Main thread saturated · Block Physics
One game thread is already over its tick budget under the benchmark's own load. Players and redstone will push it further.
- mspt_avg = 51 ms
- mspt_target = 35 ms
Weakest test: Mob AI, 0.12 TPS on average. The 15 other tests hold at least 11.62 TPS.
What matters
Read from the measurements, never estimatedMain thread saturated
One game thread is already over its tick budget under the benchmark's own load. Players and redstone will push it further.
Likely cause: single core
- mspt_avg = 51 ms
- mspt_target = 35 ms
Main thread saturated
One game thread is already over its tick budget under the benchmark's own load. Players and redstone will push it further.
Likely cause: single core
- mspt_avg = 64.8 ms
- mspt_target = 35 ms
Non-default server configuration
Compare with care against a server running default settings.
- view-distance 4 (default 10)
- simulation-distance 3 (default 10)
- spawn-limit.monster 3 (default 70)
+4 more modified settings
›+2 more findings
mobAIbench artifactThe plugin's own witnesses say part of the workload never ran. The figure is kept, but it weighs less.
- confidence = 0.25 ratio
networkThe worst ticks are several times the mean. Average TPS stays fine while players feel the stutter.
- mspt_avg = 89 ms
- mspt_p99 = 190.35 ms
- spike_ratio = 2.14 ratio
- sample_count = 148 count
A finding stays silent rather than attribute a cause without evidence. An absent cause is an answer, not an omission.
Test by test
Problems first — select a row for the full test- RedstoneSingle-Core Stress15.44StableCritical
- Block PhysicsSingle-Core Stress19.61StableCritical
- Mob AIEntities0.12StableUnstable
- Network (NBT)Network11.62UnstableUnstable
- Lighting UpdateExploration18.32StableUnstable
- Single-Core (tick-aligned)CPU19.86StableUnstable
- Memory PerformanceMemory & Storage19.94StableUnstable
- Disk PerformanceMemory & Storage19.98StableUnstable
- ExplosionMechanics19.39StableHealthy
- Chunk LoadingExploration19.53StableHealthy
- Chunk TickingExploration19.73StableHealthy
- World SaveMechanics19.80StableHealthy
- Bone Meal GrowthMechanics19.88StableHealthy
- HopperMechanics19.92StableHealthy
- Multi-Core CPUCPU19.97StableHealthy
- Ticking Tile EntitiesMechanics19.97StableHealthy
Settings your measurements justify
Each setting below is triggered by a measurement from this report, and names the test to replay to find out whether it helped.
Stop firing an event on every hopper move
hopper.disable-move-event: trueMeasured : hopper.mspt.avg 50.2 ms (threshold 35.0 ms)
The hopper test spends more time per tick than the budget allows. Paper fires an inventory-move event for every transfer; turning it off removes that call for plugins that never listen.
Test to replay : hopper
Switch the redstone implementation
misc.redstone-implementation: ALTERNATE_CURRENTMeasured : redstone.mspt.avg 64.8 ms (threshold 35.0 ms)
Redstone costs more per tick than the budget allows. Paper ships an alternative wire implementation that computes the same result without the repeated updates of the vanilla one.
Test to replay : redstone
Shrink the monster activation range
world-settings.default.entity-activation-range.monsters: 24Measured : mobAI.mspt.avg 8044.9 ms (threshold 35.0 ms)
Mob AI costs more per tick than the budget allows. The activation range decides how far from a player a monster keeps thinking; shrinking it puts distant ones to sleep without removing them.
Test to replay : mobAI
Cap the TNT processed per tick
world-settings.default.max-tnt-per-tick: 40Measured : explosion.mspt.avg 51.6 ms (threshold 35.0 ms)
Explosions cost more per tick than the budget allows. Capping how many primed entities are processed per tick spreads a large blast over several ticks instead of one long freeze.
Test to replay : explosion
These settings trade a little gameplay fidelity for tick budget. Apply one at a time and replay the named test: the effect stays a hypothesis until you have measured it.
Briefing for a language model
Paste this into the assistant of your choice to have it rewritten in prose. It already carries every number, so the model has none left to invent.
{
"format": "voxelbench.tuning.v1",
"instruction": "Rewrite each recommendation below as plain prose for a server operator. Every number you are allowed to use is already in this document: the measured value, its threshold, and the proposed setting. Do not compute, estimate or predict any other number, and never state an expected gain. Keep every setting name and value exactly as written.",
"server": {
"type": "folia",
"minecraft": "1.21.0",
"java": "25.0.4.1"
},
"recommendations": [
{
"id": "paper.hopper-sans-evenement",
"family": "paper",
"measure": "hopper.mspt.avg",
"value": 50.2,
"unit": "ms",
"threshold": 35,
"control_test": "hopper",
"file": "config/paper-world-defaults.yml",
"setting": "hopper.disable-move-event",
"proposed_value": "true",
"current_value": null
},
{
"id": "paper.redstone-alternatif",
"family": "paper",
"measure": "redstone.mspt.avg",
"value": 64.8,
"unit": "ms",
"threshold": 35,
"control_test": "redstone",
"file": "config/paper-world-defaults.yml",
"setting": "misc.redstone-implementation",
"proposed_value": "ALTERNATE_CURRENT",
"current_value": null
},
{
"id": "spigot.portee-activation-monstres",
"family": "spigot",
"measure": "mobAI.mspt.avg",
"value": 8044.9,
"unit": "ms",
"threshold": 35,
"control_test": "mobAI",
"file": "spigot.yml",
"setting": "world-settings.default.entity-activation-range.monsters",
"proposed_value": "24",
"current_value": null
},
{
"id": "spigot.tnt-par-tick",
"family": "spigot",
"measure": "explosion.mspt.avg",
"value": 51.6,
"unit": "ms",
"threshold": 35,
"control_test": "explosion",
"file": "spigot.yml",
"setting": "world-settings.default.max-tnt-per-tick",
"proposed_value": "40",
"current_value": null
}
]
}Every measurement
Each test in full. Tests in trouble are open; the others fold to their summary line.
Single-Core Stress
RedstoneCritical15.44 TPS64.8 ms35.5s
Single-core stress via intensive redstone circuit updates.
- Heavy garbage collection3 collections, 107ms total
- The run did not hold steadyTPS avg=15.44<19
Percentiles and stability
The regions of the test's zones. Each figure is the worst region's.
- Regions measured
- 1
- Seconds measured
- 18 s
- Median TPS
- 15.50 TPS
- Effective TPS
- 15.44 TPS
- Worst sustained MSPT
- 92.4 ms
- Time degraded
- 100.0 %
- Seconds degraded
- 18 s
- Ticks over budget
- 25.9 %
Raw data (12 fields)
Block PhysicsCritical19.61 TPS51.0 ms25.6s
Single-core stress via falling-block physics.
- Heavy garbage collection5 collections, 165ms total
Percentiles and stability
The regions of the test's zones. Each figure is the worst region's.
- Regions measured
- 1
- Seconds measured
- 14 s
- Median TPS
- 20.00 TPS
- Effective TPS
- 19.61 TPS
- Worst sustained MSPT
- 67.8 ms
- Time degraded
- 0.0 %
- Seconds degraded
- 0 s
- Ticks over budget
- 24.7 %
Raw data (18 fields)
Exploration
Chunk LoadingHealthy19.53 TPS51.2 ms18.6s
Chunk generation and loading speed across dispersed zones.
- Heavy garbage collection5 collections, 582ms total
Percentiles and stability
The regions of the test's zones. Each figure is the worst region's.
- Regions measured
- 8
- Seconds measured
- 17 s
- Median TPS
- 20.00 TPS
- Effective TPS
- 19.53 TPS
- Time degraded
- 0.0 %
- Seconds degraded
- 0 s
- Ticks over budget
- 42.6 %
Raw data (12 fields)
Chunk TickingHealthy19.73 TPS50.7 ms37.1s
Server performance when ticking many loaded chunks with random tick updates.
- Heavy garbage collection7 collections, 270ms total
Percentiles and stability
The regions of the test's zones. Each figure is the worst region's.
- Regions measured
- 1
- Seconds measured
- 28 s
- Median TPS
- 20.00 TPS
- Effective TPS
- 19.74 TPS
- Time degraded
- 3.6 %
- Seconds degraded
- 1 s
- Ticks over budget
- 14.4 %
Raw data (12 fields)
Lighting UpdateUnstable18.32 TPS54.6 ms11.4s
Lighting recomputation throughput on block place / break.
- Heavy garbage collection4 collections, 93ms total
- The run did not hold steadyTPS avg=18.32<19
Percentiles and stability
The regions of the test's zones. Each figure is the worst region's.
- Regions measured
- 8
- Seconds measured
- 7 s
- Median TPS
- 20.00 TPS
- Effective TPS
- 18.32 TPS
- Time degraded
- 14.3 %
- Seconds degraded
- 1 s
- Ticks over budget
- 2.5 %
Raw data (7 fields)
Entities
Mob AIUnstable0.12 TPS8,044.9 ms31.5s
Complex mob AI behaviours (villagers, iron golems, hostile mobs).
- Unfiltered samples usedUsing unfiltered samples (filter rate: 100%)
- Heavy garbage collection9 collections, 373ms total
- The run did not hold steadyTPS avg=0.12<19, TPS P1=13.08<15
Percentiles and stability
The regions of the test's zones. Each figure is the worst region's.
- Regions measured
- 1
- Seconds measured
- 21 s
- Median TPS
- 0.00 TPS
- Effective TPS
- 0.12 TPS
- Worst sustained MSPT
- 5,203.8 ms
- Time degraded
- 100.0 %
- Seconds degraded
- 21 s
- Region starved
- yes
Raw data (30 fields)
Mechanics
HopperHealthy19.92 TPS50.2 ms10.0s
Intensive hopper chain item transfers.
Percentiles and stability
The regions of the test's zones. Each figure is the worst region's.
- Regions measured
- 8
- Seconds measured
- 8 s
- Median TPS
- 20.00 TPS
- Effective TPS
- 19.92 TPS
- Time degraded
- 0.0 %
- Seconds degraded
- 0 s
- Ticks over budget
- 1.0 %
Raw data (16 fields)
ExplosionHealthy19.39 TPS51.6 ms6.7s
Massive TNT explosions and block destruction.
Percentiles and stability
The regions of the test's zones. Each figure is the worst region's.
- Regions measured
- 8
- Seconds measured
- 5 s
- Median TPS
- 20.00 TPS
- Effective TPS
- 19.25 TPS
- Time degraded
- 20.0 %
- Seconds degraded
- 1 s
- Ticks over budget
- 4.8 %
Raw data (59 fields)
Ticking Tile EntitiesHealthy19.97 TPS50.1 ms36.9s
Many ticking tile entities (furnaces, hoppers, etc.).
- Heavy garbage collection6 collections, 193ms total
Percentiles and stability
The regions of the test's zones. Each figure is the worst region's.
- Regions measured
- 8
- Seconds measured
- 28 s
- Median TPS
- 20.00 TPS
- Effective TPS
- 19.97 TPS
- Time degraded
- 0.0 %
- Seconds degraded
- 0 s
- Ticks over budget
- 1.0 %
Raw data (9 fields)
World SaveHealthy19.80 TPS50.5 ms4.9s
World save speed and TPS impact during save operations.
Percentiles and stability
The regions of the test's zones. Each figure is the worst region's.
- Regions measured
- 1
- Seconds measured
- 3 s
- Median TPS
- 20.00 TPS
- Effective TPS
- 19.80 TPS
- Worst sustained MSPT
- 50.2 ms
- Time degraded
- 0.0 %
- Seconds degraded
- 0 s
- Ticks over budget
- 0.0 %
Raw data (5 fields)
Bone Meal GrowthHealthy19.88 TPS50.3 ms23.4s
Rapid crop and tree growth via bone meal.
- Heavy garbage collection4 collections, 106ms total
Percentiles and stability
The regions of the test's zones. Each figure is the worst region's.
- Regions measured
- 1
- Seconds measured
- 22 s
- Median TPS
- 20.00 TPS
- Effective TPS
- 19.88 TPS
- Worst sustained MSPT
- 77.8 ms
- Time degraded
- 0.0 %
- Seconds degraded
- 0 s
- Ticks over budget
- 9.9 %
Raw data (13 fields)
CPU
Single-Core (tick-aligned)Unstable19.86 TPS2.6 ms10.0s
Single-thread compute throughput measured on the main thread under the tick scheduler.
- The run did not hold steadyTPS P1=14.01<15, TPS min=14.01 (71% of avg)
Percentiles and stability
Raw data (8 fields)
Multi-Core CPUHealthy19.97 TPS0.5 ms24.6s
Parallel computation throughput across worker threads.
Percentiles and stability
Raw data (7 fields)
Memory & Storage
Memory PerformanceUnstable19.94 TPS0.6 ms85.8s
RAM bandwidth and latency.
JVM measurements — do not reflect actual bare-metal performance.
- Heavy garbage collection5 collections, 59ms total
- The run did not hold steadyTPS min=14.52 (73% of avg)
Percentiles and stability
Raw data (13 fields)
Disk PerformanceUnstable19.98 TPS0.7 ms214.7s
Sequential and random 4K read/write performance.
JVM measurements — do not reflect actual bare-metal performance.
- Heavy garbage collection17 collections, 320ms total
- The run did not hold steadyTPS min=14.68 (73% of avg)
Percentiles and stability
Raw data (10 fields)
Network
Network (NBT)Unstable11.62 TPS89.0 ms34.7s
Serialisation throughput — proxy for the rate at which the server can ship world data to clients.
- Heavy garbage collection49 collections, 469ms total
- The run did not hold steadyTPS avg=11.62<19, TPS P1=5.00<15, TPS StdDev=2.56>2, TPS min=5.00 (43% of avg)
Percentiles and stability
Raw data (11 fields)
›Machine and serverAMD EPYC 7763 64-Core Processor · Folia · 1.21.0
Score Breakdown
Detailed sub-category scores