Sepals

An extremely radical and experimental optimization for Minecraft server performances.

Compatibility

Currently, sepals are compatible to almost all mods.

Here is verified mod list:

Target mod	Target mod version	Recommend target version	Sepals version	Recommend sepals version
sodium	all	all	all	latest
iris	all	all	all	latest
ferritecore	all	all	all	latest
krypton	all	all	all	latest
lithium	>=0.14.1	latest	>=1.0.4	latest
c2me-fabric	>=0.3.0+alpha.0.295	latest	>=1.0.5	latest
moonrise-opt	>=0.1.0-beta8	unknown	>=1.0.6	latest
async	>=0.1.5+alpha.8-1.21.4	unknown	>=1.0.11	latest

If you don't know which version to choose, please use the latest version of all the optimization mods, Sepals will ensure compatibility with them.

Specially

When Sepals work with mod Async testing on different platforms has different results, please use configurations to disable Sepals features or Async features to test which is more worthwhile if you want to use both together.

Sepals feature related: enableSepalsEntitiesCramming.

Config

Config name	Allowed value
forceEnableSepalsPoi	bool value
enableSepalsVillager	bool value
enableSepalsFrogLookAt	bool value
enableSepalsFrogAttackableSensor	bool value
enableSepalsLivingTargetCache	bool value
nearestLivingEntitiesSensorUseQuickSort	bool value
enableSepalsBiasedLongJumpTask	bool value
enableSepalsEntitiesCramming	bool value
enableSepalsItemMerge	bool value

Performance

A test was done in server of Mars provided by feimia, CPU: Intel i7-14700K; Game memory: 4G; OS: Ubuntu 24.04.1 LTS; Minecraft: 1.21.

Entities cramming

Use box cache to prevent too much 'getOtherEntities' calls

-- Notice --
This feature was ignored scoreboard/team predicates for every sigle entity
may cause unexpected behaviors

This problem will not affects anything in vanilla(survival and creative)
it only things to consider when game servers that using command blocks    

-- Status --
Default enabled

1390 villagers cramming in a 7x7 space:

Environment	tickCramming	Percent(Avg.)
Vanilla	53.6 ms	100 %
With Lithium	54.4 ms	101 %
With Sepals	10.2 ms	19 %
With Sepals+Lithium	8.5 ms	15 %

Weighted random

Use binary search to replaced vanilla weight random

-- Warning --
This feature may not provided good optimization
because it was proved by spark that slower than vanilla when constructing range table
even if the binary search almost close to 0ms

-- Status --
Default disabled

-- Warining -- 
Not tested.

Biased long jump task

Use sepals long jump task impletation to replaced vanilla impletation

As mentioned above, the binary search is almost no costs
sepals impletation will construct the range table at the same time as generating targets
and used Catheter to replaced java stream

The same time, rearrange should jump conditions in frog brain, make a good performance

-- Notice --
This feature is unable to change in game runtime
required restart the server to apply changes

-- Status --
Default enabled

800 frogs cramming in a 7x7 space:

Environment	keepRunning	Percent(Avg.)
Vanilla (LongJumpTask)	43.1 ms	100 %
With Lithium (LongJumpTask)	7.5 ms	17 %
With Sepals (SepalsLongJumpTask)	0.2 ms	0.4 %
With Sepals+Lithium (SepalsLongJumpTask)	0.05 ms	0.1 %

Environment	getTarget	Percent(Avg.)	Percent(in keepRunning)
Vanilla (LongJumpTask)	43.1 ms	100 %	100 %
With Lithium (LongJumpTask)	3.6 ms	9 %	48 %
With Sepals (SepalsLongJumpTask)	N/A ms	0 %	0 %
With Sepals+Lithium (SepalsLongJumpTask)	N/A ms	0 %	0 %

Quick sort in NearestLivingEntitiesSensor

Quick sort supports by fastutil to replace java tim sort

-- Status --
Default enabled

800 frogs cramming in a 7x7 space:

Environment	sort (NearestLivingEntitiesSensor#sense)	Percent(Avg.)
Vanilla	3.8 ms	100 %
With Lithium	3.6 ms	94 %
With Sepals	2.2 ms	57 %
With Sepals+Lithium	2.2 ms	57 %

Frog attackable target filter

Rearrange the attackable conditions
let less costs predicate running first
reduce the probability of high-costs calculating

-- The complaints --
Mojang's attackable predicate is:

!entity.getBrain().hasMemoryModule(MemoryModuleType.HAS_HUNTING_COOLDOWN)
 && Sensor.testAttackableTargetPredicate(entity, target)
 && FrogEntity.isValidFrogFood(target)
 && !this.isTargetUnreachable(entity, target)
 && target.isInRange(entity, 10.0)

in this case, 'Sensor#testAttackableTargetPredicate' has calls 'TargetPredicate#test'
that cause a very lots raycast calculate when entities too much in the area
but... minecraft's raycast is absolutely bad, very slow

the 'TargetPredicate#test' in this case (800 frogs) has make 9.8ms costs in once game tick
among them, 'BlockView.raycast' contributed 7.3ms

then i make it be:

FrogEntity.isValidFrogFood(target) &&
 entity.getBrain().hasMemoryModule(MemoryModuleType.HAS_HUNTING_COOLDOWN) && 
 target.isInRange(entity, 10.0) && 
 Sensor.testAttackableTargetPredicate(entity, target) && 
 isTargetUnreachable(entity, target);
 
the 'isValidFrogFood' is simple conditions, check the entity's tag has in 'frog_food'
and a extra check when entity is slime then skip it when it size not 1

the 'isInRange' and 'hasMemoryModule' also simple, it only a few math calculates

-- Status --
Default enabled

800 frogs cramming in a 7x7 space:

Environment	time	Percent(Avg.)
Vanilla (FrogAttackablesSensor#matches)	10 ms	100 %
With Lithium (FrogAttackablesSensor#matches)	5.7 ms	57 %
With Sepals (SepalsFrogBrain#attackable)	0.1 ms	1 %
With Sepals+Lithium (SepalsFrogBrain#attackable)	0.1 ms	1 %

Frog look-at target filter

Use 'SepalsLivingTargetCache' to improves target search performance
Required enable the SepalsLivingTargetCache to apply this feature, otherwise not difference to vanilla

-- Notice --
The raycast is in TargetPredicate test
at the 'findFirst' in LivingTargetCache when input predicate is success

but if subsequent conditions is failures, then to calculate this anymore is ueless
because even if the findFirst has found (raycast success)
but we don't used this result in subsequent contexts  

-- Status --
Default enabled

800 frogs cramming in a 7x7 space:

Environment	findFirst	Percent
Vanilla (LookAtMobWithIntervalTask$$Lambda#findFirst)	2.7 ms	100 %
With Lithium (LookAtMobWithIntervalTask$$Lambda#findFirst)	2.5 ms	92 %
With Sepals (SepalsLookAtMobWithIntervalTask$$Lambda#findFirstPlayer)	0.1 ms	3 %
With Sepals+Lithium (SepalsLookAtMobWithIntervalTask$$Lambda#findFirstPlayer)	0.1 ms	3 %

Villager miscellaneous optimizations

Listed what sepals to do here:

1. Used Catheter to replaced java stream, The newest version Catheter has more quite stunning performance and scalability than Stream 
2. Cached tasks, activities, running tasks and memories to improves starting and updating task
3. Use sepals composite task to replaced vanilla composite task
4. Whenever possible, find opportunities to skips more raycast and useless predicates
5. Use 'SepalsLivingTargetCache' to replaced vanilla cache, At the cost in sensors tick make less cost in finding interaction target or look at mob task
6. Rearranged predicates and extra lower cost predicate, The purpose of this is do higher cost predicate later or best don't do that, Skip the remaining high cost predicates in advance
7. Copied and modified 'SerializingRegionBasedStorage' optimizations from lithium
8. With more targeted task, don't use the generics to reduce useless operations
9. Used binary search list to replaced hashset search

-- Notice --
Recommend use this feature with lithium and c2me to make best performance

-- Warning --
Not long-term stability tested, only a month running shown it's ok currently
this feature does not be proved 100% vanilla
also it does not be proved has not vanilla in statistical significance 

-- Status --
Default enabled

800 villagers cramming in a 7x7 space at noon:

Environment	Brain#tick (Total)	Percent	Brain#startTasks	Percent(startTasks)	Brain#tickSensors	Percent(tickSensors)	Brain#updateTasks	Percent(updateTasks)	Brain#tickMemories	Percent(tickMemories)
Vanilla	18 ms	100 %	9.3 ms	100 %	5.2 ms	100 %	3 ms	100 %	0.5 ms	100 %
With lithium	12.4 ms	68 %	4.8 ms	51 %	5.9 ms	113 %	1.2 ms	40 %	0.5 ms	100 %
With Sepals	9.7 ms	53 %	3.6 ms	38 %	3.7 ms	71 %	2 ms	66 %	0.4 ms	80 %
With Sepals and lithium	10 ms	55 %	3.4 ms	36 %	3.7 ms	71 %	2.5 ms	83 %	0.4 ms	80 %

800 villagers cramming in a 7x7 space at night:

Environment	Brain#tick (Total)	Percent	Brain#startTasks	Percent(startTasks)	Brain#tickSensors	Percent(tickSensors)	Brain#updateTasks	Percent(updateTasks)	Brain#tickMemories	Percent(tickMemories)
Vanilla	16.7 ms	100 %	8.2 ms	100 %	6 ms	100 %	2 ms	100 %	0.5 ms	100 %
With lithium	10.2 ms	61 %	3.2 ms	24 %	6 ms	113 %	0.5 ms	25 %	0.5 ms	100 %
With Sepals	9 ms	53 %	3.3 ms	16 %	4.7 ms	78 %	0.7 ms	35 %	0.3 ms	60 %
With Sepals and lithium	8.7 ms	52 %	2.9 ms	11 %	4.6 ms	76 %	0.7 ms	35 %	0.5 ms	100 %