Skip to main content

Cross-platform terminal-native hardware diagnostics and safe performance profiling

Project description

SYSTEM-PERF™

Measure clearly. Stress safely. Know what your hardware can handle.

SYSTEM-PERF™ is a cross-platform terminal-native hardware diagnostic, bounded performance test, and game-readiness CLI for Windows, Linux, and macOS.

Created by Anika Mukherjeecuteypieanika@gmail.com

What v0.2.0 does

  • Deep hardware inventory: system model, motherboard, BIOS/firmware, CPU topology/clocks/cache, RAM modules/speeds, graphics adapters/VRAM/drivers, and storage devices.
  • Safe multi-process CPU and RAM tests with cancellation and thermal limits where CPU sensors are available.
  • NVIDIA live telemetry for temperature, power, utilization, memory use, and clocks.
  • Portable temporary-file storage read/write check that cleans up after itself.
  • Named-game 1080p readiness estimates using CPU results, installed RAM, GPU class, and VRAM.
  • Calm card-based terminal design, plus stable JSON/JSONL and Markdown output.
  • Versioned, integrity-stamped result bundles and baseline comparisons.
  • Safe per-user report locations, even when launched from a protected directory.

Install

python -m pip install system_perf-0.2.0-py3-none-any.whl

On Windows, use py -m pip if python is not on PATH.

Start here

system-perf doctor
system-perf detect
system-perf run full

The full profile prints measured CPU, RAM, storage, and GPU telemetry results, followed by game-readiness estimates. The saved result path is printed at the end.

Check one game during the run:

system-perf run full --game "Cyberpunk 2077"

Or evaluate an existing result:

system-perf games
system-perf games --result path/to/result.json
system-perf games --result path/to/result.json --game forza

Focused tests

system-perf test cpu --duration 5 --workers 4
system-perf test memory --size 128 --rounds 8
system-perf test gpu
system-perf test storage --size 64
system-perf monitor --format jsonl --count 5

The GPU command is currently a capability and telemetry probe, not a rendered GPU benchmark. It identifies the adapters, dedicated graphics memory, drivers, current NVIDIA sensor values, and a broad model class used by the game estimator.

Supported game models

  • Minecraft Java
  • Valorant
  • Counter-Strike 2
  • Fortnite
  • Grand Theft Auto V
  • Forza Horizon 5
  • Red Dead Redemption 2
  • Elden Ring
  • Cyberpunk 2077
  • Hogwarts Legacy
  • Starfield
  • Microsoft Flight Simulator

Game output uses four states:

  • GREAT — the hardware meets the model's recommended CPU, RAM, GPU, and VRAM tiers.
  • PLAYABLE — it meets the modeled minimum, usually with lower settings.
  • LIMITED — at least one component is below the modeled minimum.
  • UNKNOWN — there is not enough GPU evidence for a responsible estimate.

These are explainable hardware-class estimates, not guaranteed FPS. Resolution, presets, ray tracing, upscaling, drivers, cooling, background applications, mods, and game updates can materially change real performance.

Reports

Default result folders:

Platform Location
Windows %LOCALAPPDATA%\SYSTEM-PERF\reports
macOS ~/Library/Application Support/SYSTEM-PERF/reports
Linux $XDG_STATE_HOME/system-perf/reports or ~/.local/state/system-perf/reports

Choose a path explicitly with --output:

system-perf run full --output result.json
system-perf report result.json
system-perf report result.json --format markdown --output report.md
system-perf compare before.json after.json

Platform capabilities

Capability Linux Windows macOS
CPU/RAM/system inventory Yes Yes Yes
Motherboard/firmware inventory sysfs where exposed CIM Apple system profile
CPU and RAM workloads Yes Yes Yes
Storage inventory and portable I/O test Yes Yes Yes
Linux CPU temperature Yes, when sysfs exposes it
NVIDIA telemetry When nvidia-smi is available When nvidia-smi is available Legacy NVIDIA systems only
Named-game estimates Yes Yes Yes, when GPU model is recognized
JSON/JSONL/Markdown Yes Yes Yes

Missing sensors or inventory sources are shown as unavailable and do not crash the run.

Safety and privacy

Workloads are bounded and can be cancelled with Ctrl+C. Where a supported CPU temperature sensor exists, --temperature-limit stops the CPU workload cleanly. The inventory intentionally excludes hardware serial numbers and hostnames from shared reports.

SYSTEM-PERF cannot guarantee hardware safety. Use conservative limits and adequate cooling.

License and branding

Copyright © 2026 Anika Mukherjee.

The software is licensed under the GNU Affero General Public License, version 3 or any later version (AGPL-3.0-or-later). See LICENSE.

SYSTEM-PERF™ and the original project branding are claimed as trademarks of Anika Mukherjee. See NOTICE.md. is a trademark claim, not a representation of formal registration.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distributions

No source distribution files available for this release.See tutorial on generating distribution archives.

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

system_perf-0.2.0-py3-none-any.whl (49.6 kB view details)

Uploaded Python 3

File details

Details for the file system_perf-0.2.0-py3-none-any.whl.

File metadata

  • Download URL: system_perf-0.2.0-py3-none-any.whl
  • Upload date:
  • Size: 49.6 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.10.11

File hashes

Hashes for system_perf-0.2.0-py3-none-any.whl
Algorithm Hash digest
SHA256 382530e17f6dabaeffb79d8bcd4afc95ba91d36eee7535f38e4fac41cf8a42e0
MD5 80be873ef67134b796a4217dd393a5db
BLAKE2b-256 d913e3eb6f1e104c44e5bd99e60417660d17b6f3d9141f5d2e1eaa6e0a9852d1

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page