{
  "id": 18150967,
  "title": "This overlooked metric can reveal CPU bottlenecks that GPU usage alone can't",
  "url": "https://urgent.news/2026/11/04/this-overlooked-metric-can-reveal-cpu-bottlenecks-that-gpu-usage",
  "topic": "tech",
  "section": "Tech",
  "published": "2026-11-04T17:01:23.000Z",
  "source": {
    "name": "XDA Developers",
    "slug": "xda-developers",
    "url": "https://www.xda-developers.com/overlooked-metric-can-reveal-cpu-bottlenecks-that-gpu-usage-alone-cant/"
  },
  "original_language": "en",
  "account": "For years, many have relied on GPU usage as the primary indicator to determine if their CPU is limiting their GPU performance. However, this single metric can be misleading. While GPU usage may consistently sit around 70–80% when the GPU is waiting on the CPU, it doesn't provide the full picture. Factors such as FPS caps, V-Sync, game engine limitations, and RAM bottlenecks can also reduce GPU usage, making it an unreliable sole measure of CPU bottlenecks.\n\nIntel PresentMon introduced a more accurate metric called GPU Busy, which reveals exactly how many milliseconds your GPU actively spends working on each frame. By comparing GPU Busy to your frame time, you can gain a more precise understanding of whether your CPU or GPU is the true bottleneck. GPU Busy is measured in milliseconds and provides a frame-by-frame measurement, unlike GPU usage, which is typically measured over sampling intervals.\n\nTo determine if your CPU is the bottleneck, compare the GPU Busy value to your frame time. If GPU Busy hovers around the frame time (e.g., 6ms for a 150FPS average), your GPU is likely GPU-bound. Conversely, if GPU Busy is significantly lower than your frame time (e.g., 4ms when the average frame time is 6.67ms), this gap may indicate a CPU bottleneck. However, this is not a definitive conclusion, as other factors like FPS caps, V-Sync, slower RAM, and game engine limitations can also influence frame times.\n\nGPU Busy helps narrow down the possibilities by offering a more granular view of GPU processing time. While it cannot definitively pinpoint a CPU bottleneck, it is a valuable tool for further investigation. It is advisable to first monitor GPU usage to identify obvious GPU-related issues. If GPU usage appears normal, then use Intel PresentMon to monitor GPU Busy and compare it to frame time. This approach allows you to rule out GPU bottlenecks and focus on potential CPU limitations. Remember, multiple factors can influence frame times, and no single metric can provide a definitive answer. It's essential to interpret the data carefully and consider all possible causes before making assumptions about your system's performance.",
  "summary": "You get frame-by-frame insights with this one.",
  "key_points": [
    "Intel PresentMon introduced GPU Busy metric to measure GPU active time per frame.",
    "GPU Busy compared to frame time reveals CPU or GPU bottlenecks.",
    "GPU Busy helps narrow down performance issues beyond GPU usage alone."
  ],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}