Geekbench says Intel is artificially boosting CPU benchmark tests


The latest Intel CPUs may be impressive, but not as impressive as some benchmarks suggest. Intel’s new tool breaks down Geekbench 6 results and gives the company’s latest CPUs high scores in some tests.

Primate Labs, the company behind the popular open-source CPU and GPU benchmarking program Geekbench, released a blog post today explaining this. Intel’s new Binary Optimization Tool Intel generates incorrect scores for CPUs. A tool is designed to change the instruction sequence to improve performance, which would normally be useful, but benchmarks are only useful when measuring performance. exactly the same workload per attempt. Otherwise, final scores cannot be compared between different devices.

When Geekbench 6 is run through the Binary Optimization Tool, some workload scores go up to 40% and overall scores go up to 8%. These scores cannot be compared to other devices running Geekbench and are not a true reflection of computing performance. Intel Binary Optimization Tool only supports a few applications, and it only works on certain Core Ultra Series 3 processors (“Panther Lake”) and Core Ultra 200 Plus chips (“Arrow Lake Refresh”).

The blog post explained: “The techniques used are not clearly documented, and it is unclear how widely applicable these techniques are to different applications. The tool only supports a short list of applications, and Geekbench 6 is one of the few supported applications.”

Task manager in Windows 10.

3 ways Windows Task Manager is lying to you

The task manager looks reputable, but some of its numbers are misleading. Here’s how to fix errors from Windows’ favorite diagnostic tool.

Unfortunately, this means that some Geekbench scores for these Core Ultra CPUs are not accurate, and Geekbench currently has “no way to detect whether a Geekbench 6 result was run with or without the Binary Optimization Tool.” As a result, all benchmark results from CPUs that support the tool will have a message explaining that the result may be invalid. “It’s fake!”

It’s nothing new for hardware manufacturers to find ways to cheat benchmarks. The practice was especially widespread in smartphones Samsung, HTG, Sony, LG and OnePlus all try at one point or another – usually by temporarily overclocking the device when a comparator is detected. It was Intel using a compiler hack artificially increasing benchmark results for Xeon processors for several years.

If you’re trying to compare CPUs and there’s at least one new Intel CPU in the mix, you’ll want to check that the test is complete. without Intel’s fraud. Geekbench 6.3 is the only benchmark tool currently supported, but Intel may add others to the list.

Source: Geekbench Blog



Source link

Leave a Reply

Your email address will not be published. Required fields are marked *