Earlier this year, GeekBench, a well-known benchmark testing platform, released version 6.0, which significantly adjusted the scoring algorithm.
Currently, GeekBench has released version 6.1 update, including Clang 16 parameters on all platforms, to more accurately reflect the multi-core performance of the device.
This change is reflected in the results of the running score, that is, GeekBench version 6.1 compared to the previous 6.0 version, single-core score increased by 5%, multi-core score increased by 10%.
Therefore, GeekBench officially advises users not to compare the score gap between the two versions of the device, because of the difference in scoring parameters, this is not practical.
In addition to the difference in run scores due to the introduction of Clang 16 parameters, GeekBench version 6.1 includes the following updates:
Geekbench 6.1 extends the workload gap (pause between workloads) from two seconds to five seconds, minimizing the difference in results due to heat downshift on smartphones such as the Galaxy S23 series.
– Support for scalable vector extension instructions in multiple image processing and machine learning functions, as well as support for AVX512-FP16 instructions in various image processing tasks.
For systems without FP16 (floating point 16-bit) instructions, Geekbench 6.1 will utilize a fixed-point algorithm to consider multiple image processing functions.
Geekbench 6.1 for 12 – and 16-core AMD Ryzens, AMD Threadrippers, and Intel Xeons processors enhances multi-core performance for background blur and horizon detection workloads.
Specific updates and more relevant information can be viewed on the official website page.