---
title: Performance
description: Allocation behavior, benchmark results, and the limits of what they show.
sidebar:
  order: 6
---

The measured fixed-capacity traversal paths allocate zero managed bytes, including the four-worker paths. The BenchmarkDotNet comparison places generated `[Each]` near a handwritten baseline at 100,000 entities, while the named query costs about twice as much.

:::note
This is an initial baseline with visible timing variation. It is not a numeric regression policy or a claim of general superiority over another ECS.
:::

## Dense movement at 100,000 entities

All four cases use the same `Position` and `Velocity` structs, initial values, and three scalar updates, with untracked components. The handwritten baseline has no scheduler. Archevolt's numbers include schedule dispatch, resource access, lease checks, and query traversal.

| Case | Mean (µs) | Mean (ns/entity) | Managed B/op |
| --- | ---: | ---: | ---: |
| Handwritten | 104.279 | 1.043 | 0 |
| Generated `[Each]` | 108.866 | 1.089 | 0 |
| Generated named query | 210.411 | 2.104 | 0 |
| Arch 2.1.0 inline query | 146.054 | 1.461 | 0 |

Measured on an AMD Ryzen 9 5950X, Windows 11, .NET 10.0.13, Release, BenchmarkDotNet 0.15.8. Arch owns its own managed chunk storage, so the two engines do not offer identical alignment, scheduling, or access guarantees.

## Things that cost

- Structural churn allocates.
- Small workloads can cost more to dispatch in parallel than to run serially.
- Writable access stamps a change on [tracked components](/guides/change-tracking) even when the value is equal.

## Reproduce

```powershell
dotnet run --project benchmarks/Archevolt.Benchmarks -c Release -- --quick
dotnet run --project benchmarks/Archevolt.Benchmarks -c Release -- --filter "*MovementBenchmarks*" --job short
./scripts/measure.ps1 -Mode All
```

`measure.ps1` runs the matrix, BenchmarkDotNet comparison, and JIT disassembly (`-Mode Matrix`, `Bdn`, or `Disassembly` run them separately). It fails if any command fails or BenchmarkDotNet does not execute all four comparison cases.

The full [performance report](https://github.com/Stephanvs/archevolt/blob/main/.audit/performance-final.md) records the workload matrix, handwritten and Arch comparisons, allocations, native storage, generated code sizes, and JIT and AOT inspection.