SuperTuxKart: test runs

Played sessions

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

Same in every session: Walnut 0.1

Date	Launcher	Game build	Platform builds that differ	Rules	Play	Ended	Picture
2026-10-26	SuperTuxKart	supertuxkart 1.5-2+ps3.833	Linux kernel 6.12.111+ps3.786 Mesa 26.2.4-1+ps3.850 ps3rsx 0.788+k6.12.111+ps3.786	2.3	13 min	quit	ok	Details
2026-10-26	SuperTuxKart	supertuxkart 1.5-2+ps3.833	Linux kernel 6.12.111+ps3.747 Mesa 26.2.4-1+ps3.827 ps3rsx 0.788+k6.12.111+ps3.786	2.3	12 min	crash	minor	Details
2026-10-25	SuperTuxKart	supertuxkart 1.5-2+ps3.786	Linux kernel 6.12.111+ps3.747 Mesa 26.2.4-1+ps3.827 ps3rsx 0.735+k6.12.111+ps3.747	1.0	2 min	quit	wrong	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.

Version	Tested	Medal	Rules	Frame rate	Smoothness	Sound	Picture	Stability	Controls	Display	Features	Load	Play
Walnut 0.1 supertuxkart 1.5-2	2026-10-26	Silver · 69	2.3	5	58	98	95	91	100	100	100	71	0.2 h 2 tests
Walnut 0.1 supertuxkart 1.5-2	2026-10-25	Scrap · 31	1.0	-	-	73	-	90	-	-	-	97	0.0 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: SuperTuxKart 1.5. Its profile mode starts a race at once with every kart driven by the AI, drives for the given seconds, prints "Number of frames: N time T, Average FPS: F" and quits.

The recipe's full description

SuperTuxKart 1.5 (our build, gamedb/games/supertuxkart/packages/supertuxkart/: big-endian .spm loading, legacy-renderer fixes; 839 MB with its data, deployed by name). Its profile mode (--profile-time) starts a race at once with every kart driven by the AI, drives for the given seconds, prints "Number of frames: N time T, Average FPS: F" and quits. GL 2.1 picks the legacy renderer by itself ("Using the fixed pipeline"). RAM and VRAM are the risk (its FAQ wants 1 GB against our 217 MB): a small track and 4 karts. setup() writes the whole config (~/.config/supertuxkart/config-0.10/config.xml) every run, so the run doesn't depend on what an earlier one saved: the settings below, everything else at the game's defaults (a file without a key keeps that key's default). Names from src/config/ user_config.hpp; the file has one element per group with the settings as attributes. The game rewrites it when it quits (BACKUP restores it). base: the game's defaults (HD textures on: up to 2048x2048; texture compression on, which our 0004 makes reach the legacy renderer: opaque power-of-two textures as DXT1, compressed by Mesa at upload; the minimap shown, 512x512 render target with 0003), but vsync off and the old-driver popup off. Every variant changes one thing, except lean (the candidate PS3 default). Memory per run: during.log has MemAvailable and SwapFree every 5 s, bench.txt their lowest. Steady state: the HUD window, 60 s from the end of the track load (LOAD_MARK, the Track: line printed when the scene is built; the race starts at once, no countdown), then 120 s. Our patch 0005 (ProfileWorld::update) with STK_PROFILE_WARMUP=60 restarts the tick count and the clock once the race time passes 60 s and ends the race at 60 + the profile time (180 s: 240 s of race, 180 s measured); the "Average FPS" line it prints is ticks per second (see FPS_RE below). The race is AI-driven (rand() seeded from the clock, frame-time physics): close between runs, not identical.

The recipe today runs it at the game's own resolution, with STK_PROFILE_WARMUP=60. Older batches ran the recipe as it was then: their titles and observations say what changed (a new game build, a driver change).

11 variants of the recipe

`base`	the recipe as it is, without a variant
`nohd`	HD textures off (bit 1: the user's choice, so the game doesn't turn them back on): textures are scaled down to max_texture_size (512) at load instead of 2048. The biggest single VRAM and RAM cut expected: a 2048x2048 RGBA8 texture is 16 MB, at 512 1 MB (DXT1: 128 KB)
`tex256`	the same capped at 256
`notc`	no texture compression: every texture RGBA8, as Debian's build (isolates 0004: VRAM, load time)
`nomap`	minimap hidden: its render target (FBO texture plus depth buffer) isn't made (0003)
`noparticles`	no particles (skid, nitro, weather, explosions): fewer billboards and their textures
`static`	drivers not animated: the legacy renderer skins every animated kart on the CPU each frame
`geo0`	LOD switched at the shortest distances (Very Low): draw cost, not memory (every level is loaded)
`vsync`	vsync on, the game's default (base turns it off)
`karts1`	fewer karts on the track (their models load at startup anyway; this is draw and AI cost)
`lean` stands for the game	the candidate PS3 default if the single steps hold: HD textures off (512), compression on, no minimap, important particles only, static drivers

Conditions: what a variant adds to the recipe (its environment switches) and what ran alongside (the stall recorder, a profiler). Rows marked invalid or not comparable say why. The other numbers, the build and the run length are under "All numbers" in each batch.

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.

Suite summary

2026-10-20 · batch suite-summary-2026-10

Purpose: The 13 benchmarked games (15 rows) with their known numbers, the rules-1 calibration of the rating formula (docs/GAMEDB.md "Calibration", tools/gamedb/score.py). Each value names its source: a batch's cell, or the suite table of docs/games/README.md where the number is a reading or a range (no single cell holds it). Stutter is the stall seconds over the 2-minute window; load unmeasured, stability assumed clean, picture and display right.

Row	ram_free MiB	vram MiB	fps_steady frames a second	stall_play_s seconds	low1_play frames a second	longest_play_ms ms	sound_faults_per_min faults a minute	Picture, sound
SuperTuxKart lean lean	26	97	21.0	1.8	11.8	513	3	track draws, sound running

batch ssdroot12 snd_ps3 DMA watchdog (kernel local/0047): dma_watchdog_ms=200 against 0 (off) on SuperTuxKart lean, OpenArena gl2min and ClassiCube base x3 (SSD root .472 Mesa ps3.545~dev, .472 with the test module)

2026-10-20 · batch 2026-10-20-ssdroot12-dma-watchdog · 17 runs in the batch (all games)

Variant	Conditions	Steady fps in the window	1% low fps	Stalls frames over 100 ms
`lean`	stall recorder at 10 ms	21	11.8	6
`lean`	stall recorder at 10 ms switch snd_dma_watchdog_ms0	20	11.7	5

All numbers and setup of this batch

Variant	Numbers	Setup
`lean`	fps whole run 20 cpu 318.4 seconds underruns 0 underruns ram_free 26.1 MiB vram 95.0 MiB longest_ms 291.1 ms stall_disk_s 0.1 seconds stall_swap_s 0 seconds stall_reclaim_s 0.1 seconds stall_gpu_s 0 seconds stall_cpu_s 0.8 seconds stall_play_s 1.0 seconds longest_play_ms 291.1 ms	Kernel ps3.472, Mesa ps3.545~dev the game's resolution, warm-up + 120 s
`lean`	fps whole run 19 cpu 323.5 seconds underruns 0 underruns ram_free 29.2 MiB vram 96.0 MiB longest_ms 126.1 ms stall_disk_s 0 seconds stall_swap_s 0 seconds stall_reclaim_s 0 seconds stall_gpu_s 0 seconds stall_cpu_s 0.6 seconds stall_play_s 0.6 seconds longest_play_ms 126.1 ms	Kernel ps3.472, Mesa ps3.545~dev the game's resolution, warm-up + 120 s

Observations: same boot (2026-10-20 06:53-08:11Z), 18 runs rotated, DUMPON=1, arms base (watchdog 200 ms) and snd_dma_watchdog_ms0 (off); one OpenArena off run (07:45Z, two logins) interfered and out of the cells. Wedges (Wait timeout lines): off arm 2 of 9 runs (OpenArena snd_stall=95 oal_wait=65, ClassiCube snd_stall=164 oal_wait=155); watchdog on 0 of 9, with 2 restarts caught (OpenArena, ClassiCube) and the sound kept going. fps, 1% low and stalls equal: SuperTuxKart 21 against 20 fps, ClassiCube 48.4 both, OpenArena 61.9 against 62.3

batch ssdroot10 OpenAL wedge rate on SuperTuxKart lean x5 per arm: our openal-soft +ps3.549 (int16), stock +ps3.273 (LDPATH), ours with OAL=f32 (SSD root .472 Mesa ps3.545~dev)

2026-10-20 · batch 2026-10-20-ssdroot10-openal-wedge · 14 runs in the batch (all games)

Variant	Conditions	Steady fps in the window	1% low fps	Stalls frames over 100 ms
`lean`	stall recorder at 10 ms	20	11.6	8
`lean`	stall recorder at 10 ms switch ldstock-openal	21.0	11.9	7.5
`lean`	stall recorder at 10 ms switch oalf32	20	11.5	9

All numbers and setup of this batch

Variant	Numbers	Setup
`lean`	fps whole run 19 cpu 322.7 seconds underruns 0 underruns ram_free 23.4 MiB vram 95.7 MiB longest_ms 255.3 ms stall_disk_s 0.1 seconds stall_swap_s 0.0 seconds stall_reclaim_s 0.0 seconds stall_gpu_s 0 seconds stall_cpu_s 1.0 seconds stall_play_s 1.3 seconds longest_play_ms 255.3 ms	Kernel ps3.472, Mesa ps3.545~dev the game's resolution, warm-up + 120 s
`lean`	fps whole run 20.0 cpu 315.7 seconds underruns 0.0 underruns ram_free 25.9 MiB vram 95.0 MiB longest_ms 216.0 ms stall_disk_s 0.1 seconds stall_swap_s 0.0 seconds stall_reclaim_s 0.0 seconds stall_gpu_s 0.0 seconds stall_cpu_s 0.9 seconds stall_play_s 1.1 seconds longest_play_ms 216.0 ms	Kernel ps3.472, Mesa ps3.545~dev the game's resolution, warm-up + 120 s
`lean`	fps whole run 20 cpu 313.2 seconds underruns 0 underruns ram_free 21.4 MiB vram 95.9 MiB longest_ms 157.8 ms stall_disk_s 0.1 seconds stall_swap_s 0 seconds stall_reclaim_s 0 seconds stall_gpu_s 0 seconds stall_cpu_s 0.8 seconds stall_play_s 1.1 seconds longest_play_ms 157.8 ms	Kernel ps3.472, Mesa ps3.545~dev the game's resolution, warm-up + 120 s

Observations: same boot (2026-10-20 03:50-05:12Z), 15 runs rotated, one (04:44Z stock, two logins) interfered and out of the cells. Runs with Wait timeout lines (the wedge): ours 2 of 5, stock 3 of 5, ours f32 0 of 5. Our build does not raise the rate (ssdroot5's 0 of 12 for stock was luck): the wedge is the snd_ps3 DMA stop whatever OpenAL does (kernel local/0047 watchdog). f32's 0 of 5 is not significant at n=5. fps equal: 20-21 steady, 1% low 11.5-11.9

batch ssdroot9 Mesa 0044: NV30_MINMAX_VMX=1 against unset on SuperTuxKart lean x5 (SSD root .472 Mesa ps3.545~dev)

2026-10-20 · batch 2026-10-20-ssdroot9-minmax-vmx · 10 runs in the batch (all games)

Variant	Conditions	Steady fps in the window	1% low fps	Stalls frames over 100 ms
`lean`	stall recorder at 10 ms	20	11.6	8
`lean`	stall recorder at 10 ms switch minmax_vmx1	22	12.3	5

All numbers and setup of this batch

Variant	Numbers	Setup
`lean`	fps whole run 20 cpu 319.6 seconds underruns 0 underruns ram_free 24.1 MiB vram 95.1 MiB longest_ms 205.6 ms stall_disk_s 0.1 seconds stall_swap_s 0 seconds stall_reclaim_s 0.0 seconds stall_gpu_s 0 seconds stall_cpu_s 1.0 seconds stall_play_s 1.2 seconds longest_play_ms 205.6 ms	Kernel ps3.472, Mesa ps3.545~dev the game's resolution, warm-up + 120 s
`lean`	fps whole run 21 cpu 307.4 seconds underruns 0 underruns ram_free 27.8 MiB vram 95.9 MiB longest_ms 143.7 ms stall_disk_s 0.0 seconds stall_swap_s 0 seconds stall_reclaim_s 0 seconds stall_gpu_s 0 seconds stall_cpu_s 0.5 seconds stall_play_s 0.6 seconds longest_play_ms 143.7 ms	Kernel ps3.472, Mesa ps3.545~dev the game's resolution, warm-up + 120 s

Observations: same boot (2026-10-20 02:37-03:27Z), 10 runs rotated. Steady 22 against 20 fps (+10%), 1% low 12.3 against 11.6, gameplay stalls 0.6 against 1.2 s, CPU 307 against 320 s (n=5 per arm, none interfered). A gain, candidate default

batch ssdroot8 DUMPON=1 repro of the OpenAL wedge: SuperTuxKart x4 OpenArena x3 ClassiCube x3 (SSD root .472 Mesa ps3.545~dev, stock snd_ps3)

2026-10-20 · batch 2026-10-20-ssdroot8-openal-wedge · 10 runs in the batch (all games)

Variant	Conditions	Steady fps in the window	1% low fps	Stalls frames over 100 ms
`lean--`	stall recorder at 10 ms	-	-	-
`lean`	stall recorder at 10 ms	20.5	11.7	6.0

All numbers and setup of this batch

Variant	Numbers	Setup
`lean--`	fps whole run 19.5 ram_free 13.2 MiB	Kernel ps3.472, Mesa ps3.545~dev the game's resolution, whole run
`lean`	fps whole run 20.0 cpu 320.3 seconds underruns 0.0 underruns ram_free 23.0 MiB vram 95.5 MiB longest_ms 142.2 ms stall_disk_s 0.0 seconds stall_swap_s 0.0 seconds stall_reclaim_s 0.0 seconds stall_gpu_s 0.0 seconds stall_cpu_s 0.7 seconds stall_play_s 0.8 seconds longest_play_ms 142.2 ms	Kernel ps3.472, Mesa ps3.545~dev the game's resolution, warm-up + 120 s

Observations: investigation runs, not for fps (a dump freezes the game for seconds and DUMPON costs), so the cells come only from the runs without a dump and are not comparable with other rows. 2026-10-20 00:39-01:25Z, same boot as the rest of ssdroot8. Webcam picture right in all 10. Dumps taken in 8 of 10 runs. Which runs wedged OpenAL (Wait timeout lines in game.log; the trigger, dump= and dump_at in bench.txt): SuperTuxKart 00:39Z none (1 xrun line), 00:45Z dump at 10.9 s on a Wait timeout, 291 lines, the card's hw pointer stood still 1.0 s before the trigger (the DMA stopped, as in cf64bd4), 00:51Z dump at 30.6 s on a mmap commit error, then 29 Wait timeout lines (the pointer still moved at the trigger), 00:57Z dump at 33.1 s on a mmap commit error, 2 lines, no Wait timeout, pointer moving. OpenArena 01:03Z, 01:07Z and 01:10Z dumps at 21.5, 20.5 and 33.4 s, all on a mmap commit error (2, 2 and 5 lines), no Wait timeout, pointer moving in all three. ClassiCube 01:15Z dump at 2.9 s on a mmap commit error, then 12 Wait timeout lines; 01:18Z none (0 lines); 01:22Z dump at 4.4 s on a mmap commit error, 1 line, pointer moving. So Wait timeout (the wedge) hit 2 of 4 SuperTuxKart, 0 of 3 OpenArena and 1 of 3 ClassiCube, 3 of 10; the other 5 dumps are client xruns that recovered. The SuperTuxKart runs of 00:51Z and 00:57Z have an empty bench.txt and a game.log cut at 12288 and 10855 bytes (a 164 MB core in /dev/shm on a 256 MB console, the likely cause, not checked): their counts come from game.log and the runs make no cell

batch ssdroot8 Mesa 0038 force-DXT on the rest of the suite: Luanti and ClassiCube off/opaque/opaque+NEAREST, OpenArena gl2min and gl2min_tc0 off/opaque, SuperTuxKart lean opaque/all (SSD root .472 Mesa ps3.545~dev)

2026-10-20 · batch 2026-10-20-ssdroot8-force-dxt · 24 runs in the batch (all games)

Variant	Conditions	Steady fps in the window	1% low fps	Stalls frames over 100 ms
`lean`	stall recorder at 10 ms switch force_dxtall_force_dxt_stats50	20.5	11.4	10.5
`lean`	stall recorder at 10 ms switch force_dxtopaque_force_dxt_stats50	20.0	11.2	12.0

All numbers and setup of this batch

Variant	Numbers	Setup
`lean`	fps whole run 19.5 cpu 316.2 seconds underruns 0.0 underruns ram_free 28.6 MiB vram 58.0 MiB longest_ms 225.7 ms stall_disk_s 0.0 seconds stall_swap_s 0.0 seconds stall_reclaim_s 0.0 seconds stall_gpu_s 0.0 seconds stall_cpu_s 1.4 seconds stall_play_s 1.4 seconds longest_play_ms 225.7 ms	Kernel ps3.472, Mesa ps3.545~dev the game's resolution, warm-up + 120 s
`lean`	fps whole run 20.0 cpu 316.6 seconds underruns 0.0 underruns ram_free 28.1 MiB vram 95.5 MiB longest_ms 309.8 ms stall_disk_s 0.1 seconds stall_swap_s 0.0 seconds stall_reclaim_s 0.0 seconds stall_gpu_s 0.0 seconds stall_cpu_s 1.6 seconds stall_play_s 1.8 seconds longest_play_ms 309.8 ms	Kernel ps3.472, Mesa ps3.545~dev the game's resolution, warm-up + 120 s

Observations: same boot (2026-10-19 23:01Z to 2026-10-20 00:39Z), 24 runs rotated per game, MEM=1 STALLS=1, NV30_FORCE_DXT_STATS=50 in every arm (the plain column carries only that), none interfered, webcam picture right in all (Luanti world, ClassiCube terrain, OpenArena spectator view, SuperTuxKart track). Nothing was forced in four of the five games, so the numbers are equal inside the noise (VRAM in MiB as the vram matrix has it): Luanti (n=2 per arm) tex=0, skip_nodata 516 (every upload is a NULL glTexImage2D filled by sub-images) and skip_format 1; peak VRAM 62.4 (off, 63.1 and 61.6), 60.9 (opaque), 61.9 (+NEAREST); steady 46.5, 46.0, 46.0 fps (HUD); 1% low 17.4, 17.2, 16.4; stalls 17.5, 17.5, 20.5 frames (3.3, 3.3, 3.9 s); load 25.5, 25.2, 25.1 s. ClassiCube tex=0, skip_small 214 (under 64 px) and 10 textures kept as NEAREST-filtered (opaque) or with alpha (+NEAREST): VRAM 60.1 MiB in every arm, 47.2-47.3 fps, 1% low 30.0, 30.6, 31.0, no stalls, load 7.8-8.0 s. OpenArena gl2min tex=0 (skip_format 2593, skip_nodata 8): off 61.8 against opaque 62.1 steady fps, VRAM 114.4-116.2 against 116.9-117.0 MiB, gameplay stalls 0.2-0.3 s both, CPU 187 s off against 187 and 216 s (one outlier) with opaque; gl2min_tc0 (the game's own compression off, 145-149 MiB VRAM) tex=0 (skip_nodata 2244: textures made with a NULL glTexImage2D): off 58.6 against opaque 58.9 fps, VRAM 144.8-148.7 against 147.5-149.0 MiB. SuperTuxKart lean opaque forces nothing again (skip_alpha 182, skip_format 173, skip_small 22, skip_npot 6), while all takes 182 textures as DXT5 (52-53 MB of levels sent as 13 MB; 5 textures, 2.8 MB, go back to uncompressed): peak VRAM 95.5 MiB (96.1, 95.0) with opaque against 58.1 MiB (58.2, 57.9) with all, -37 MiB or -39%; steady 20 fps (HUD) against 20.5 (20, 21); 1% low 11.3 against 11.4; frames over 100 ms 12 against 10.5 (1.8 against 1.4 s), longest gameplay frame 310 against 226 ms; CPU 323 and 311 s against 307 and 326 s; load 28.1 against 28.4 s (mipgen_ms 0, scan_ms 1-3: nothing visible); lowest MemAvailable 26-32 MiB in both arms (the VRAM saved is VRAM, not RAM). Picture: the SuperTuxKart photos with all show the same track, sky, kart and dirt road as with opaque (different moments of the race), no blockiness or banding at the webcam's resolution; DXT5's loss on fine detail is not judged by a webcam, so the user has the last word. OpenAL wedge lines (Wait timeout) in 9 of the 24 runs: Luanti 2 of 6 (160, 194 lines), ClassiCube 2 of 6 (14, 9), OpenArena 3 of 8 (11, 19, 7), SuperTuxKart 2 of 4 (275, 277), in off arms as well, the wedged runs' fps unchanged

batch ssdroot7 DUMPON=1 repro of the OpenAL wedge: SuperTuxKart x4 OpenArena x3 ClassiCube x3 Sauerbraten x3 (SSD root .472 Mesa ps3.545~dev)

2026-10-19 · batch 2026-10-19-ssdroot7-openal-wedge · 8 runs in the batch (all games)

Variant	Conditions	Steady fps in the window	1% low fps	Stalls frames over 100 ms
`lean`	stall recorder at 10 ms	20	11.3	8

All numbers and setup of this batch

Variant	Numbers	Setup
`lean`	fps whole run 19 cpu 323.1 seconds underruns 0 underruns ram_free 25.6 MiB vram 95.0 MiB longest_ms 126.9 ms stall_disk_s 0.0 seconds stall_swap_s 0 seconds stall_reclaim_s 0.0 seconds stall_gpu_s 0 seconds stall_cpu_s 0.8 seconds stall_play_s 0.9 seconds longest_play_ms 126.9 ms	Kernel ps3.472, Mesa ps3.545~dev the game's resolution, warm-up + 120 s

Observations: investigation runs, not for fps: a dump freezes the game for seconds, and DUMPON itself costs (OpenArena's window was 71 s and 57 fps against 94 s and 62 fps without it), so the cells come from the 8 of 13 runs without a dump and are not comparable with other rows. Webcam picture right in all 13 (SuperTuxKart track, OpenArena spectator view, ClassiCube terrain, Sauerbraten cave). Which runs wedged OpenAL (oal_wedge, dump trigger; 2026-10-19 UTC): SuperTuxKart 20:53Z dump at 48 s (76 log lines of Wait timeout), 21:00Z none, 21:05Z dump at 33 s (2 mmap commit error), 21:11Z dump at 286 s (25 lines); the three dumped SuperTuxKart runs have an empty bench.txt, their counts are from game.log and the runs are not in the cells. OpenArena 21:25Z dump at 21.6 s (75, mmap commit error), 21:18Z and 21:21Z none. ClassiCube 21:36Z dump at 103.6 s (101, Wait timeout), 21:29Z and 21:33Z none. Sauerbraten 3 of 3 clean (underruns 96-171). So the wedge hit 3 of 4 SuperTuxKart, 1 of 3 OpenArena, 1 of 3 ClassiCube and 0 of 3 Sauerbraten, as early as 21 s into a run. The finding from the dumps (the sound card's DMA stops: snd_ps3's interrupt chain) is in commit cf64bd4

batch ssdroot7 Mesa 0044: NV30_MINMAX_VMX=1 against unset on SuperTuxKart lean x3 (SSD root .472 Mesa ps3.545~dev)

2026-10-19 · batch 2026-10-19-ssdroot7-minmax-vmx · 6 runs in the batch (all games)

Variant	Conditions	Steady fps in the window	1% low fps	Stalls frames over 100 ms
`lean`	stall recorder at 10 ms	21	12.0	6
`lean`	stall recorder at 10 ms switch minmax_vmx1	22	12.3	6

All numbers and setup of this batch

Variant	Numbers	Setup
`lean`	fps whole run 20 cpu 305.2 seconds underruns 0 underruns ram_free 24.4 MiB vram 94.6 MiB longest_ms 231.5 ms stall_disk_s 0.1 seconds stall_swap_s 0.0 seconds stall_reclaim_s 0 seconds stall_gpu_s 0 seconds stall_cpu_s 0.6 seconds stall_play_s 0.8 seconds longest_play_ms 231.5 ms	Kernel ps3.472, Mesa ps3.545~dev the game's resolution, warm-up + 120 s
`lean`	fps whole run 21 cpu 313.2 seconds underruns 0 underruns ram_free 26.5 MiB vram 96.1 MiB longest_ms 185.2 ms stall_disk_s 0.1 seconds stall_swap_s 0 seconds stall_reclaim_s 0.1 seconds stall_gpu_s 0 seconds stall_cpu_s 0.7 seconds stall_play_s 0.8 seconds longest_play_ms 185.2 ms	Kernel ps3.472, Mesa ps3.545~dev the game's resolution, warm-up + 120 s

Observations: same boot (2026-10-19 20:21-20:51Z), 6 runs rotated unset, 1, 1, unset, unset, 1 (the rec10 and minmax_vmx1 columns); webcam picture right in all 6 (the same track, other karts in some). n=3 per arm, none interfered. HUD steady fps: unset 21, 19, 21; minmax_vmx1 22, 22, 22 (the HUD median is an integer; the rate over the frames under 100 ms is 20.7 against 21.9, +6%). 1% low 12.0 against 12.3; gameplay stalls 6 (0.8 s) against 6 (0.8 s). CPU did not fall: 305 s unset, 313 s with it (the expected -5%). The ranges do not overlap, but one fps in 21 is the HUD's resolution, so indicative +5%, not proven. oal_wedge above 0 in all 6 and above 200 in 3 (unset 266 and 240 lines, vmx 246; the others 1-3): the wedged unset runs made 21 and 21, the one with 2 lines 19, so no sign that the wedge costs fps. load 28.1-28.4 s everywhere

batch ssdroot6 libpng: libpng16 1.6.59 +ps3.560 against Debian's stock on SuperTuxKart lean x3 and Sauerbraten base x3 and OpenArena gl2min x2 (SSD root .472 Mesa ps3.545~dev zlib-ng on both arms)

2026-10-19 · batch 2026-10-19-ssdroot6-libpng · 15 runs in the batch (all games)

Variant	Conditions	Steady fps in the window	1% low fps	Stalls frames over 100 ms
`lean`	stall recorder at 10 ms start-up profile switch zng	19.5	11.5	10.0
`lean`	stall recorder at 10 ms start-up profile switch zng switch ldstock-png	20.0	11.8	10.5

All numbers and setup of this batch

Variant	Numbers	Setup
`lean`	fps whole run 19.5 cpu 320.4 seconds underruns 0.0 underruns ram_free 19.8 MiB vram 95.4 MiB longest_ms 172.6 ms stall_disk_s 0.1 seconds stall_swap_s 0.0 seconds stall_reclaim_s 0.0 seconds stall_gpu_s 0.0 seconds stall_cpu_s 1.1 seconds stall_play_s 1.2 seconds longest_play_ms 172.6 ms	Kernel ps3.472, Mesa ps3.545~dev the game's resolution, warm-up + 120 s
`lean`	fps whole run 19.5 cpu 322.5 seconds underruns 0.0 underruns ram_free 27.3 MiB vram 95.7 MiB longest_ms 198.8 ms stall_disk_s 0.1 seconds stall_swap_s 0.0 seconds stall_reclaim_s 0.0 seconds stall_gpu_s 0.0 seconds stall_cpu_s 1.2 seconds stall_play_s 1.4 seconds longest_play_ms 198.8 ms	Kernel ps3.472, Mesa ps3.545~dev the game's resolution, warm-up + 120 s

Observations: same boot (2026-10-19 16:51-18:10Z), the stock arm is the column with the ldstock-png token, webcam picture right in all 22 runs (60 s and 150 s photos; OpenArena's 150 s photo is the map reload). SuperTuxKart lean, 3 rounds rotated, PERF_LOAD=1: round 1 (16:51Z stock, 16:57Z ours) is interfered (a login at 9 s, deliberate mouse moves) and out of the cells, so n=2 per arm, and its profile agrees with the rest. The png group of the load profile: stock 4.00 and 4.05 s (4.17 s in the interfered run), ours 1.61 and 1.61 s (1.67 s): -2.4 s, -60%, inside the predicted 2-3.3 s; the zlib group is 3.4 s in both. Load 30.3 s (30.42, 30.25) against 27.6 s (26.93, 28.20): -2.8 s, -9%; the main thread's CPU over the load 24.0 against 21.1-21.7 s. Gameplay unchanged: steady 20 against 19.5 fps (HUD), 1% low 11.9 against 11.5, 10 against 10 frames over 100 ms (1.4 against 1.25 s). The stock run of 17:16Z ended with rc=134 (SIGABRT) after the full race (21602 ticks), cause not looked at. Sauerbraten base, no load profile: load 7.19 s stock against 7.25 s ours, steady 39.5 fps both, 1% low 27.1 against 26.6, 0-1 stalls both, so libpng is too small a part of its load for the timer to see. OpenArena gl2min x2 as the control: load 6.80 against 6.62 s, steady 63.1 against 61.1 fps (59.6 to 63.5, noise), gameplay stalls 0 both; CPU 231-241 s with the stock libs against 211 s with ours, not explained (OpenArena loads few PNGs; the plain runs of the next row also have 211-213 s). OpenAL wedge lines (Wait timeout, mmap commit error) in 8 of the 10 SuperTuxKart runs of this and the previous batch, both arms, not counted in bench.txt yet. logscan: nothing new. Proven: the png group and SuperTuxKart's load; nothing measurable for Sauerbraten and OpenArena

batch ssdroot5 load-time libraries: libvorbis +ps3.549 and zlib-ng against stock on SuperTuxKart lean and Sauerbraten base x3, JSIMD=none x2 (SSD root .472 Mesa ps3.544)

2026-10-19 · batch 2026-10-19-ssdroot5-load-libraries · 22 runs in the batch (all games)

Variant	Conditions	Steady fps in the window	1% low fps	Stalls frames over 100 ms
`lean`	stall recorder at 10 ms start-up profile	21	11.9	6
`lean`	stall recorder at 10 ms start-up profile switch jsimdnone	20.5	11.9	8.5
`lean`	stall recorder at 10 ms start-up profile switch ldstock-vorbis	21	12.0	5
`lean`	stall recorder at 10 ms start-up profile switch zng	20	11.7	7

All numbers and setup of this batch

Variant	Numbers	Setup
`lean`	fps whole run 20 cpu 321.9 seconds underruns 0 underruns ram_free 27.9 MiB vram 95.9 MiB longest_ms 151.8 ms stall_disk_s 0 seconds stall_swap_s 0 seconds stall_reclaim_s 0.1 seconds stall_gpu_s 0 seconds stall_cpu_s 0.7 seconds stall_play_s 0.8 seconds longest_play_ms 151.8 ms	Kernel ps3.472, Mesa ps3.544 the game's resolution, warm-up + 120 s
`lean`	fps whole run 20.0 cpu 321.4 seconds underruns 0.0 underruns ram_free 31.0 MiB vram 95.4 MiB longest_ms 442.6 ms stall_disk_s 0.1 seconds stall_swap_s 0.0 seconds stall_reclaim_s 0.2 seconds stall_gpu_s 0.0 seconds stall_cpu_s 1.0 seconds stall_play_s 1.4 seconds longest_play_ms 442.6 ms	Kernel ps3.472, Mesa ps3.544 the game's resolution, warm-up + 120 s
`lean`	fps whole run 20 cpu 320.5 seconds underruns 0 underruns ram_free 25.4 MiB vram 96.1 MiB longest_ms 427.2 ms stall_disk_s 0.1 seconds stall_swap_s 0 seconds stall_reclaim_s 0.3 seconds stall_gpu_s 0 seconds stall_cpu_s 0.6 seconds stall_play_s 1.2 seconds longest_play_ms 427.2 ms	Kernel ps3.472, Mesa ps3.544 the game's resolution, warm-up + 120 s
`lean`	fps whole run 20 cpu 325.4 seconds underruns 0 underruns ram_free 24.9 MiB vram 95.9 MiB longest_ms 141.2 ms stall_disk_s 0.1 seconds stall_swap_s 0 seconds stall_reclaim_s 0 seconds stall_gpu_s 0 seconds stall_cpu_s 0.8 seconds stall_play_s 0.8 seconds longest_play_ms 141.2 ms	Kernel ps3.472, Mesa ps3.544 the game's resolution, warm-up + 120 s

Observations: same boot (2026-10-19 14:29-16:35Z), PERF_LOAD=1, 22 runs. SuperTuxKart load 31.5 s with the new libvorbis against 32.2 s stock (-0.7 s), 30.6 s with zlib-ng (-0.9 s), 32.6 s with JSIMD=none (+1 s: libjpeg-turbo's AltiVec path is in use). Sauerbraten loads unchanged within the spread (8.4 / 7.5 / 8.1 s). Gameplay fps unchanged

batch ssdroot2 P2: Mesa 0038 NV30_FORCE_DXT=opaque against off (MEM=1 STALLS=1 NV30_FORCE_DXT_STATS=50)

2026-10-19 · batch 2026-10-19-ssdroot2-force-dxt · 12 runs in the batch (all games)

Variant	Conditions	Steady fps in the window	1% low fps	Stalls frames over 100 ms
`lean`	stall recorder at 10 ms switch force_dxt_stats50	20.0	10.6	17.5
`lean`	stall recorder at 10 ms switch force_dxtopaque_force_dxt_stats50	21.0	11.6	11.5

All numbers and setup of this batch

Variant	Numbers	Setup
`lean`	fps whole run 19.5 cpu 323.1 seconds underruns 0.0 underruns ram_free 27.6 MiB vram 96.2 MiB longest_ms 480.2 ms stall_disk_s 0.2 seconds stall_swap_s 0.0 seconds stall_reclaim_s 0.2 seconds stall_gpu_s 0.0 seconds stall_cpu_s 2.4 seconds stall_play_s 2.8 seconds longest_play_ms 480.2 ms	no build recorded the game's resolution, warm-up + 120 s
`lean`	fps whole run 20.0 cpu 318.3 seconds underruns 0.0 underruns ram_free 24.6 MiB vram 95.0 MiB longest_ms 385.5 ms stall_disk_s 0.1 seconds stall_swap_s 0.0 seconds stall_reclaim_s 0.3 seconds stall_gpu_s 0.0 seconds stall_cpu_s 1.3 seconds stall_play_s 1.7 seconds longest_play_ms 385.5 ms	no build recorded the game's resolution, warm-up + 120 s

Observations: same boot (2026-10-19 02:35-03:25Z), x2 interleaved per game, both arms with NV30_FORCE_DXT_STATS=50. Sauerbraten: 73 textures compressed (88 MB raw to 11 MB stored, all DXT1; 4 turned back, 17 skipped for alpha), peak VRAM 119.8 MB (the cap) to 73-78 MB, ps3rsx buffers at the middle snapshot 351 to 215 MB, RSS -3.5 MB; steady 39.9 off against 39.0 opaque (the unprofiled baseline runs span 39.1-40.3, so no fps change resolved), 1% low 26.0 against 23.9 (n=2, spread 22.9-26.2), stalls 0.5 against 0.55 s; the picture is slightly softer and darker on the webcam (grey difference 10 against 1-2 between repeats): lossy as expected. SuperTuxKart lean: 0 textures forced (skip_alpha 182, skip_format 173, skip_small 22, skip_npot 6: the game's own compression 0004 already covers its opaque ones), VRAM 98.5 against 97.3 MB, HUD 20 against 21 fps, stalls 2.8 against 1.7 s: noise. OpenArena gl2min: 0 forced (skip_format 2238: every upload is in a format the switch does not take), VRAM 120.1 against 119.7, steady 56.3 against 55.8: no effect. OpenArena runs 56 in both arms against 60-63 in the unprofiled baseline earlier in the boot (STATS=50 or drift, not separated). SuperTuxKart fps cells are the HUD.

batch ssdroot2 P1b: gameplay profiles on the SSD root (PERF=1 MEM=1 x1)

2026-10-19 · batch 2026-10-19-ssdroot2-gameplay-profiles · 7 runs in the batch (all games)

Not comparable: gameplay profiles (PERF=1 MEM=1), slower than plain runs.

Variant	Conditions	Steady fps in the window	1% low fps	Stalls frames over 100 ms
`lean` not comparable	perf recording	20	10.8	20

All numbers and setup of this batch

Variant	Numbers	Setup
`lean`	fps whole run 20 cpu 322.5 seconds underruns 0 underruns ram_free 22.9 MiB vram 94.5 MiB longest_ms 316.8 ms stall_play_s 2.7 seconds longest_play_ms 316.8 ms	no build recorded the game's resolution, warm-up + 120 s

Observations: same boot (2026-10-19 02:01-02:30Z), one run per game, perf at 499 Hz over the window (profiled runs are for profiles, not fps: OpenArena 56 against 60 unprofiled). Share of the game process's samples: Sauerbraten Mesa 62.8% (kgpu 4.8, game 10.5), ClassiCube Mesa 41.2 + kernel 20.1 + kgpu 7.1, SuperTuxKart game 36.5 + Mesa 35.5 + libc 15.1, OpenArena game 27.6 + Mesa 23.1 + other 20.4 (OpenAL 12.4, libdeflate 3.9), Luanti game 47.2 (MeshUpdate thread 20.7% of all samples) + Mesa 14.6 + libc 13.7, dsda-doom Mesa 22.3 + game 20.5 + audio 30 (FluidSynth 15, libspeq 12), OpenTTD kernel 49.5 (36.7 of them in __replay_soft_interrupts, 8.3 in HPTE invalidation, 5.1 in ps3rsx_gart_clear from buffer creation) + audio 17. OpenTTD ran at its title screen (a menu, not gameplay: 18 fps, 4172 underruns). Mesa has no symbols in these runs (perf.data not pulled: PC-side symbolisation later). SuperTuxKart fps cells are the HUD.

batch ssdroot2 P1a: start-up profiles on the SSD root (PERF_LOAD=1 STALLS=1 x1)

2026-10-19 · batch 2026-10-19-ssdroot2-startup-profiles · 5 runs in the batch (all games)

Not comparable: start-up profiles (PERF_LOAD=1 STALLS=1), slower than plain runs.

Variant	Conditions	Steady fps in the window	1% low fps	Stalls frames over 100 ms
`lean` not comparable	stall recorder at 10 ms start-up profile	21	12.0	5

All numbers and setup of this batch

Variant	Numbers	Setup
`lean`	fps whole run 20 cpu 316.1 seconds underruns 0 underruns ram_free 23.5 MiB vram 95.7 MiB longest_ms 151.3 ms stall_disk_s 0 seconds stall_swap_s 0 seconds stall_reclaim_s 0 seconds stall_gpu_s 0 seconds stall_cpu_s 0.6 seconds stall_play_s 0.6 seconds longest_play_ms 151.3 ms	no build recorded the game's resolution, warm-up + 120 s

Observations: same boot as the baseline (2026-10-19 01:34-02:00Z), cold page cache, one run per game. load_end (exec to the end of the loading screens, loadprof.py) against the USB profiles of batch 17: Sauerbraten 7.7 s (19.8), OpenArena 14.2 (28.1), Luanti 26.1 (43.1), SuperTuxKart 31.5 (54.3), ClassiCube 7.5 (13.9): -39 to -61%. The main thread's disk wait in the load fell from 4.8-18.6 s to 1.0-3.2 s (SB 1.9, OA 2.1, Luanti 3.0, STK 3.2, CB 1.0) and its CPU stayed (SB 3.9 s against 4.1, OA 8.7 against 9.0, STK 23.8 against 22.5, CB 3.2 against 2.9), so the loads are now 47-77% CPU on the main thread, not fully CPU-bound (disk 13-26%, rest: GPU, reclaim, runq, sleeps). On top now: Sauerbraten zlib 1.6 s (map inflate), OpenArena the game's own load path 3.9 s (shader scripts, mipmaps), Luanti mapgen on Emerge-0 8.6 s plus the 15 s hold, STK png 4.2 + zlib 3.9 + audio 2.2 + jpeg 2.1 s, ClassiCube the game 1.5 s and ld.so 0.6 s. Profiled runs are for profiles, not fps.

batch ssdroot1+2 A: baseline on the SSD root (kernel .472 Mesa .476 ps3rsx 0.443 ps3-gpu-tools 0.477~dev)

2026-10-19 · batch 2026-10-19-ssdroot1-baseline · 12 runs in the batch (all games)

Variant	Conditions	Steady fps in the window	1% low fps	Stalls frames over 100 ms
`lean`	stall recorder at 10 ms	22	12.8	2

All numbers and setup of this batch

Variant	Numbers	Setup
`lean`	fps whole run 21 cpu 308.3 seconds underruns 0 underruns ram_free 25.1 MiB vram 94.8 MiB longest_ms 108.3 ms stall_disk_s 0 seconds stall_swap_s 0 seconds stall_reclaim_s 0 seconds stall_gpu_s 0 seconds stall_cpu_s 0.2 seconds stall_play_s 0.2 seconds longest_play_ms 108.3 ms	Kernel ps3.472, Mesa ps3.476, ps3rsx 0.443 the game's resolution, warm-up + 120 s

Observations: same boot 2026-10-19 00:22-01:30Z, kernel 6.12.111+ps3.472, root on the internal SSD image (this console has an upgraded SSD; a stock PS3 has an HDD), STALLS=1 ROTATE=1, the stall recorder samples ps3db now. 5 of 17 runs excluded as interfered (SSH logins and a key press): Sauerbraten 0022 (the check run) and 0034, SuperTuxKart 0101 0106 0112 (a polling loop at the batch switch). Against the USB root (cross-boot: ssd1 '-' arm for Sauerbraten and OpenArena, batch 16 on .433 and older Mesa for the rest) the game's load is 7.1 s against 16.9 (Sauerbraten), 6.9 against 10.7 (OpenArena), 25.0 against 40.8 (Luanti), 31.1 against 53.6 (STK), 7.6 against 14.2 (ClassiCube); the load's disk wait is 1.3-3.4 s against 5.5-10.9; gameplay stalls Sauerbraten 0.05 s against 1.6, OpenArena 0.2 against 0.3, Luanti 3.5 against 2.5 (reclaim 1.6 s against 0.4), STK 0.2 s (n=1); steady fps about the same (Sauerbraten 39.1 against 38.5, Luanti 46 against 46, STK 22 by the HUD against 20, ClassiCube 48.1 against 45.9, falling to 46.1 from 02:00 on with nothing changed) except OpenArena 60.4 against 55.9. SuperTuxKart fps cells are the HUD median now: its own 'Average FPS' (105-111) is the physics tick rate (120 Hz x 180 s = 21602 ticks in every run), not rendered frames. Photos show the right scenes; Luanti's 60 s and 150 s photos fall in the path's turn over the sea (cyan, no terrain) because the load is shorter than on USB, not a missing world. Mic: OpenArena audible, Sauerbraten/STK/Luanti/ClassiCube mostly at the room-hum floor.

batch 17: ETR uncapped, start-up profiles x5, shader cache cold/warm, OpenArena gl2 memory variants, Mesa sysmem switches, Sauerbraten shortnormals

2026-10-18 · batch 2026-10-18-batch17 · 29 runs in the batch (all games)

Variant	Conditions	Steady fps in the window	1% low fps	Stalls frames over 100 ms
`lean`	stall recorder at 10 ms start-up profile	19	11.5	8

All numbers and setup of this batch

Variant	Numbers	Setup
`lean`	fps whole run 19 cpu 335.3 seconds underruns 0 underruns ram_free 22.5 MiB vram 96.1 MiB longest_ms 691.2 ms stall_disk_s 0.4 seconds stall_swap_s 0.0 seconds stall_reclaim_s 0 seconds stall_gpu_s 0 seconds stall_cpu_s 0.7 seconds stall_play_s 2.0 seconds longest_play_ms 691.2 ms	no build recorded the game's resolution, warm-up + 120 s

Observations: same boot as batch 16 (kernel 433), ioquake3 462, Mesa 442~dev, cube2 442~dev, ps3-gpu-tools 0.469; 1 run excluded (sauerbraten shortnormals round 3: a key press). OpenArena gl2min on ioquake3 462 (snorm colours default): steady 57-60, stalls 0.7-2.6 s (batch 16 on 396: 51-53, 9-11 s); gl2min_color16 (old colour path) 52-53, stalls 12.9 s, swap 20 s; md3pack -20 MB VRAM, same stalls; mippowf lowest stalls (0.7-0.9 s). Mesa NV30_SYSMEM_COPY=2+NV30_INVALIDATE_BUFFER=1 on color16: stalls 9.9/12.3 -> 1.5/1.4 s, swap 20-26 -> 0.6-1.1 s, steady 53-54 -> 59. Sauerbraten shortnormals 38.7 -> 45.4 (+17%), stalls equal, VRAM +4 MB. Shader cache cold vs warm: no load gain (OA 10.3 cold vs 10.6 warm, Sauerbraten 17.4 vs 18.4). ETR uncapped still 59: not a vsync cap

batch 16: game prefetch A/B, idle fix on/off (433), ClassiCube chunks1, X-Moto stream, Luanti lean parts

2026-10-18 · batch 2026-10-18-batch16 · 41 runs in the batch (all games)

Variant	Conditions	Steady fps in the window	1% low fps	Stalls frames over 100 ms
`lean`	gpflearn	21	12.1	6
`lean`	stall recorder at 10 ms	20	11.4	7
`lean`	stall recorder at 10 ms switch gpffetch	20	11.5	12

All numbers and setup of this batch

Variant	Numbers	Setup
`lean`	fps whole run 20 cpu 331.1 seconds underruns 0 underruns ram_free 26.7 MiB vram 95.9 MiB longest_ms 432.6 ms stall_play_s 1.3 seconds longest_play_ms 432.6 ms	no build recorded the game's resolution, warm-up + 120 s
`lean`	fps whole run 19 cpu 330.8 seconds underruns 0 underruns ram_free 23.9 MiB vram 95.2 MiB longest_ms 686.1 ms stall_disk_s 0.4 seconds stall_swap_s 0.0 seconds stall_reclaim_s 0 seconds stall_gpu_s 0 seconds stall_cpu_s 0.7 seconds stall_play_s 1.8 seconds longest_play_ms 686.1 ms	no build recorded the game's resolution, warm-up + 120 s
`lean`	fps whole run 20 cpu 330.6 seconds underruns 0 underruns ram_free 24.4 MiB vram 96.1 MiB longest_ms 682.4 ms stall_disk_s 0.5 seconds stall_swap_s 0.0 seconds stall_reclaim_s 0 seconds stall_gpu_s 0 seconds stall_cpu_s 1.1 seconds stall_play_s 2.7 seconds longest_play_ms 682.4 ms	no build recorded the game's resolution, warm-up + 120 s

Observations: same boot 2026-10-18 ~18:00-20:25 local, kernel 433 (idle fix, no kicker), interference detector on (3 interfered runs excluded: ssh logins). Prefetch fetch vs base x3: no gain (Sauerbraten load 18.1-18.6 vs 15.2-18.8 s; OA gl2min ~0.5 s faster; Luanti WORSE: load +1-4 s, two runs with 3 s hitches): turn the default off. Idle fix on/off x2: stall_play_s 0.4/0.4 vs 9.0/7.9 s, longest 0.22-0.38 vs 2.2-2.8 s (clean). ClassiCube chunks1 48.1 vs 46.0 (+4.7%). X-Moto stream 162.1 vs 160.6 (+1%). Luanti chunk3 renders a black void at 150 s (the lean culprit); srvunload/active2 render but stall more (6.3/6.7 s vs ~2.5)

batch 8: STK lean, OpenTTD gl, ClassiCube/dsda batching, malloc presets, profiles (Mesa ps3.408, kick)

2026-10-18 · batch 2026-10-18-batch8 · 26 runs in the batch (all games)

Variant	Conditions	Steady fps in the window	1% low fps	Stalls frames over 100 ms
`base`	waker kick at 10 ms	12	8.4	113
`lean`	waker kick at 10 ms	21	12.2	8
`lean`	waker kick at 10 ms perf recording	20	10.8	15

All numbers and setup of this batch

Variant	Numbers	Setup
`base`	fps whole run 12 cpu 412.3 seconds underruns 0 underruns ram_free 24.4 MiB vram 129.7 MiB longest_ms 675.6 ms stall_play_s 13.5 seconds longest_play_ms 675.6 ms	Mesa ps3.408 the game's resolution, warm-up + 120 s
`lean`	fps whole run 21 cpu 318.1 seconds underruns 0 underruns ram_free 18.5 MiB vram 95.7 MiB longest_ms 663.4 ms stall_play_s 2.4 seconds longest_play_ms 663.4 ms	Mesa ps3.408 the game's resolution, warm-up + 120 s
`lean`	fps whole run 20 cpu 325.4 seconds underruns 0 underruns ram_free 27.6 MiB vram 95.0 MiB longest_ms 576.3 ms stall_play_s 3.4 seconds longest_play_ms 576.3 ms	Mesa ps3.408 the game's resolution, warm-up + 120 s

Observations: same boot 2026-10-18 ~09:30-11:50 local. SuperTuxKart: TRACK DRAWS (webcam), lean smooth 21.4 vs base 13, stalls 2.4 vs 13.5 s. OpenTTD gl: right picture, 21 fps (software 8-12). ClassiCube base (batching) 47.3 vs gloff 49.4 (-4%), low1 31 vs 27.6. dsda gl 55.9 vs gloff 54.2 (+3%), low1 40.8 vs 36.5. Malloc presets inconsistent between rounds (noise). OA gl2min 55 s stalls, longest 16 s, 66 MB swapped; Luanti 10 s stalls, 51 MB swapped INTERFERENCE (retro-scan 2026-10-18, docs/STALLS.md): luanti-base@wkkick10+perf (20261018T081125Z) had a right click in the window: distrust its cell.

Group 4 smokes (Mesa ps3.246 + ps3rsx 0.249)

2026-10-17 · batch 2026-10-17-group4-smokes · 6 runs in the batch (all games)

Variant	Conditions	Steady fps in the window	1% low fps	Stalls frames over 100 ms
`base`	the recipe as it is	-	-	-

All numbers and setup of this batch

Variant	Numbers	Setup
`base`	cpu 13.8 seconds underruns 0 underruns ram_free 48.9 MiB vram 100.0 MiB	Mesa ps3.246, ps3rsx 0.249 the game's resolution, whole run

Observations: 2026-10-17 09:49-10:30Z, heavy games, VRAM swap on. OpenArena opengl2 (gl2min, 97 GLSL programs): the first launch hit the 90 s cap before the level loaded; the rerun with 240 s ran at 13.6 fps average with one 36.3 s stall, HUD median 31, 157.8 MB VRAM and MemAvailable down to 9.8 MB. With Debian's defaults (gl2, 115 programs) 10.7 fps, a 28.3 s stall, HUD median 23, 159.7 MB VRAM. The picture was right (webcam) and no driver bug showed; the games are memory-bound. SuperTuxKart quit after 29 s with a fatal error: 'SPMeshLoader: Not little endian machine' (thunderbird.spm). Sauerbraten with GPU skinning (base) ran to the 170 s cap with HUD median 0 and no fps line (GAME-PLAN has 11 fps by its own counter against 28 for CPU skinning); CPU skinning gives 19.6 fps in 601 frames (longest 380 ms) at 135 MB VRAM, load 27 s. Single runs.