[Perf] Windows/x64: 3 Improvements on 8/28/2026 9:49:00 AM +00:00

### Run Information

Name | Value
-- | --
Architecture | x64
OS | Windows 10.0.22631
Queue | ViperWindows
Baseline | [bf3289d05dff3962cdccb0e81b5315990e5b1198](https://github.com/dotnet/sdk/commit/bf3289d05dff3962cdccb0e81b5315990e5b1198)
Compare | [100002779ceec649fcfd00c233f673b494130827](https://github.com/dotnet/sdk/commit/100002779ceec649fcfd00c233f673b494130827)
Diff | [Diff](https://github.com/dotnet/sdk/compare/bf3289d05dff3962cdccb0e81b5315990e5b1198...100002779ceec649fcfd00c233f673b494130827)
Configs | CompilationMode:tiered, RunKind:bepuphysics

### Improvements in DemoBenchmarks.TwoBodyConstraintBenchmarks

Benchmark | Baseline | Test | Test/Base | Test Quality | Edge Detector | Baseline IR | Compare IR | IR Ratio
-- | -- | -- | -- | -- | -- | -- | -- | --
|

• [AngularHinge - Duration of single invocation]()
• 📝 - [Benchmark Source]()
• [ADX - Test Multi Config Graph]()

| 116.24 μs | 100.51 μs | 0.86 | 0.00 | True | | |
|

• [Contact4Nonconvex - Duration of single invocation]()
• 📝 - [Benchmark Source]()
• [ADX - Test Multi Config Graph]()

| 551.82 μs | 485.23 μs | 0.88 | 0.00 | True | | |

![graph]()
![graph]()
[Test Report]()

### Repro
General Docs link: https://github.com/dotnet/performance/blob/main/docs/benchmarking-workflow-dotnet-runtime.md

```cmd
git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'DemoBenchmarks.TwoBodyConstraintBenchmarks*'
```

### DemoBenchmarks.TwoBodyConstraintBenchmarks.AngularHinge

#### ETL Files

#### Histogram

#### JIT Disasms

### DemoBenchmarks.TwoBodyConstraintBenchmarks.Contact4Nonconvex

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

[Profiling workflow for dotnet/runtime repository](https://github.com/dotnet/performance/blob/master/docs/profiling-workflow-dotnet-runtime.md)
[Benchmarking workflow for dotnet/runtime repository](https://github.com/dotnet/performance/blob/master/docs/benchmarking-workflow-dotnet-runtime.md)

---

### Run Information

Name | Value
-- | --
Architecture | x64
OS | Windows 10.0.22631
Queue | ViperWindows
Baseline | [bf3289d05dff3962cdccb0e81b5315990e5b1198](https://github.com/dotnet/sdk/commit/bf3289d05dff3962cdccb0e81b5315990e5b1198)
Compare | [100002779ceec649fcfd00c233f673b494130827](https://github.com/dotnet/sdk/commit/100002779ceec649fcfd00c233f673b494130827)
Diff | [Diff](https://github.com/dotnet/sdk/compare/bf3289d05dff3962cdccb0e81b5315990e5b1198...100002779ceec649fcfd00c233f673b494130827)
Configs | CompilationMode:tiered, RunKind:bepuphysics

### Improvements in DemoBenchmarks.SweepBenchmarks

Benchmark | Baseline | Test | Test/Base | Test Quality | Edge Detector | Baseline IR | Compare IR | IR Ratio
-- | -- | -- | -- | -- | -- | -- | -- | --
|

• [BoxCompound - Duration of single invocation]()
• 📝 - [Benchmark Source]()
• [ADX - Test Multi Config Graph]()

| 44.66 μs | 37.86 μs | 0.85 | 0.00 | True | | |

![graph]()
[Test Report]()

### Repro
General Docs link: https://github.com/dotnet/performance/blob/main/docs/benchmarking-workflow-dotnet-runtime.md

```cmd
git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'DemoBenchmarks.SweepBenchmarks*'
```

### DemoBenchmarks.SweepBenchmarks.BoxCompound

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

[Profiling workflow for dotnet/runtime repository](https://github.com/dotnet/performance/blob/master/docs/profiling-workflow-dotnet-runtime.md)
[Benchmarking workflow for dotnet/runtime repository](https://github.com/dotnet/performance/blob/master/docs/benchmarking-workflow-dotnet-runtime.md)

Start with src/benchmarks/real-world/bepuphysics2/TwoBodyConstraintBenchmarks.cs and SweepBenchmarks.cs, then run the provided benchmarks_ci.py command with the listed filters. Compare the Windows/x64 results and linked test reports to understand the reported improvements. Done should mean the performance change is explained and any follow-up is validated against the affected benchmarks.

Written by the indexing model from the issue text.