Recently, in the GeekBench score library in the Reddit community, Twitter social platform, and some digital forums. Users found to discuss the topic of “Apple and GeekBench have PY transactions, making the running score more biased towards iPhone.”
As reported by Foreign technology media PhoneArena pointed out in the Geekbench 5. The Galaxy S23 Ultra has received a single-core score of about 1600 and the multi-core score of about 5000. On the other hand, the iPhone 14 Pro has a score of 1900 and 5500 points respectively.
Whereas, in Geekbench 6, the Galaxy S23 Ultra scored approx 1900 on single-core tests and also 5100 on multi-core. In the interim, the iPhone 14 Pro has a single-core score of 2500 and a multi-core score of 6500.
Under the hood of Geekbench 5, the iPhone has 18% better single-core performance. Comparatively, it’s 10% higher than the multi-core performance of Samsung. Also, the Geekbench 6 test results, the lead extends to 31% and 18% categorically.
Thus, it’s being questioned by many users about the reliability of GeekBench 6. At the same time, various users held the view that Cook paid GeekBench to make the Galaxy S23 Ultra appear slower than it actually is in reality. In addition to this, the CEO of Geekbench contradicts, and the A16 Bionic further extends its lead over competing Android chips in Geekbench 6. Although it is not as he thinks, the data is wrong. But because the test is more accurate.
Furthermore, he stated that in addition to machine learning and artificial intelligence. The GeekBench 6 scoring algorithm also incorporates more data features. The same are as follows:
- Larger datasets
- AI object detection (when your iPhone/Android recognizes your face in a photo or knows to tell the difference between food and dog)
- Apply filters to your photos
- Blurred background (like you can add to a video call in apps like Zoom, Slack, iMessage, etc.)
- Photos of more than one million pixels (new phones come with 48, 50, 64, 108, and even 200MP cameras, which wasn’t common when Geekbench 2019 launched in 5)
- Larger maps, larger PDF files, new HTML workloads, navigation, ray tracing, and more.
John Poole, the founder of Primate Labs and founder of Geekbench, recently reveals some facts, which are as follows:
Perhaps the biggest change in Geekbench 6 is the way multi-core scores are calculated, measuring “how cores work together to accomplish a shared task,” rather than assigning each core a different task.
Including most modern ARM processors and Intel’s 12- and 13th generation CPUs, hybrid CPU architectures (a mix of large, fast cores and small, energy-efficient cores) are the future. GeekBench 6 wanted to better reflect how real-world multicore workloads would run, with more realistic CPU performance.
FOLLOW US ON SOCIAL MEDIA – Telegram, Twitter, Facebook, & Google News.