GeekBench 6 boasted that the gap between iPhone and S23 was questioned by users

Since the launch of GeekBench 6.0 in February this year,the topic discussed by users on the Reddit community,Twitter social platform and some digital forums is”Apple and GeekBench have a PY transaction,which makes running score more inclined to iPhone”.

PhoneArena pointed out in the report that under Geekbench 5,the single-core fraction of Galaxy S23 Ultra is about 1600,and the multi-core fraction is about 5000;The iPhone 14 Pro scored 1900 and 5500 points.

In GeekBench 6,the Galaxy S23 Ultra has a single-core test score of about 1900 and a multi-core test score of about 5100.At the same time,the iPhone 14 Pro has a single core score of 2500 and a multi-core score of 6500.

Under Geekbench 5,the single-core performance of iPhone is 18%higher than that of Samsung,and the multi-core performance is 10%higher.In the Geekbench 6 test results,the leading edge expanded to 31%and 18%respectively.

Therefore,many users questioned the reliability of GeekBench 6.Many users held the view that Cook paid GeekBench to make the Galaxy S23 Ultra look slower than the actual speed.

The CEO of Geekbench doesn’t agree with this.A16 Bionic has expanded its leading edge over competing Android chips in GeekBench 6.He doesn’t think it’s data fraud,but the test is more accurate.

He said that in addition to machine learning and artificial intelligence,the GeekBench 6 scoring algorithm also incorporates more data features.