Geekbench 7 is Here, Introducing Larger Datasets and Brand-New Audio-Video Benchmarks
3 day ago / Read about 0 minute
Author:小编   

With the rapid development of hardware architectures and software workloads, Primate Labs has officially released Geekbench 7, a new-generation benchmarking tool. As the latest version in the series, Geekbench 7 maintains the same appearance and operation as its predecessors but features significant innovations in its underlying architecture, aiming to provide a more comprehensive and accurate assessment of the true computing performance of computers and smartphones. This update redesigns the CPU multi-core and GPU benchmarks to better align with real-world usage scenarios. The multi-core test now activates multi-threading only when real applications require it, avoiding the indiscriminate use of all cores. The GPU test significantly increases the weight of machine learning and content creation scenarios, adding tasks such as face tracking, real-time video filters, background blurring, and support for NVIDIA's CUDA API. Additionally, Geekbench 7 introduces several new media workload tests, such as AV1 encoding for screen-shared videos and Opus compression for music, to measure the CPU's capabilities in audio-visual processing. To celebrate the release, Primate Labs is offering a 20% discount on Geekbench 7 Pro until August 6, priced at $79, while the standard version remains free.