GPUCHECK Scraper

An Apify actor for scraping hardware performance data (GPU/CPU combinations, games, and FPS) from GPUCHECK. Built on the Apify SDK, it can be run on the Apify platform or locally.

Overview

This actor extracts performance benchmarks from GPUCHECK.com. By default, it only scrapes new results not seen in previous runs to avoid duplicate data. Results are saved to a dataset, with each item detailing a specific game's FPS for a given GPU, CPU, quality setting, and resolution.

Key Features

• Incremental Scraping: Saves the state of the last crawl. Subsequent runs only extract new data, saving time and compute units.
• Filtered Crawling: Use regex patterns to include or exclude specific GPU and CPU models.
• Proxy Support: Requires a proxy; uses Apify Proxy by default.
• Local & Cloud Execution: Can be run on the Apify platform or in your local environment.

How to Use

1. Configure Input: Set your desired filters (gpuIncludeRegex, cpuIncludeRegex, etc.) in the actor's input.
2. Run the Actor: Start the run on Apify or via CLI.
3. Access Output: Results are available in the run's dataset.
4. Manage State: To perform a full scrape (ignoring previous runs), delete the "gpucheck" key in Storage > Key-value stores.

Input

Configure the scrape using the following input fields:

Output

The output is stored as dataset items. Each item represents a single game's FPS result for a specific hardware and settings combination.

[
  {
    "url": "https://www.gpucheck.com/gpu/amd-radeon-rx-460/amd-ryzen-3-2200g/high/?lang=en&currency=usd",
    "gpu": "AMD Radeon RX 460 - $ 140",
    "cpu": "AMD Ryzen 3 2200G - $ 98",
    "quality": "high",
    "resolution": "1920x1080",
    "game": "Assassin's Creed Valhalla",
    "fps": "26.6"
  }
]

Compute Units Consumption

For efficiency, run longer scrapes (at least one minute) rather than multiple short ones due to startup overhead. On average, the actor consumes 1 Compute Unit for every 200 results scraped.

For feedback, contact dtrungtin@gmail.com.