings) without a visible change in the picture. **Not tested:** a real 6 or 8 GB card (only a lowered allowance on a 12 GB one), AMD and Intel GPUs, Linux, a Steam Deck. **Must be tested:** - An 8 GB card, Windows or Linux: Castle Town's market and the gate into Hyrule Field and back for a few minutes, the FPS counter on, then the same with `RX_VK_POOL_FIRST=0`, expecting fewer `textures retired … budget` lines in the log and no more dips (by hand) - A 6 GB card: the same, expecting the same; and with #1035 if it lands first (by hand) - Linux, any GPU: the same, expecting the same (Test Farm: @Beanstalk linux) - A Steam Deck: the same, expecting the same (by hand) ## Risks **What could break:** on a card at its limit the pool now empties earlier, from 85 % instead of 90 %, so some render targets that would have been reused are made new and the pool's images are destroyed (`pool_trim`, up to 64 a present). On this 12 GB card that looks to cost about what the evictions saved (the mean fps above). If it costs more than the evictions on a given card, frames are slower there. Nothing changes below the hold's line. The picture can't change: an evicted texture or a destroyed pool image is made again the same way as before. **Turning it off:** `RX_VK_POOL_FIRST=0`. **Compatibility:** Nothing changes. ## Issues Related: #995 (render target churn, fixed by #994). Overlaps #1035, #1057. ## Checklist ### Everyone - [x] One topic. - [x] The title is a Conventional Commit (`type(scope): description`) of the status line's type. - [x] Why gives the reason plainly and links the issues that fit (the tracker searched). - [x] Measurements has a verdict, or one line saying why there is nothing to measure. - [x] Testing names what was tested, what wasn't, and what must still be tested, where and how. - [x] Rebased on current `main`. - [x] `python -I -B tests/run_all.py` passes (any failure that also fails on `main` is named). - [x] No game files, dumps, build output or generated files. - [x] `third_party/` untouched (a new folder with a `PROVENANCE.md` instead). - [x] The license is OK: GPL-3.0-or-later, nothing incompatible and nothing decompiled. ### When it applies - [x] AI assistance recorded with `Assisted-by:` / `Co-Authored-By:` trailers. - [x] Numbers from the same PC, scene and settings, at least 2 interleaved runs each, the median reported. - [x] Switches documented (docs/development.md; the README for a menu row) and shipped at their recommended value. - [x] A file in `docs/tested/` for each new measured run (format in `docs/tested-configurations.md`). - [x] A check under `tests/` that reads changed source updated in this PR, not weakened. - [x] Docs updated. - [ ] Host code builds on both Windows (llvm-mingw) and Linux, with UTF-8 paths. (Built on Windows only.) 🤖 Generated with [Claude Code](https://claude.com/claude-code)
perf(vk): over the hold's line the render target pool is given back before any texture is evicted
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 / Conventional Commit title (pull_request) Successful in 16s
Website / Website on Cloudflare (pull_request_target) Has been skipped
Guide screenshots / Guide images and local UI source review (pull_request) Failing after 26s
CI / Workflow lint (actionlint) (pull_request) Successful in 27s
Third-party parts / As their notes say (pull_request) Successful in 37s
PR description / Description follows the template (pull_request_target) Successful in 9s
beanstalk/approval approved by HumanBean (reviewed)
beanstalk/vram6gb 0/1 pass, 1 error
beanstalk/vram8gb 1/1 fail
beanstalk/linux 0/1 pass, 1 error
beanstalk/amd 1/1 fail
beanstalk/steamdeck 0/1 pass, 0 running, 1 queued
a416edb056
With the pool holding up to 768 MB of idle render target images, a card near
the compaction's hold line evicted textures while the pool kept its images: at
a 9.5 GB allowance (a 12 GB card with other programs open) the gate route
evicted 2,538 textures in 140 s and made 1,876 of them again. A pooled image
is made again in a fraction of a millisecond; an evicted texture is read,
decoded and uploaded again.

Now the pool's idle images count as room the hold need not take from the
textures, and over the hold's line the pool is given back first
(pool_pressed). Two rounds in turn, RX_VK_POOL_FIRST=0 against the change:
textures evicted 2,538 -> 347 and 4,076 -> 685; frames over 25 ms 123 -> 88
and 100 -> 107; mean fps 150 -> 138 and 160 -> 146, partly the route's
timing, partly more render target images made again. RX_VK_POOL_FIRST=0: as
before.

Assisted-by: Claude Opus 5.5
Co-Authored-By: Claude Opus 5.5 <[email protected]>
dyst left a comment

Nice, tight change. Giving idle pool images back before live textures is the right order, and the same-binary A/B makes the trade easy to see. Three things from the code:

  1. The give-back flips on and off every look. pool_pressed is set again at every look from the 90 % rule (vk_texture.cpp:4492), and the new line raises it only while idle is non-zero. Between 85 and 90 %, a look with idle > 0 gives the pool back and pool_trim empties it. The next look sees idle ≈ 0 and lowers the flag, so the pool refills, and the cycle repeats every 20 presents. On the idle-free looks the hold also evicts for the full excess as before. This may be where the extra images made new and the lower field fps come from. Trimming only the excess and holding it while over the line (your Overlap idea) would fix it.
  2. The credit assumes used drops by idle. Pooled images share the render targets' 128 MB blocks with live targets (vk_vma.cpp:52-56), and VMA keeps a block until it is empty. reclaimable leaves that space out for this reason (vram_compact.h:93-97). Logging used before and after a give-back would show how much really comes off.
  3. The default. The one card measured is 8-9 % slower on the mean, with a worse 1 % low in round 2. I'd ship it off until a 6-8 GB card shows a win.

Overlap with #875: no textual conflict in vk_texture.cpp. They're complementary: this PR changes how much the hold takes, #875 how old a texture must be.

This is a code read only. Nothing was built or run.

🤖 Generated with Claude Code

Nice, tight change. Giving idle pool images back before live textures is the right order, and the same-binary A/B makes the trade easy to see. Three things from the code: 1. **The give-back flips on and off every look.** `pool_pressed` is set again at every look from the 90 % rule (vk_texture.cpp:4492), and the new line raises it only while `idle` is non-zero. Between 85 and 90 %, a look with idle > 0 gives the pool back and `pool_trim` empties it. The next look sees idle ≈ 0 and lowers the flag, so the pool refills, and the cycle repeats every 20 presents. On the idle-free looks the hold also evicts for the full excess as before. This may be where the extra images made new and the lower field fps come from. Trimming only the excess and holding it while over the line (your Overlap idea) would fix it. 2. **The credit assumes `used` drops by `idle`.** Pooled images share the render targets' 128 MB blocks with live targets (vk_vma.cpp:52-56), and VMA keeps a block until it is empty. `reclaimable` leaves that space out for this reason (vram_compact.h:93-97). Logging `used` before and after a give-back would show how much really comes off. 3. **The default.** The one card measured is 8-9 % slower on the mean, with a worse 1 % low in round 2. I'd ship it off until a 6-8 GB card shows a win. Overlap with #875: no textual conflict in `vk_texture.cpp`. They're complementary: this PR changes how much the hold takes, #875 how old a texture must be. This is a code read only. Nothing was built or run. 🤖 Generated with [Claude Code](https://claude.com/claude-code)
@ -4509,0 +4518,4 @@
std::lock_guard lk(pool_mu);
idle = pool_bytes;
}
if (idle && pressure > allowed / 100 * pct) pool_pressed = true;
Member

Because of the idle &&, this only holds while the pool still has something in it. pool_pressed is reassigned from the 90 % rule at every look (line 4492). So once pool_trim has emptied the pool, the next look clears the flag, the pool refills over 10 presents, and the look after that sets it again. Between 85 and 90 % the give-back alternates every look. Could the flag key off the count alone (over the hold's line) and pool_trim take a cap of pool_bytes - over rather than 0? Then only the excess goes and it stays gone.

Because of the `idle &&`, this only holds while the pool still has something in it. `pool_pressed` is reassigned from the 90 % rule at every look (line 4492). So once `pool_trim` has emptied the pool, the next look clears the flag, the pool refills over 10 presents, and the look after that sets it again. Between 85 and 90 % the give-back alternates every look. Could the flag key off the count alone (over the hold's line) and `pool_trim` take a cap of `pool_bytes - over` rather than 0? Then only the excess goes and it stays gone.
@ -4509,0 +4519,4 @@
idle = pool_bytes;
}
if (idle && pressure > allowed / 100 * pct) pool_pressed = true;
reclaim += idle;
Member

This credits all of idle as room. The pooled images sit in target_image_pool's 128 MB blocks beside live render targets, and VMA keeps a block until it empties (vk_vma.cpp:224). Destroying them may not lower used by this much. That is why reclaimable leaves other blocks' free space out (vram_compact.h:93-97). Worth logging used before and after a give-back, or crediting only what actually came off at the previous look.

This credits all of `idle` as room. The pooled images sit in `target_image_pool`'s 128 MB blocks beside live render targets, and VMA keeps a block until it empties (vk_vma.cpp:224). Destroying them may not lower `used` by this much. That is why `reclaimable` leaves other blocks' free space out (vram_compact.h:93-97). Worth logging `used` before and after a give-back, or crediting only what actually came off at the previous look.
Owner

Tested on Linux, as asked under Must be tested for a 6–8 GB card: this PR's commit rebased onto main @ 43f40b87 (no conflicts), on a 6 GB card's rules.

Hardware: RTX 3090, NVIDIA 620.70.06, CachyOS.
Settings: DLAA 2560x1440 with RX_VK_TEST_VRAM_MB=6144 RX_VK_TEST_ALLOWED_MB=5233. RX_VK_TEST_VRAM_MB makes the renderer size everything for a 6 GB card, including the starting blocks, texture budget and mirrors.
Route: one session each: Hyrule Field, then the Market (teleports), walking and then turning. One build, RX_VK_POOL_FIRST default against =0, plus two main runs.

main (2 runs) this PR, RX_VK_POOL_FIRST=0 this PR (default)
Textures retired by the budget (sum of budget lines) 40,608 / 44,019 50,582 47,082
Render target images made new (last pool line) 24,783 / 31,376 32,447 32,386
GPU ms, Field turning / Market walking / Market turning 13.50, 12.30, 11.10 / 11.25, 12.10, 10.90 11.25, 12.40, 11.20 10.80, 12.40, 10.50
Market turning fps 45.8 / 32.5 44.4 48.5
VRAM peak (MB) 5,428 / 5,379 5,748 6,196

On this route, with a 6 GB card's rules, the pool given back first cut the textures retired by 7% (50,582 to 47,082) on the same build. That is inside the spread between main's two runs, and nowhere near the 2,538 to 347 of the description. Render targets were made new as often either way, and GPU time and fps are the same within this PC's noise. This PC swaps heavily, so the frame-time tails are noisy, which is why I compare GPU time.

So it neither helps nor hurts here. The route is different from yours (Field and the Market rather than the gate), and the pressure on a real 6 GB card may differ. I didn't see the fps drop of your rounds either.

🤖 Generated with Claude Code

Tested on Linux, as asked under Must be tested for a 6–8 GB card: this PR's commit rebased onto `main` @ 43f40b87 (no conflicts), on a 6 GB card's rules. **Hardware:** RTX 3090, NVIDIA 620.70.06, CachyOS. **Settings:** DLAA 2560x1440 with `RX_VK_TEST_VRAM_MB=6144 RX_VK_TEST_ALLOWED_MB=5233`. `RX_VK_TEST_VRAM_MB` makes the renderer size everything for a 6 GB card, including the starting blocks, texture budget and mirrors. **Route:** one session each: Hyrule Field, then the Market (teleports), walking and then turning. One build, `RX_VK_POOL_FIRST` default against `=0`, plus two main runs. | | main (2 runs) | this PR, `RX_VK_POOL_FIRST=0` | this PR (default) | |---|---|---|---| | Textures retired by the budget (sum of `budget` lines) | 40,608 / 44,019 | 50,582 | 47,082 | | Render target images made new (last pool line) | 24,783 / 31,376 | 32,447 | 32,386 | | GPU ms, Field turning / Market walking / Market turning | 13.50, 12.30, 11.10 / 11.25, 12.10, 10.90 | 11.25, 12.40, 11.20 | 10.80, 12.40, 10.50 | | Market turning fps | 45.8 / 32.5 | 44.4 | 48.5 | | VRAM peak (MB) | 5,428 / 5,379 | 5,748 | 6,196 | On this route, with a 6 GB card's rules, the pool given back first cut the textures retired by 7% (50,582 to 47,082) on the same build. That is inside the spread between main's two runs, and nowhere near the 2,538 to 347 of the description. Render targets were made new as often either way, and GPU time and fps are the same within this PC's noise. This PC swaps heavily, so the frame-time tails are noisy, which is why I compare GPU time. So it neither helps nor hurts here. The route is different from yours (Field and the Market rather than the gate), and the pressure on a real 6 GB card may differ. I didn't see the fps drop of your rounds either. 🤖 Generated with [Claude Code](https://claude.com/claude-code)
Owner

@Beanstalk

@Beanstalk
Member

Beanstalk test farm

Commit a416edb · tier reviewed · approved by HumanBean · dashboard

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

The plan compares the Market and Hyrule Field on 6 GB and 8 GB cards, Linux, AMD, and Steam Deck, taking roughly 25–30 minutes per tester. It lowers the compaction threshold to exercise memory pressure and tests the new pool-first switch both on and off. It measures frame times, FPS, VRAM and load times and compares screenshots; crashes, substantial performance or memory regressions, and visual differences fail the plan. The available probes do not verify eviction ordering or count texture recreations.

  • Where: vram6gb (any PC with 6 GB VRAM, 1 report), vram8gb (any PC with 8 GB VRAM, 1 report), linux (1 report), amd (1 report), steamdeck (1 report).
  • Matrix: 2 cells, each on main and the PR: RX_VK_POOL_FIRST 0 or 1 (PR only).
  • market-field-pressure: from a new save (settings frame_rate_cap=6; switches RX_VK_COMPACT_PCT=50), then teleport to Market, move the stick, measure market for 30 s (after 5 s warm-up), take the picture market to compare with main, teleport to Hyrule Field, measure field for 30 s (after 5 s warm-up), take the picture field to compare with main, measure return for 20 s (after 5 s warm-up); 1 run on each side for each of 2 cells (2 game starts a run).
  • Measured: crashes, frame times, stutters and hitches, draws and frame rate, VRAM use, load times.
  • 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 peak VRAM is more than 15% above main; or longest load is more than 30% above main; or frame time (p99) is more than 20% above main.
  • Time: about 31 min of game runs per tester (4 game starts), plus the two builds.

Started. Run 2: the farm's reports were cleared after its Windows and Linux fixes: started again.

Target Reports Crash Frame p99 (ms) vs main VRAM peak (MB) vs main Result
vram6gb · 1: linux+amd+radv+igpu+handheld AMD Custom GPU 0932 (RADV VANGOGH), 26.1.99 1/1 1 none — — error
vram8gb · 1: windows+nvidia+nvidia-proprietary+dgpu NVIDIA GeForce RTX 3070 Laptop GPU, 621.24.0 1/1 1 1 PR run — — fail
linux · 1: linux+amd+radv+dgpu AMD Radeon RX 7900 XTX (RADV NAVI31), 26.2.7 1/1 1 none — — error
amd · 1: windows+amd+amd-proprietary+dgpu AMD Radeon RX 9070 XT, 2.0.416 1/1 1 none 18.4 vs 18.9 (−3%) 8,237 vs 7,787 (+6%) fail
steamdeck 0/1 — — — queued
Scenario · matrix cell Target Frame p99 (ms) vs main Result
market-field-pressure · RX_VK_POOL_FIRST=0 (PR only) vram6gb — pass
market-field-pressure · RX_VK_POOL_FIRST=0 (PR only) vram8gb — pass
market-field-pressure · RX_VK_POOL_FIRST=0 (PR only) linux — pass
market-field-pressure · RX_VK_POOL_FIRST=0 (PR only) amd 18.4 vs 18.9 (−3%) fail
market-field-pressure · RX_VK_POOL_FIRST=0 (PR only) steamdeck — —
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) vram6gb — pass
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) vram8gb — fail
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) linux — pass
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) amd 17.5 vs 17.2 (+2%) fail
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) steamdeck — —

Measurements · vram6gb

Verdict: could not be judged on vram6gb (1/1 report).

Hardware: 1: linux+amd+radv+igpu+handheld AMD Custom GPU 0932 (RADV VANGOGH), 26.1.99 AMD Custom APU 0932, 16 GB RAM
Scene: market-field-pressure, from a new save (frame_rate_cap=6, RX_VK_COMPACT_PCT=50); 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); market-field-pressure: 1 game start a side, market 30 s after 5 s warm-up, field 30 s after 5 s warm-up, return 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 @ d76feb2 After: this PR @ a416edb
market-field-pressure · RX_VK_POOL_FIRST=0 (PR only) · longest load (s) 21.2 6.8 (−68%)
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) · longest load (s) 29.0 28.9 (−0%)

Flags and settings

Flag What it is Values tried Before this PR After this PR Effect (frame rate) Helps / doesn't help / untested
RX_VK_POOL_FIRST switch the PR adds (set on the PR side only) 0, 1 not in main the PR's default market-field-pressure, main without it → the PR with 0: not measured untested
Markdown for the PR description
## Measurements

**Verdict:** could not be judged on vram6gb (1/1 report).

Hardware: 1: linux+amd+radv+igpu+handheld `AMD Custom GPU 0932 (RADV VANGOGH), 26.1.99` `AMD Custom APU 0932, 16 GB RAM`  
Scene: market-field-pressure, from a new save (frame_rate_cap=6, RX_VK_COMPACT_PCT=50); 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); market-field-pressure: 1 game start a side, market 30 s after 5 s warm-up, field 30 s after 5 s warm-up, return 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 @ d76feb2 | After: this PR @ a416edb |
|---|---|---|
| market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · longest load (s) | 21.2 | 6.8 (−68%) |
| market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · longest load (s) | 29.0 | 28.9 (−0%) |

### Flags and settings

| Flag | What it is | Values tried | Before this PR | After this PR | Effect (frame rate) | Helps / doesn't help / untested |
|---|---|---|---|---|---|---|
| `RX_VK_POOL_FIRST` | switch the PR adds (set on the PR side only) | `0`, `1` | not in main | the PR's default | market-field-pressure, main without it → the PR with 0: not measured | untested |

Measurements · vram8gb

Verdict: fail on vram8gb (1/1 report): market-field-pressure crash failed.

Hardware: 1: windows+nvidia+nvidia-proprietary+dgpu NVIDIA GeForce RTX 3070 Laptop GPU, 621.24.0 AMD Ryzen 9 5900HX with Radeon Graphics, 32 GB RAM
Scene: market-field-pressure, from a new save (frame_rate_cap=6, RX_VK_COMPACT_PCT=50); 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); market-field-pressure: 1 game start a side, market 30 s after 5 s warm-up, field 30 s after 5 s warm-up, return 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 @ d76feb2 After: this PR @ a416edb
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) · field · frame rate 56.4 —
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) · market · frame rate 47.7 —
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) · return · frame rate 45.9 —
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) · field · peak VRAM (MB) 6,070 —
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) · market · peak VRAM (MB) 5,346 —
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) · return · peak VRAM (MB) 6,252 —
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) · longest load (s) 62.9 —
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) · field · frame time (p99) (ms) 26.4 —
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) · market · frame time (p99) (ms) 34.7 —
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) · return · frame time (p99) (ms) 33.9 —

Flags and settings

Flag What it is Values tried Before this PR After this PR Effect (frame rate) Helps / doesn't help / untested
RX_VK_POOL_FIRST switch the PR adds (set on the PR side only) 0, 1 not in main the PR's default market-field-pressure, main without it → the PR with 0: not measured untested
Markdown for the PR description
## Measurements

**Verdict:** fail on vram8gb (1/1 report): `market-field-pressure crash` failed.

Hardware: 1: windows+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce RTX 3070 Laptop GPU, 621.24.0` `AMD Ryzen 9 5900HX with Radeon Graphics, 32 GB RAM`  
Scene: market-field-pressure, from a new save (frame_rate_cap=6, RX_VK_COMPACT_PCT=50); 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); market-field-pressure: 1 game start a side, market 30 s after 5 s warm-up, field 30 s after 5 s warm-up, return 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 @ d76feb2 | After: this PR @ a416edb |
|---|---|---|
| market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · field · frame rate | 56.4 | — |
| market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · market · frame rate | 47.7 | — |
| market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · return · frame rate | 45.9 | — |
| market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · field · peak VRAM (MB) | 6,070 | — |
| market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · market · peak VRAM (MB) | 5,346 | — |
| market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · return · peak VRAM (MB) | 6,252 | — |
| market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · longest load (s) | 62.9 | — |
| market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · field · frame time (p99) (ms) | 26.4 | — |
| market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · market · frame time (p99) (ms) | 34.7 | — |
| market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · return · frame time (p99) (ms) | 33.9 | — |

### Flags and settings

| Flag | What it is | Values tried | Before this PR | After this PR | Effect (frame rate) | Helps / doesn't help / untested |
|---|---|---|---|---|---|---|
| `RX_VK_POOL_FIRST` | switch the PR adds (set on the PR side only) | `0`, `1` | not in main | the PR's default | market-field-pressure, main without it → the PR with 0: not measured | untested |

Measurements · linux

Verdict: could not be judged on linux (1/1 report).

Hardware: 1: linux+amd+radv+dgpu AMD Radeon RX 7900 XTX (RADV NAVI31), 26.2.7 AMD Ryzen 7 5800X3D 8-Core Processor, 64 GB RAM
Scene: market-field-pressure, from a new save (frame_rate_cap=6, RX_VK_COMPACT_PCT=50); 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); market-field-pressure: 1 game start a side, market 30 s after 5 s warm-up, field 30 s after 5 s warm-up, return 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 @ d76feb2 After: this PR @ a416edb
market-field-pressure · RX_VK_POOL_FIRST=0 (PR only) · longest load (s) 39.9 30.7 (−23%)
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) · longest load (s) 37.6 26.9 (−28%)

Flags and settings

Flag What it is Values tried Before this PR After this PR Effect (frame rate) Helps / doesn't help / untested
RX_VK_POOL_FIRST switch the PR adds (set on the PR side only) 0, 1 not in main the PR's default market-field-pressure, main without it → the PR with 0: not measured untested
Markdown for the PR description
## Measurements

**Verdict:** could not be judged on linux (1/1 report).

Hardware: 1: linux+amd+radv+dgpu `AMD Radeon RX 7900 XTX (RADV NAVI31), 26.2.7` `AMD Ryzen 7 5800X3D 8-Core Processor, 64 GB RAM`  
Scene: market-field-pressure, from a new save (frame_rate_cap=6, RX_VK_COMPACT_PCT=50); 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); market-field-pressure: 1 game start a side, market 30 s after 5 s warm-up, field 30 s after 5 s warm-up, return 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 @ d76feb2 | After: this PR @ a416edb |
|---|---|---|
| market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · longest load (s) | 39.9 | 30.7 (−23%) |
| market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · longest load (s) | 37.6 | 26.9 (−28%) |

### Flags and settings

| Flag | What it is | Values tried | Before this PR | After this PR | Effect (frame rate) | Helps / doesn't help / untested |
|---|---|---|---|---|---|---|
| `RX_VK_POOL_FIRST` | switch the PR adds (set on the PR side only) | `0`, `1` | not in main | the PR's default | market-field-pressure, main without it → the PR with 0: not measured | untested |

Measurements · amd

Verdict: fail on amd (1/1 report): market-field-pressure screenshot.max_diff 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: market-field-pressure, from a new save (frame_rate_cap=6, RX_VK_COMPACT_PCT=50); 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); market-field-pressure: 1 game start a side, market 30 s after 5 s warm-up, field 30 s after 5 s warm-up, return 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 @ d76feb2 After: this PR @ a416edb
market-field-pressure · RX_VK_POOL_FIRST=0 (PR only) · field · frame rate 83.6 82.1 (−2%)
market-field-pressure · RX_VK_POOL_FIRST=0 (PR only) · market · frame rate 60.5 63.3 (+5%)
market-field-pressure · RX_VK_POOL_FIRST=0 (PR only) · return · frame rate 54.9 55.9 (+2%)
market-field-pressure · RX_VK_POOL_FIRST=0 (PR only) · field · peak VRAM (MB) 7,787 8,237 (+6%)
market-field-pressure · RX_VK_POOL_FIRST=0 (PR only) · market · peak VRAM (MB) 8,274 8,149 (−2%)
market-field-pressure · RX_VK_POOL_FIRST=0 (PR only) · return · peak VRAM (MB) 7,218 7,284 (+1%)
market-field-pressure · RX_VK_POOL_FIRST=0 (PR only) · longest load (s) 17.6 10.1 (−43%)
market-field-pressure · RX_VK_POOL_FIRST=0 (PR only) · field · frame time (p99) (ms) 18.9 18.4 (−3%)
market-field-pressure · RX_VK_POOL_FIRST=0 (PR only) · market · frame time (p99) (ms) 24.2 22.9 (−5%)
market-field-pressure · RX_VK_POOL_FIRST=0 (PR only) · return · frame time (p99) (ms) 27.1 25.3 (−6%)
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) · field · frame rate 85.9 86.7 (+1%)
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) · market · frame rate 63.4 63.8 (+1%)
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) · return · frame rate 55.7 55.0 (−1%)
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) · field · peak VRAM (MB) 8,615 8,135 (−6%)
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) · market · peak VRAM (MB) 7,763 8,754 (+13%)
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) · return · peak VRAM (MB) 7,979 8,525 (+7%)
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) · longest load (s) 9.9 10.3 (+5%)
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) · field · frame time (p99) (ms) 17.2 17.5 (+2%)
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) · market · frame time (p99) (ms) 22.6 22.8 (+1%)
market-field-pressure · RX_VK_POOL_FIRST=1 (PR only) · return · frame time (p99) (ms) 24.8 26.3 (+6%)

Flags and settings

Flag What it is Values tried Before this PR After this PR Effect (frame rate) Helps / doesn't help / untested
RX_VK_POOL_FIRST switch the PR adds (set on the PR side only) 0, 1 not in main the PR's default market-field-pressure, main without it → the PR with 0: 83.6 → 82.1 (−2%) no clear effect
Markdown for the PR description
## Measurements

**Verdict:** fail on amd (1/1 report): `market-field-pressure screenshot.max_diff` 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: market-field-pressure, from a new save (frame_rate_cap=6, RX_VK_COMPACT_PCT=50); 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); market-field-pressure: 1 game start a side, market 30 s after 5 s warm-up, field 30 s after 5 s warm-up, return 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 @ d76feb2 | After: this PR @ a416edb |
|---|---|---|
| market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · field · frame rate | 83.6 | 82.1 (−2%) |
| market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · market · frame rate | 60.5 | 63.3 (+5%) |
| market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · return · frame rate | 54.9 | 55.9 (+2%) |
| market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · field · peak VRAM (MB) | 7,787 | 8,237 (+6%) |
| market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · market · peak VRAM (MB) | 8,274 | 8,149 (−2%) |
| market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · return · peak VRAM (MB) | 7,218 | 7,284 (+1%) |
| market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · longest load (s) | 17.6 | 10.1 (−43%) |
| market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · field · frame time (p99) (ms) | 18.9 | 18.4 (−3%) |
| market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · market · frame time (p99) (ms) | 24.2 | 22.9 (−5%) |
| market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · return · frame time (p99) (ms) | 27.1 | 25.3 (−6%) |
| market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · field · frame rate | 85.9 | 86.7 (+1%) |
| market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · market · frame rate | 63.4 | 63.8 (+1%) |
| market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · return · frame rate | 55.7 | 55.0 (−1%) |
| market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · field · peak VRAM (MB) | 8,615 | 8,135 (−6%) |
| market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · market · peak VRAM (MB) | 7,763 | 8,754 (+13%) |
| market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · return · peak VRAM (MB) | 7,979 | 8,525 (+7%) |
| market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · longest load (s) | 9.9 | 10.3 (+5%) |
| market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · field · frame time (p99) (ms) | 17.2 | 17.5 (+2%) |
| market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · market · frame time (p99) (ms) | 22.6 | 22.8 (+1%) |
| market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · return · frame time (p99) (ms) | 24.8 | 26.3 (+6%) |

### Flags and settings

| Flag | What it is | Values tried | Before this PR | After this PR | Effect (frame rate) | Helps / doesn't help / untested |
|---|---|---|---|---|---|---|
| `RX_VK_POOL_FIRST` | switch the PR adds (set on the PR side only) | `0`, `1` | not in main | the PR's default | market-field-pressure, main without it → the PR with 0: 83.6 → 82.1 (−2%) | no clear effect |
<!-- beanstalk:sticky --> ### Beanstalk test farm Commit `a416edb` · tier **reviewed** · approved by HumanBean · [dashboard](https://beanstalk.magicbean.dev/prs/1104) **What will be tested** · AI plan (`gpt-6.1-terra`) The plan compares the Market and Hyrule Field on 6 GB and 8 GB cards, Linux, AMD, and Steam Deck, taking roughly 25–30 minutes per tester. It lowers the compaction threshold to exercise memory pressure and tests the new pool-first switch both on and off. It measures frame times, FPS, VRAM and load times and compares screenshots; crashes, substantial performance or memory regressions, and visual differences fail the plan. The available probes do not verify eviction ordering or count texture recreations. - **Where:** vram6gb (any PC with 6 GB VRAM, 1 report), vram8gb (any PC with 8 GB VRAM, 1 report), linux (1 report), amd (1 report), steamdeck (1 report). - **Matrix:** 2 cells, each on main and the PR: RX_VK_POOL_FIRST `0` or `1` (PR only). - **market-field-pressure:** from a new save (settings frame_rate_cap=6; switches `RX_VK_COMPACT_PCT=50`), then teleport to Market, move the stick, measure market for 30 s (after 5 s warm-up), take the picture market to compare with main, teleport to Hyrule Field, measure field for 30 s (after 5 s warm-up), take the picture field to compare with main, measure return for 20 s (after 5 s warm-up); 1 run on each side for each of 2 cells (2 game starts a run). - **Measured:** crashes, frame times, stutters and hitches, draws and frame rate, VRAM use, load times. - **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 peak VRAM is more than 15% above main; or longest load is more than 30% above main; or frame time (p99) is more than 20% above main. - **Time:** about 31 min of game runs per tester (4 game starts), plus the two builds. Started. Run 2: the farm's reports were cleared after its Windows and Linux fixes: started again. | Target | Reports | Crash | Frame p99 (ms) vs main | VRAM peak (MB) vs main | Result | | --- | --- | --- | --- | --- | --- | | vram6gb · 1: linux+amd+radv+igpu+handheld `AMD Custom GPU 0932 (RADV VANGOGH), 26.1.99` | 1/1 [1](https://beanstalk.magicbean.dev/reports/rp_d664cc8c3068bb6e) | none | — | — | error | | vram8gb · 1: windows+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce RTX 3070 Laptop GPU, 621.24.0` | 1/1 [1](https://beanstalk.magicbean.dev/reports/rp_4ebc20d05cb96d5e) | 1 PR run | — | — | fail | | linux · 1: linux+amd+radv+dgpu `AMD Radeon RX 7900 XTX (RADV NAVI31), 26.2.7` | 1/1 [1](https://beanstalk.magicbean.dev/reports/rp_ac9705a094958abd) | none | — | — | error | | amd · 1: windows+amd+amd-proprietary+dgpu `AMD Radeon RX 9070 XT, 2.0.416` | 1/1 [1](https://beanstalk.magicbean.dev/reports/rp_acc9360e4cd8c420) | none | 18.4 vs 18.9 (−3%) | 8,237 vs 7,787 (+6%) | fail | | steamdeck | 0/1 | — | — | — | queued | | Scenario · matrix cell | Target | Frame p99 (ms) vs main | Result | | --- | --- | --- | --- | | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` | vram6gb | — | pass | | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` | vram8gb | — | pass | | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` | linux | — | pass | | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` | amd | 18.4 vs 18.9 (−3%) | fail | | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` | steamdeck | — | — | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` | vram6gb | — | pass | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` | vram8gb | — | fail | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` | linux | — | pass | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` | amd | 17.5 vs 17.2 (+2%) | fail | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` | steamdeck | — | — | #### Measurements · vram6gb **Verdict:** could not be judged on vram6gb (1/1 report). Hardware: 1: linux+amd+radv+igpu+handheld `AMD Custom GPU 0932 (RADV VANGOGH), 26.1.99` `AMD Custom APU 0932, 16 GB RAM` Scene: market-field-pressure, from a new save (frame_rate_cap=6, RX_VK_COMPACT_PCT=50); 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); market-field-pressure: 1 game start a side, market 30 s after 5 s warm-up, field 30 s after 5 s warm-up, return 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 @ d76feb2 | After: this PR @ a416edb | |---|---|---| | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · longest load (s) | 21.2 | 6.8 (−68%) | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · longest load (s) | 29.0 | 28.9 (−0%) | ### Flags and settings | Flag | What it is | Values tried | Before this PR | After this PR | Effect (frame rate) | Helps / doesn't help / untested | |---|---|---|---|---|---|---| | `RX_VK_POOL_FIRST` | switch the PR adds (set on the PR side only) | `0`, `1` | not in main | the PR's default | market-field-pressure, main without it → the PR with 0: not measured | untested | <details><summary>Markdown for the PR description</summary> ```markdown ## Measurements **Verdict:** could not be judged on vram6gb (1/1 report). Hardware: 1: linux+amd+radv+igpu+handheld `AMD Custom GPU 0932 (RADV VANGOGH), 26.1.99` `AMD Custom APU 0932, 16 GB RAM` Scene: market-field-pressure, from a new save (frame_rate_cap=6, RX_VK_COMPACT_PCT=50); 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); market-field-pressure: 1 game start a side, market 30 s after 5 s warm-up, field 30 s after 5 s warm-up, return 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 @ d76feb2 | After: this PR @ a416edb | |---|---|---| | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · longest load (s) | 21.2 | 6.8 (−68%) | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · longest load (s) | 29.0 | 28.9 (−0%) | ### Flags and settings | Flag | What it is | Values tried | Before this PR | After this PR | Effect (frame rate) | Helps / doesn't help / untested | |---|---|---|---|---|---|---| | `RX_VK_POOL_FIRST` | switch the PR adds (set on the PR side only) | `0`, `1` | not in main | the PR's default | market-field-pressure, main without it → the PR with 0: not measured | untested | ``` </details> #### Measurements · vram8gb **Verdict:** fail on vram8gb (1/1 report): `market-field-pressure crash` failed. Hardware: 1: windows+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce RTX 3070 Laptop GPU, 621.24.0` `AMD Ryzen 9 5900HX with Radeon Graphics, 32 GB RAM` Scene: market-field-pressure, from a new save (frame_rate_cap=6, RX_VK_COMPACT_PCT=50); 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); market-field-pressure: 1 game start a side, market 30 s after 5 s warm-up, field 30 s after 5 s warm-up, return 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 @ d76feb2 | After: this PR @ a416edb | |---|---|---| | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · field · frame rate | 56.4 | — | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · market · frame rate | 47.7 | — | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · return · frame rate | 45.9 | — | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · field · peak VRAM (MB) | 6,070 | — | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · market · peak VRAM (MB) | 5,346 | — | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · return · peak VRAM (MB) | 6,252 | — | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · longest load (s) | 62.9 | — | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · field · frame time (p99) (ms) | 26.4 | — | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · market · frame time (p99) (ms) | 34.7 | — | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · return · frame time (p99) (ms) | 33.9 | — | ### Flags and settings | Flag | What it is | Values tried | Before this PR | After this PR | Effect (frame rate) | Helps / doesn't help / untested | |---|---|---|---|---|---|---| | `RX_VK_POOL_FIRST` | switch the PR adds (set on the PR side only) | `0`, `1` | not in main | the PR's default | market-field-pressure, main without it → the PR with 0: not measured | untested | <details><summary>Markdown for the PR description</summary> ```markdown ## Measurements **Verdict:** fail on vram8gb (1/1 report): `market-field-pressure crash` failed. Hardware: 1: windows+nvidia+nvidia-proprietary+dgpu `NVIDIA GeForce RTX 3070 Laptop GPU, 621.24.0` `AMD Ryzen 9 5900HX with Radeon Graphics, 32 GB RAM` Scene: market-field-pressure, from a new save (frame_rate_cap=6, RX_VK_COMPACT_PCT=50); 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); market-field-pressure: 1 game start a side, market 30 s after 5 s warm-up, field 30 s after 5 s warm-up, return 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 @ d76feb2 | After: this PR @ a416edb | |---|---|---| | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · field · frame rate | 56.4 | — | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · market · frame rate | 47.7 | — | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · return · frame rate | 45.9 | — | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · field · peak VRAM (MB) | 6,070 | — | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · market · peak VRAM (MB) | 5,346 | — | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · return · peak VRAM (MB) | 6,252 | — | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · longest load (s) | 62.9 | — | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · field · frame time (p99) (ms) | 26.4 | — | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · market · frame time (p99) (ms) | 34.7 | — | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · return · frame time (p99) (ms) | 33.9 | — | ### Flags and settings | Flag | What it is | Values tried | Before this PR | After this PR | Effect (frame rate) | Helps / doesn't help / untested | |---|---|---|---|---|---|---| | `RX_VK_POOL_FIRST` | switch the PR adds (set on the PR side only) | `0`, `1` | not in main | the PR's default | market-field-pressure, main without it → the PR with 0: not measured | untested | ``` </details> #### Measurements · linux **Verdict:** could not be judged on linux (1/1 report). Hardware: 1: linux+amd+radv+dgpu `AMD Radeon RX 7900 XTX (RADV NAVI31), 26.2.7` `AMD Ryzen 7 5800X3D 8-Core Processor, 64 GB RAM` Scene: market-field-pressure, from a new save (frame_rate_cap=6, RX_VK_COMPACT_PCT=50); 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); market-field-pressure: 1 game start a side, market 30 s after 5 s warm-up, field 30 s after 5 s warm-up, return 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 @ d76feb2 | After: this PR @ a416edb | |---|---|---| | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · longest load (s) | 39.9 | 30.7 (−23%) | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · longest load (s) | 37.6 | 26.9 (−28%) | ### Flags and settings | Flag | What it is | Values tried | Before this PR | After this PR | Effect (frame rate) | Helps / doesn't help / untested | |---|---|---|---|---|---|---| | `RX_VK_POOL_FIRST` | switch the PR adds (set on the PR side only) | `0`, `1` | not in main | the PR's default | market-field-pressure, main without it → the PR with 0: not measured | untested | <details><summary>Markdown for the PR description</summary> ```markdown ## Measurements **Verdict:** could not be judged on linux (1/1 report). Hardware: 1: linux+amd+radv+dgpu `AMD Radeon RX 7900 XTX (RADV NAVI31), 26.2.7` `AMD Ryzen 7 5800X3D 8-Core Processor, 64 GB RAM` Scene: market-field-pressure, from a new save (frame_rate_cap=6, RX_VK_COMPACT_PCT=50); 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); market-field-pressure: 1 game start a side, market 30 s after 5 s warm-up, field 30 s after 5 s warm-up, return 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 @ d76feb2 | After: this PR @ a416edb | |---|---|---| | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · longest load (s) | 39.9 | 30.7 (−23%) | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · longest load (s) | 37.6 | 26.9 (−28%) | ### Flags and settings | Flag | What it is | Values tried | Before this PR | After this PR | Effect (frame rate) | Helps / doesn't help / untested | |---|---|---|---|---|---|---| | `RX_VK_POOL_FIRST` | switch the PR adds (set on the PR side only) | `0`, `1` | not in main | the PR's default | market-field-pressure, main without it → the PR with 0: not measured | untested | ``` </details> #### Measurements · amd **Verdict:** fail on amd (1/1 report): `market-field-pressure screenshot.max_diff` 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: market-field-pressure, from a new save (frame_rate_cap=6, RX_VK_COMPACT_PCT=50); 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); market-field-pressure: 1 game start a side, market 30 s after 5 s warm-up, field 30 s after 5 s warm-up, return 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 @ d76feb2 | After: this PR @ a416edb | |---|---|---| | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · field · frame rate | 83.6 | 82.1 (−2%) | | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · market · frame rate | 60.5 | 63.3 (+5%) | | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · return · frame rate | 54.9 | 55.9 (+2%) | | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · field · peak VRAM (MB) | 7,787 | 8,237 (+6%) | | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · market · peak VRAM (MB) | 8,274 | 8,149 (−2%) | | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · return · peak VRAM (MB) | 7,218 | 7,284 (+1%) | | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · longest load (s) | 17.6 | 10.1 (−43%) | | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · field · frame time (p99) (ms) | 18.9 | 18.4 (−3%) | | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · market · frame time (p99) (ms) | 24.2 | 22.9 (−5%) | | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · return · frame time (p99) (ms) | 27.1 | 25.3 (−6%) | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · field · frame rate | 85.9 | 86.7 (+1%) | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · market · frame rate | 63.4 | 63.8 (+1%) | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · return · frame rate | 55.7 | 55.0 (−1%) | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · field · peak VRAM (MB) | 8,615 | 8,135 (−6%) | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · market · peak VRAM (MB) | 7,763 | 8,754 (+13%) | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · return · peak VRAM (MB) | 7,979 | 8,525 (+7%) | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · longest load (s) | 9.9 | 10.3 (+5%) | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · field · frame time (p99) (ms) | 17.2 | 17.5 (+2%) | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · market · frame time (p99) (ms) | 22.6 | 22.8 (+1%) | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · return · frame time (p99) (ms) | 24.8 | 26.3 (+6%) | ### Flags and settings | Flag | What it is | Values tried | Before this PR | After this PR | Effect (frame rate) | Helps / doesn't help / untested | |---|---|---|---|---|---|---| | `RX_VK_POOL_FIRST` | switch the PR adds (set on the PR side only) | `0`, `1` | not in main | the PR's default | market-field-pressure, main without it → the PR with 0: 83.6 → 82.1 (−2%) | no clear effect | <details><summary>Markdown for the PR description</summary> ```markdown ## Measurements **Verdict:** fail on amd (1/1 report): `market-field-pressure screenshot.max_diff` 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: market-field-pressure, from a new save (frame_rate_cap=6, RX_VK_COMPACT_PCT=50); 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); market-field-pressure: 1 game start a side, market 30 s after 5 s warm-up, field 30 s after 5 s warm-up, return 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 @ d76feb2 | After: this PR @ a416edb | |---|---|---| | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · field · frame rate | 83.6 | 82.1 (−2%) | | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · market · frame rate | 60.5 | 63.3 (+5%) | | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · return · frame rate | 54.9 | 55.9 (+2%) | | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · field · peak VRAM (MB) | 7,787 | 8,237 (+6%) | | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · market · peak VRAM (MB) | 8,274 | 8,149 (−2%) | | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · return · peak VRAM (MB) | 7,218 | 7,284 (+1%) | | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · longest load (s) | 17.6 | 10.1 (−43%) | | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · field · frame time (p99) (ms) | 18.9 | 18.4 (−3%) | | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · market · frame time (p99) (ms) | 24.2 | 22.9 (−5%) | | market-field-pressure · `RX_VK_POOL_FIRST=0 (PR only)` · return · frame time (p99) (ms) | 27.1 | 25.3 (−6%) | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · field · frame rate | 85.9 | 86.7 (+1%) | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · market · frame rate | 63.4 | 63.8 (+1%) | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · return · frame rate | 55.7 | 55.0 (−1%) | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · field · peak VRAM (MB) | 8,615 | 8,135 (−6%) | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · market · peak VRAM (MB) | 7,763 | 8,754 (+13%) | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · return · peak VRAM (MB) | 7,979 | 8,525 (+7%) | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · longest load (s) | 9.9 | 10.3 (+5%) | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · field · frame time (p99) (ms) | 17.2 | 17.5 (+2%) | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · market · frame time (p99) (ms) | 22.6 | 22.8 (+1%) | | market-field-pressure · `RX_VK_POOL_FIRST=1 (PR only)` · return · frame time (p99) (ms) | 24.8 | 26.3 (+6%) | ### Flags and settings | Flag | What it is | Values tried | Before this PR | After this PR | Effect (frame rate) | Helps / doesn't help / untested | |---|---|---|---|---|---|---| | `RX_VK_POOL_FIRST` | switch the PR adds (set on the PR side only) | `0`, `1` | not in main | the PR's default | market-field-pressure, main without it → the PR with 0: 83.6 → 82.1 (−2%) | no clear effect | ``` </details>
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 / Conventional Commit title (pull_request) Successful in 16s
Website / Website on Cloudflare (pull_request_target) Has been skipped
Guide screenshots / Guide images and local UI source review (pull_request) Failing after 26s
CI / Workflow lint (actionlint) (pull_request) Successful in 27s
Third-party parts / As their notes say (pull_request) Successful in 37s
PR description / Description follows the template (pull_request_target) Successful in 9s
beanstalk/approval approved by HumanBean (reviewed)
beanstalk/vram6gb 0/1 pass, 1 error
beanstalk/vram8gb 1/1 fail
beanstalk/linux 0/1 pass, 1 error
beanstalk/amd 1/1 fail
beanstalk/steamdeck 0/1 pass, 0 running, 1 queued
This pull request is marked as a work in progress.
This branch is out-of-date with the base branch
View command line instructions

Checkout

From your project repository, check out a new branch and test the changes.
git fetch -u 'perf/pool-first-eviction:GummiNaomi-perf/pool-first-eviction'
git switch 'GummiNaomi-perf/pool-first-eviction'
Sign in to join this conversation.
No reviewers
No milestone
No project
No assignees
4 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!1104
No description provided.