On Tuesday, Primate Labs unveiled Geekbench 6, the latest version of its popular benchmark tool. Used to benchmark CPU and GPU performance in Macs, iPhones, iPads, and non-Apple computing devices, Geekbench 6’s benchmark suite offers updates that better reflect the workloads that current devices can handle.
“A lot has changed in the tech world over the past three years. Smartphone cameras take bigger and better photos. Artificial intelligence, especially machine learning, has become ubiquitous in general and mobile applications,” wrote John Poole in a blog post. “The number of cores in computers and mobile devices continues to grow. And the way we interact with our computers and mobile devices has changed dramatically – who would have guessed that video conferencing would suddenly increase in 2020? »
To better reflect modern workflows, Geekbench 6 implements new tasks, such as background blurring in video conferences, image filtering and adjustment for social media, automatic removal of unwanted objects from photos , detecting and tagging objects in photos using machine learning models, and using scripting languages to analyze, process, and convert text. Datasets have also been updated with larger files and higher resolution images.
As Geekbench explains, the new CPU benchmark scores “are used to assess and optimize CPU and memory performance using workloads including data compression, image processing, machine learning and code compilation”.
The new tests are similar in nature to Geekbench 5, but perform newer tasks with new datasets and libraries to better represent modern applications. You cannot directly compare Geekbench 5 and Geekbench 6 scores. Multi-core performance, in particular, uses new methods of sharing work between cores that should better represent the performance of processors with “high performance” and “high efficiency” cores.
igamesnews uses Geekbench in our iPhone, iPad, and Mac review, and we’ll be using Geekbench 6. However, previous results recorded with Geekbench 5 cannot be compared to Geekbench 6, as the two use different test suites. In the short term, we may include Geekbench 5 results to provide historical context for older devices that we cannot test with Geekbench 6. It’s hard to compare scores without doing extensive testing, but testing the MacBook 14-inch Pro with an 8-core CPU and 14-core GPU returned scores of 2138 (single-core) and 8475 (multi-core) vs an average of 1750 (single) and 9550 (multi-core) using Geekbench 5. We have also tested the 14-inch MacBook Pro with a 10-core M1 Pro and a 16-core GPU and got Geekbench 6 scores of 2399 (single) and 12289 (multi), vs Geekench 5 scores of 1778 (single) and 12544 (multi ).
Geekbench 6 is free for personal use and can be downloaded from the Geekbench site (Mac) or the App Store (iOS). The Pro version allows users to run the application from a portable storage device or network drive, save results offline and automate testing, and provides a license for commercial use . Geekbench 6 Pro is $99, but Primate Labs is offering 20% off ($79) until February 28.
What do Geekbench scores mean?
What do these Geekbench scores mean? What does it measure and why should you care?
Geekbench runs a set of specific repeatable tests that stress your CPU or graphics processing unit (GPU) in a specific way, and turn it into a numerical score. The tests and the data they use are the same on all platforms, but some platforms may have hardware that can speed up some parts of it (like file compression). Higher is better, but it doesn’t represent a concrete value like “elapsed time” or “pixels per second”. It is a whole synthetic mark and pointing.
CPU tests
The CPU test includes things like file compression (compressing and decompressing a 75MB archive with 9,841 files in a variety of methods), generating directions between a sequence of locations, rendering 8 popular web pages, rendering PDFs , run developer scripts and machine-learn tasks such as object detection and background blurring.
The CPU test has two result scores: single-core, which measures the performance of running tasks on a single CPU core, and multi-core, which uses all CPU cores. Both are important – many applications have their performance limited by a single main thread, so single-core processor performance will determine how fast they run. But for well-multithreaded applications or for running multiple applications at once, multi-core is a good measure of total maximum CPU performance.
GPU Compute Tests
You might think that measuring graphics processing unit (GPU) performance comes down to knowing what frame rate you can expect in 3D games. When it comes to Geekbench, that’s not necessarily the case. Geekbench tests do not measure 3D graphics rendering performance. Instead, it measures performance on “GPU Compute” tasks – using the GPU to handle computational tasks such as detecting edges in images, applying Gaussian blur to a 24-megapixel photo , a particle physics simulation or the detection of faces in photos.
It’s entirely possible for one product to run 3D games faster than the other while scoring lower on the Geekbench GPU Compute tests. While GPU compute and 3D graphics performance are often linked, it’s fair to say that if you want to measure 3D graphics performance, you should be running a 3D graphics benchmark, not a GPU Compute benchmark like Geekbench.
When you run the Geekbench 6 GPU Compute test, you can choose which API (Application Programming Interface – the way developers talk to hardware) you want to use. On a Mac, you can choose between OpenCL and Metal. On iPhone and iPad, the only supported API is Metal. Your choice does not change the tests in any way, only how the application communicates with the hardware. On Mac, Metal is probably the highest score and should probably be compared to Vulkan’s performance on Windows and Android PCs.
If you’re technically inclined, you can read a PDF white paper that describes the tests in detail. There is a whitepaper for Compute CPU testing and GPU testing.
