Walnut Games

All games

Neverball

Gold · 87 0.1 h played, 0 crashes

Tilt the floor to roll a ball through 3D levels, collecting coins before the time runs out.

Genre
puzzle arcade
Theme
family
Players
single player
Input
supports gamepad
View
3D
Keywords
physics ball levels

Played sessions

Testing is done with Walnut's rating tool (walnut-rate, on the console): how ratings are made and submitted.

Builds: Walnut 0.1, neverball 1.6.0+git20180603-3+ps3.721

The platform builds under it (3)
Linux kernel6.12.111+ps3.747 Walnut's build
Mesa26.2.4-1+ps3.721 Walnut's build
ps3rsx0.735+k6.12.111+ps3.747 Walnut's build
DateLauncherRulesPlayEndedPicture
2026-10-24 Neverball 1.04 minquit ok Details

Download all sessions as CSV (one row per session, the aspects as columns)

Rating history

The rating on each version it was tested on, newest first: each row is rated from that version's sessions alone, under the newest rules those sessions have the data for.

VersionTestedMedalRules Frame rateSmoothnessSoundPictureStabilityDisplayLoad Play
Walnut 0.1
neverball 1.6.0+git20180603-3		 2026-10-24 Gold · 87 1.0 88649710090-100 0.1 h
1 test

Scripted benchmarks

Not rated: tests of driver and game changes. A script plays the same scene on the PS3 again and again (the desktop stopped, vsync off) to compare two builds of the game or of the graphics driver. The numbers say which build is faster; they never feed the medal.

How the script plays it: Neverball: its shipped replay gui/demo1.nbr, real-time paced.

The recipe's full description

Neverball (1.6 snapshot of 2018): its shipped replay gui/demo1.nbr, real-time paced. Our package (gamedb/games/neverball/packages/neverball/patches) prints one line with NEVERBALL_BENCH=1 when the replay ends, from the first replay frame (frames, seconds, fps, longest frame) and quits, so the run is exactly the replay; DURATION is a safety cap. Config ~/.neverball/ neverballrc: fullscreen 1920x1080, vsync off, the fps counter on. It generates mipmaps on the GPU (GL_GENERATE_MIPMAP) and uses GL_LUMINANCE_ALPHA textures: both broken on nv30 until the Mesa fixes (the game's technical.md); variant nomip turns mipmaps off. Steady state: passes. The replay is ~20 s; our patch 0002 (st_demo.c bench_end) with NEVERBALL_BENCH_LOOPS=N replays it N times (demo_replay_stop, progress_replay on a copy of curr_demo(), goto_state(&st_demo_play)), one line per pass, and quits after the Nth. 12 passes cover 60 s of warm-up + 120 s; the client stops the run once enough is measured. fps= is the first pass's.

The recipe today runs it at the game's own resolution, with NEVERBALL_BENCH=1 NEVERBALL_BENCH_LOOPS=20. Older batches ran the recipe as it was then: their titles and observations say what changed (a new game build, a driver change).

7 variants of the recipe
base stands for the gamethe recipe as it is, without a variant
nomip-
gloffour neverball (gamedb/games/neverball/packages/neverball/patches/0003): fewer GL calls and draws a frame, each change behind an environment switch, base has them all on. gui/demo1.nbr on llvmpipe at GL 2.1: 2154 -> 844 calls, 137 -> 74 draws a frame (levels at their intro view: greed 6289 -> 1915, mover 8205 -> 2463 calls). gloff is NEVERBALL_GL=0 (the game as Debian draws it), a no* variant is base with one change off.
nomesh-
noitems-
noback-
nostate-

One table, a header row per batch (what it tested and what came out). Conditions: what a variant adds to the recipe (its environment switches) and what ran alongside (the stall recorder, a profiler); a column appears only when this game has a value in it. Rows marked invalid or not comparable say why.

What the run length means

"warm-up + 120 s" is a steady-state run, which leaves out a 60 s warm-up and measures 120 s or more; "whole run" counted everything from the start, the way runs before 2026-10-17 did.

Variant Conditions Steady fps
in the window		 1% low
fps		 Stalls
frames over 100 ms		 fps
whole run		 Build Run length ram free
MiB		vram
MiB		stall play s
seconds		longest play ms
ms		sound faults per min
faults a minute		cpu
seconds		underruns
underruns		longest ms
ms		 Note
Suite summary 2026-10-20 · suite-summary-2026-10
Tested: Not a test run: the suite's 13 benchmarked games (15 rows, with OpenArena's two renderers and OpenTTD's two blitters) gathered with their known numbers as of October 2026, each value naming the batch or document it came from.
Result: It is the calibration of the rating formula (rules 1): the scale was tried on these numbers before any game was rated, and the GameDB calibration table is generated from it. Benchmarks never give a medal; medals come from played sessions only.
Neverball the recipe as it is 66.5 41.3 - - - warm-up + 120 s 58 34 2.0 3900 50 - - - right, 118 underruns
Real-time audio as shipped: ps3-desktop 0.1013 (limits.d rtprio 10 + ps3-rt-audio, RTAUDIO=1) against none (RTPRIO=0) on dsda-doom software (owncfg), Extreme Tux Racer, X-Moto and Neverball (kernel .874) 2026-10-28 · 2026-10-28-rt-audio-shipped
Tested: Real-time audio as shipped in ps3-desktop 0.1013 (a priority limit plus ps3-rt-audio, which game mode uses to put the game's audio thread on SCHED_RR), against none, on dsda-doom, Extreme Tux Racer, X-Moto and Neverball.
Result: Shipped. Audible dropouts a run fell from 29 to 2 in dsda-doom, 26 to 2 in Extreme Tux Racer and 3 to 0 in Neverball, with steady fps unchanged within 0.4%. The 1% lows came out a little lower in the light games, with the ranges touching.
What we saw

25 runs in three rounds, games interleaved, arm order alternating. Two Neverball runs with it were interfered by the agent's own SSH checks and left out, so Neverball has 2 runs with it. Every run fullscreen at 1920x1080 and in gameplay at its 40 s screenshot.

  • ps3-rt-audio put the audio thread on SCHED_RR 1 in every run with it: SDL's audio thread in dsda-doom, X-Moto and Neverball, and Extreme Tux Racer's unnamed audio thread.
  • Audible dropouts a run (about 200 s), median, without and with it: dsda-doom 29 to 2 (audible time 0.8-1.2 s to 50-113 ms), Extreme Tux Racer 26 to 2 (0.65-0.92 s to 58-124 ms), Neverball 3 to 0, X-Moto 0 to 0 (logged underruns 78 to 11).
  • Steady fps unchanged within 0.4%: dsda-doom 39.75 and 39.8, X-Moto 161.7 and 161.0, Neverball 82.73 and 82.72.
  • 1% lows a little lower with it in the light games, ranges touching: X-Moto 113.4 to 109.4, Neverball 57.9 to 56.6, dsda-doom 20.6 to 19.0.
  • Extreme Tux Racer's were higher with it: HUD 1% low 19.8 to 29.2, steady 56 to 59. The previous batch had no such gap (29.9 against 28.6).
base rtaudio 82.7 57.3 0.0 82.5 - warm-up + 120 s 56.8 34.2 0.0 28.6 - 99.5 53.0 193.4
base real-time priority 0 82.7 58.4 0 82.6 - warm-up + 120 s 54.2 34.2 0.0 28.3 - 94.8 106 214.4
Luanti ps3.364 batching, Neverball ps3.370 batching, Sauerbraten libspeq 0.381 skinning (spawn) 2026-10-17 · 2026-10-17-luanti-neverball-batching
Tested: Draw batching in our Luanti (ps3.364) and Neverball (ps3.370) packages against the old drawing, and BananaBread's model skinning on the SPUs, on the PPE through libspeq, or off (libspeq 0.381).
Result: Batching gives Luanti fewer stalls and shorter worst frames, and cuts Neverball's CPU from 80 to 65 s at the same frame rate. For BananaBread libspeq's PPE skinning is best; the SPU path adds hitches. One Luanti batching-off run was later found disturbed by input.
What we saw
  • Same boot.
  • Luanti, batching on against off, two runs each: steady 45.0 against 46.0 fps, stall time 6.25 against 9.1 s, longest frame 1.7 against 3.4 s, no underruns, picture right.
  • Neverball, batching on against old drawing: smooth 78.7 against 79.1 fps, CPU 65 against 80 s, stalls similar. One batched run ended before the measured window (12 demo loops were too few; now 20).
  • BananaBread (spawn point with the gun), one run each: skinning on the SPUs 23.6 fps steady with 21 stalls (4.3 s); on the PPE 28.1 fps with no stalls; off 23.0 with no stalls. libspeq's PPE path is the best; the SPU path adds hitches.
  • Interference (found by the later scan): one Luanti batching-off run had three digs (a mouse button in the game) in its warm-up. The measured window had no logged action, but the batching-off cell includes that run.
base the recipe as it is 66.5 41.3 51 76.5 - warm-up + 120 s 58.5 34.2 - - - 65.5 118 3893.0
gloff NEVERBALL_GL=0 62.4 48.1 39.5 71.9 - warm-up + 120 s 56.9 34.3 - - - 80.3 115.0 3674.5
Mesa ps3.262 repeats and Armagetron without 0020 (texinv) 2026-10-17 · 2026-10-17-mesa-262-repeats
Tested: Whether the frame rates of the Mesa ps3.262 suite round hold up when repeated, and whether Armagetron's drop on ps3.262 comes from Mesa patch 0020 (the texture cache flushed only after texture changes).
Result: Single QuakeSpasm runs spread from 66 to 84 fps on the same Mesa, so one-run differences of that size are noise. Armagetron runs at 52-53 fps with or without 0020, so 0020 is not the cause of its loss.
What we saw
  • Same boot as the Mesa ps3.262 suite round.
  • QuakeSpasm again: 73.0 and 84.1 fps (66.3 in the suite round). Single QuakeSpasm runs spread widely on the same Mesa.
  • Neverball again: 77.7 and 74.2 fps (71.5 in the suite round).
  • Armagetron with ps3.262 as built: 53 and 53 fps. With patch 0020 undone (the texture cache flushed before every draw again): 52 and 52. So the drop from 60 to 52 is not patch 0020.
  • All Armagetron numbers here are the Gallium HUD's median. Two runs per variant.
base the recipe as it is - - - 75.9 Mesa ps3.262 whole run 62.0 34.3 - - - 11.5 19.0 -
Mesa ps3.262 + ps3rsx 0.249 2026-10-17 · 2026-10-17-mesa-262
Tested: Mesa ps3.262 (patches 0019 to 0025: fewer words sent to the GPU per draw) across the suite, against ps3.246.
Result: Big gains in several games: OpenArena +23%, Yamagi +24%, X-Moto +21%, Extreme Tux Racer and dsda-doom +15%. Armagetron fell from 60 to 51 fps, which the next batches traced. Single runs; BananaBread's drop is a memory stall, not a driver number.
What we saw
  • The patches: no relocations, the texture cache invalidated only after texture changes, no duplicate vertex-format writes, batched vertex program constants. One run per game.
  • Against ps3.246: OpenArena with compressed textures 44.2 to 54.4 fps (+23%), Yamagi 51.4 to 63.9 (+24%), X-Moto 39 to 47 (+21%), Extreme Tux Racer 46 to 53 (+15%), dsda-doom software 33 to 38 with the demo now finishing (+15%), ClassiCube 47.6 to 50.1 (+5%).
  • Neverball 78.2 to 71.5 (noise), QuakeSpasm 70.5 to 66.3 (repeated in the next batch), Luanti 53 to 53.
  • Armagetron dropped from 60 to 51 (driver frame counter).
  • BananaBread reads 28 to 8 on the counter, but it was memory-stalled (13 MB free, killed at the cap): not a driver number, and the cause is uncertain.
  • The kernel's command validator rejected nothing.
  • Single runs.
base the recipe as it is - - - 71.5 Mesa ps3.262, ps3rsx 0.249 whole run 64.6 34.3 - - - 10.7 12 -
Mesa ps3.246 + ps3rsx 0.249 2026-10-17 · 2026-10-17-mesa-246-ps3rsx-249
Tested: The first full suite round after the reboot into ps3rsx 0.249, one run per game, as a baseline.
Result: Every game ran with the right picture. The sound was right except in dsda-doom (MIDI music), BananaBread and OpenTTD. Single runs, so differences of a few fps are noise.
What we saw
  • One run per game. Frame rates by the game's own counter, or the driver's frame counter where marked:
  • Armagetron 60 (counter), ClassiCube 47.6, dsda-doom software 33 (counter, demo unfinished), Extreme Tux Racer 46 (counter), Luanti 53 (counter), Neverball 78.2.
  • OpenArena with compressed textures 44.2 (longest frame 2.1 s), QuakeSpasm 70.5, BananaBread with CPU skinning 28 (counter), X-Moto 39 (counter), Yamagi Quake II 51.4.
  • Pictures right. Sound right except dsda-doom (1161 underruns, MIDI music), BananaBread (176) and OpenTTD (1429, choppy).
  • OpenTTD used 55.5 s of CPU and has no frame rate (plain X11 path).
  • Single runs: differences of a few fps are within the run-to-run spread.
base the recipe as it is - - - 78.2 Mesa ps3.246, ps3rsx 0.249 whole run 71.8 34.3 - - - 12.7 16 -
ps3rsx 0.249 first boot (Mesa ps3.246; SDL3 ps3.254) 2026-10-17 · 2026-10-17-ps3rsx-249-first-boot
Tested: The first boot with ps3rsx 0.249 and libsdl3 ps3.254: Neverball without the wedge, OpenTTD's native upload, and the first runs of Luanti and BananaBread (the Cube 2 engine).
Result: Neverball runs without wedging (76.4 and 70.2 fps), and OpenTTD rose from about 2 to 8 fps (12 with the 8-bit blitter). Luanti and BananaBread run but are memory-bound; BananaBread's first run was killed for lack of memory.
What we saw
  • The GL tests pass and switching virtual terminals works with a GL program running (those runs aren't in the records).
  • Neverball (our +ps3.257 with its own benchmark, about 23 s) runs without the wedge: 76.4 and 70.2 fps by its own count, longest frames 320 and 378 ms, 15 and 36 underruns.
  • OpenTTD with libsdl3 ps3.254 (native BGRA upload): from about 2 to 8 fps on the driver's frame counter (9.5 across the two runs), 12 fps with the 8-bit blitter. The game's own software drawing is the limit.
  • Luanti, first run: load 37 s, 67 fps on the counter (10th percentile 23, dropping to 0 while map chunks load), free memory down to 31 MB.
  • BananaBread (CPU skinning), first runs: the first was killed by the out-of-memory killer (free memory down to 10 MB); the second reached 15 fps on the counter, 236 underruns.
  • Smoke runs, single or paired.
base the recipe as it is - - - 73.3 Mesa ps3.246, ps3rsx 0.249 whole run 64.3 34.3 - - - 12.2 25.5 -
Mesa ps3.225 (0014 mip readback): tests and Neverball 2026-10-16 · 2026-10-16-mesa-225
Tested: Mesa ps3.225 with patch 0014, which reads the two smallest mipmap levels on the CPU, then Neverball, which had wedged the GPU on ps3.223.
Result: Fixed: every texture and mipmap test passes and Neverball ran 60 s without the wedge, at 70 fps. Mesa 0014 is proven; ps3rsx 0.249's own refusal of misaligned textures was still untested.
What we saw
  • The texture-format, mipmap and mipmap readback tests all pass (RGBA8 and luminance-alpha at 32x32, and a 70-case set).
  • The 2x2 and 1x1 levels are now read by the CPU.
  • Neverball ran 60 s without the wedge: 70 fps on the driver's frame counter (10th percentile 55), 50 underruns.
  • Proves Mesa 0014. ps3rsx 0.249's refusal of unaligned texture offsets was still untested.
base the recipe as it is - - - 70 Mesa ps3.225 whole run 75.6 35.4 - - - 31.8 50 -
Mesa ps3.223 Neverball wedges the RSX (graph exception 0x105) 2026-10-16 · 2026-10-16-mesa-223-neverball-wedge
Tested: Neverball's first run, on Mesa ps3.223.
Result: It locked up the GPU, which only a console reboot clears. The cause was a texture read from a misaligned address during software mipmap generation; fixed by Mesa 0014 (ps3.225), and ps3rsx 0.249 refuses such addresses too.
What we saw
  • The kernel rejected a command buffer (an input/output error) and logged a graphics exception from Neverball.
  • The RSX was wedged: only a console reboot clears it. The run was killed at the 60 s cap.
  • Cause, found next: Mesa's 3D blit read the 2x2 level of a 32x32 luminance-alpha texture at an address not aligned to 128 bytes, while generating mipmaps in software for a format nv30 cannot render to.
  • Fixed by Mesa 0014 (ps3.225); ps3rsx 0.249 also refuses such offsets.
base the recipe as it is - - - - Mesa ps3.223 whole run 71.7 25.0 - - - 63.7 5 -