Primate Labs has released Geekbench 7, the latest version of its popular cross-platform benchmarking tool. This update introduces new and improved workloads designed to better represent and measure the performance of the latest CPUs and GPUs across Android, iOS, Windows, macOS, and Linux. As with previous versions, Geekbench 7 is free for personal use. Enthusiasts and IT professionals can choose Geekbench 7 Pro, available at a 20% discount until August 6.
Revised Multi-Core Approach for Realistic Testing
One of the biggest changes is how the multi-core benchmark behaves. It now mirrors real-world application behavior instead of forcing every workload to use all available CPU cores. For example, the HTML5 browser test has been removed from the multi-core suite because web browsers are generally single- or lightly threaded. Geekbench 7 runs a benchmark in multi-threaded mode only if the real application it simulates uses multiple threads. According to Primate Labs, this provides a more realistic indication of overall system performance.
Sponsored Protocol
Updated Media Workloads
Geekbench 7 introduces a new set of media workloads to evaluate how modern CPUs handle everyday multimedia tasks. These include encoding AV1 video for screen sharing during video calls, compressing audio using the Opus codec for voice recorders and podcasts, and decoding audio and video while simultaneously generating live captions using OpenAI's Whisper speech recognition model. These workloads better represent activities such as video conferencing, content streaming, and media consumption. A new Game Physics benchmark based on the Jolt Physics engine used in modern games provides more accurate CPU performance measurement in physics-heavy gaming scenarios. The update also enhances the Photo Editor workload with more complex editing operations and expands the Photo Library test to support newer image formats like JPEG XL and DNG.
Sponsored Protocol
GPU Benchmarks with AI and CUDA
The GPU benchmark has been revamped to go beyond gaming, with a greater emphasis on machine learning and content creation. New tests measure real-time face tracking and filter effects for social media apps, AI-based image upscaling, and background blurring during video calls. Additionally, Geekbench 7 adds GPU workloads for RAW image processing, LUT-based video color grading, path tracing, and fluid simulation. Notably, it introduces support for Nvidia's CUDA API alongside OpenCL, Vulkan, and Metal, allowing users to benchmark Nvidia GPUs using the same API used in many professional and AI applications.
Larger and More Complex Datasets
To keep up with demanding modern workloads, the benchmark datasets have been updated to reflect larger and more complex files. The File Compression test now includes a broader mix of archives containing source code, object code, and text documents. The PDF Viewer workload processes a wider range of files, from park maps to technical documentation and academic papers. The update also introduces more assets and additional image formats across its developer and image processing workloads. For insights into performance debugging, check out the article on Git Bisect and Blame: Find the Commit That Introduced a Bug in 5 Minutes. For more details on Geekbench, visit Wikipedia.
Sponsored Protocol
