Interpreting Geekbench 4 Scores
With Geekbench 4, higher scores are better, with double the score indicating double the performance. Geekbench 4 CPU and Compute scores are calibrated using a Microsoft Surface Book with an Intel Core i7-6600U processor as a baseline with a score of 4,000 points. Geekbench 4 battery scores are not calibrated against a specific system.
Geekbench 4 uses a number of different tests, or workloads, to measure CPU performance. The workloads are divided into four different subsections:
Crypto Crypto workloads measure the crypto instruction performance of your computer by performing cryptography tasks that make heavy use of crypto instructions. While not all software uses crypto instructions, the software that does can benefit enormously from it.
Integer Integer workloads measure the integer instruction performance of your computer by performing processor-intensive tasks that make heavy use of integer instructions. All software makes heavy use of integer instructions, meaning a high integer score indicates good overall performance.
Floating Point Floating point workloads measure floating point performance by performing a variety of processor-intensive tasks that make heavy use of floating-point operations. While almost all software makes use of floating point instructions, floating point performance is especially important in video games, digital content creation, and high-performance computing applications.
Memory Memory workloads measure memory latency and bandwidth. Software working with large data structures (e.g., digital content creation) or with referential data structures (e.g., databases, web browsers) rely on good memory performance to keep the processor busy.
A complete description of the individual Geekbench 4 CPU workloads can be found on the Geekbench website.
Geekbench 4 uses several workloads to measure Compute performance.
Each Compute workload has an implementation for each supported Compute API. Geekbench currently supports the following Compute APIs: OpenCL, CUDA, Metal, and RenderScript. While it is possible to compare scores across APIs (e.g., a OpenCL score with a Metal score) it is important to keep in mind that due to the nature of Compute APIs the performance difference can be due to more than differences in the underlying hardware (e.g., the GPU driver can have a huge impact on performance).
Each workload's performance is compared against a baseline to determine a score. These scores are averaged together to determine an overall, or Geekbench, score for the system.
Geekbench 4 uses a Microsoft Surface Book with an Intel Core i7-6600U processor as the baseline with a score of 4,000 points.
Geekbench 4 provides three different kinds of scores:
Workload Scores Each time a workload is executed Geekbench calculates a score based on the computer's performance compared to the baseline performance.
Subsection Scores A subsection score is the geometric mean of all the workload scores for workloads that are part of the subsection. These scores are useful for determining the performance of the computer in a particular area. See the subsection descriptions above for a summary on what each subsection measures.
Geekbench Score The Geekbench score is the weighted arithmetic mean of the four subsection scores. The Geekbench score provides a way to quickly compare performance across different computers and different platforms without getting bogged down in details
Geekbench 4 battery scores measure the battery life of a device when running processor-intensive applications. It provides a consistent workload to the device, and generates a Geekbench score by evaluating the amount of work that it is able to do while the battery is discharging and the amount of time it takes for the battery to discharge.
The battery benchmark can be run in Full Discharge mode, from 100% to 0% battery level, to provide a high level of confidence in a device's battery performance. The battery benchmark can also be run in Partial Discharge mode, for 3 hours. If the battery benchmark is stopped after 10 minutes, by the user or by the battery reaching 0%, then the result will be saved and can be uploaded. Battery benchmark scores gathered by any method except the Full Discharge mode provide a medium level of confidence in a device's battery performance, and longer tests are more reliable.
When comparing scores, remember that higher scores are better, and double the score indicates double the performance.