Skip to content

BENCHMARKS.md

Latest commit

 

History

History
134 lines (92 loc) · 7.51 KB

BENCHMARKS.md

File metadata and controls

134 lines (92 loc) · 7.51 KB
Raw

Community benchmark results

This file collects real-world AMD Core Performance Boost results from supported HP systems. It starts with the production notebook used to develop this project and is open to additional results through pull requests.

These are field observations, not controlled laboratory benchmarks. Compare enabled and disabled results from the same system wherever possible. Scores from different notebooks, Windows images, power modes, cooling conditions, and background workloads are not directly comparable.

How to contribute

  1. Run the benchmark with CPU-Z. Record the CPU-Z version if possible.
  2. Test both boost states under the same conditions. A battery comparison is optional.
  3. Add a new entry at the end of this file using the template below. Do not replace or edit another contributor's results.
  4. Include enough context to explain differences between systems.
  5. Submit the change as a pull request.

Please record:

  • Notebook model, processor, and BIOS version.
  • Windows version and whether the installation is vendor-provided, clean, or enterprise-managed.
  • Power source and Windows power mode.
  • AMD Core Performance Boost state.
  • CPU-Z version, test group, and thread count.
  • Single-thread and multi-thread scores.
  • Observed clock speed and temperature, when available.
  • Any background software or cooling conditions that could affect the result.

Entry template

## Manufacturer and model: Processor

### Test context

- Contributor:
- Test date:
- BIOS version:
- Windows environment:
- CPU-Z version:
- Power mode:
- Background workload:
- Cooling conditions:

### Idle observations

| Boost setting | Power source | Idle period | Temperature | Observed CPU speed |
| --- | --- | --- | --- | --- |
| Disabled | | | | |
| Enabled | | | | |

### CPU-Z benchmark

| Boost setting | Power source | Test group | Threads | Single-thread score | Multi-thread score | Observed speed | Maximum temperature |
| --- | --- | --- | ---: | ---: | ---: | --- | --- |
| | | | | | | | |

### Stress-test observations

Describe the test duration, live stress score, sustained clock, maximum temperature, fan behavior, and any thermal- or power-limit indicators.

### Personal observations

Describe how the setting affected real workloads, responsiveness, temperature, noise, battery behavior, or stability on this system.

HP ZBook 8 G1ak: AMD Ryzen AI 5 PRO 340

Test context

  • Test date: July 2026.
  • BIOS version: 01.05.04.
  • Windows environment: Windows 11 25H2 in an enterprise-managed environment, with the organization's normal security, endpoint-management, monitoring, and background applications.
  • CPU-Z version: Not recorded.
  • Power sources: AC power and battery.
  • Cooling conditions: Production office use; temperatures and clocks were observed on the notebook itself.

This is not a clean Windows or HP factory-image benchmark. The results are intended to compare boost enabled and disabled on the same managed notebook.

The processor contains three full Zen 5 cores and three compact Zen 5c cores. With simultaneous multithreading, each group provides six threads. The tables use CPU-Z's P-Core label for the Zen 5 group and E-Core for the Zen 5c group.

Idle observations on AC power

Measurements were taken after startup and 15 minutes without a deliberate workload. Clock speed was observed in Windows Task Manager.

Boost setting Temperature Observed CPU speed
Disabled Approximately 45–47 °C Capped at approximately 2.0 GHz
Enabled Approximately 61–65 °C Boosted up to approximately 3.7 GHz

CPU-Z benchmark on AC power

Boost setting Test group Threads Single-thread score Multi-thread score Observed loaded speed Maximum observed temperature
Disabled All cores 12 235.6 2133.0 Approximately 2.0 GHz 50–51 °C
Disabled P-Core / Zen 5 6 293.6 1207.6 Approximately 2.0 GHz Not recorded
Disabled E-Core / Zen 5c 6 295.0 1204.8 Approximately 2.0 GHz Not recorded
Enabled All cores 12 502.0 3724.1 Up to approximately 4.0 GHz 70 °C
Enabled P-Core / Zen 5 6 718.5 2826.2 Up to approximately 4.4 GHz 70 °C
Enabled E-Core / Zen 5c 6 514.6 2101.5 Not recorded Not recorded

Enabling boost improved the all-core multi-thread score by roughly 75% and the single-thread score by roughly 113%. The largest gains appeared in the P-Core / Zen 5 group.

CPU-Z benchmark on battery power

Core Performance Boost remained enabled for the battery test.

Test group Threads Single-thread score Multi-thread score
All cores 12 500.0 3381.2
P-Core / Zen 5 6 461.3 2226.6
E-Core / Zen 5c 6 384.9 1589.8

Compared with the enabled AC results, the all-core multi-thread score was about 9% lower. The targeted P-Core results were roughly 36% lower in single-thread and 21% lower in multi-thread performance. The E-Core results were approximately 25% lower. The nearly unchanged all-core single-thread score may reflect preserved short-burst performance or normal scheduling variation.

Continuous stress-test observations

Power source Workload Live stress score Observed sustained or peak speed Maximum temperature
AC All cores / 12 threads Not recorded Up to approximately 4.0 GHz 70 °C
AC P-Core / 6 threads Not recorded Up to approximately 4.4 GHz 70 °C
Battery All cores / 12 threads Approximately 3520 Approximately 3.0 GHz Not recorded
Battery 6 threads, core group not recorded Approximately 2195 Not recorded Not recorded

CPU-Z stress scores are live workload readings and should not be treated as directly equivalent to completed benchmark scores.

The 70 °C plateau does not by itself confirm thermal throttling. AMD specifies 100 °C Tjmax for this processor, while HP may apply a lower temperature target, power limit, or fan policy. Confirming the limiting factor would require sensor data showing a thermal- or power-limit flag.

Personal observations

Core Performance Boost provides a clear raw-performance advantage for sustained CPU-heavy work. On this notebook, that advantage also means higher temperatures, more cooling activity, and more noticeable fan noise.

For typical office work, including email, browsing, document editing, and video meetings, the benefit is less clear. Those applications can benefit from brief bursts of speed, but they are unlikely to use the full benchmark gain. For these users, a cooler and quieter notebook may matter more than maximum CPU throughput.

The notebook draws cooling air through its bottom intake. Paper, soft materials, or an uneven surface can restrict that airflow. A thermally induced shutdown has already been observed under real-world conditions when the intake was obstructed. Restricted airflow can occur during normal enterprise use and should be considered when evaluating the additional heat produced with boost enabled.

Based on this one system, I would enable boost for users who need sustained CPU performance and can keep the notebook on a firm, unobstructed surface. I would lean toward leaving it disabled for typical office users when lower temperatures, quieter operation, and predictable behavior are the priority. This is a personal conclusion, not a universal recommendation.