As it turned out, Apple is using two third party companies (TSMC, and its own rival, Samsung) as manufacturers for the A9 chip inside its latest iPhone 6s and iPhone 6s Plus. This is nothing new, as Apple, in the past, has made similar decisions regarding other key components as well. This year, however, the SoCs are not only made by two different companies, but they are also different in the production technology used.
Some reports recently claimed that there is substantial difference between the two chips, if not necessarily in performance, but in power consumption and temperature as well. Apple felt the need to chime in officially, stating that both these chips meet the company standards, but, as usual, that didn’t stop the iPhone 6s chipgate from quickly escalating.
Consumer Reports believes that there is no such thing as the iPhone 6s chipgate. After a series of test conducted in controlled environment, with identical setups, Consumer Reports claims that there is no difference (or, if there is, it is negligible, at about one or two percent) between iPhones equipped with different versions of the A9 chip.
So, what’s really the difference between Consumer Reports’ findings, and the other reports that claimed up to 20 percent differences? While Consumer Reports’ tests were emulating real life usage, the other reports were heavily relying on synthetic benchmark data. Benchmarking applications usually force the processor to run at its maximum speed, and measure the time until the battery completely depletes under these circumstances. While these are a metric that can be taken into consideration on paper, they have nothing in common with real life usage, as the chip will most likely not be clocked at its full potential but for short periods of time during one charge cycle, under regular usage.
According to Consumer Reports’, there is no iPhone 6s chipgate, or not one that can be proven to exist under normal usage scenarios, which include phone calls, texts, social media, browsing, multimedia, gaming, etc.
In conclusion, as long as your usage scenario doesn’t include running benchmarks 100% of the time, you should be fine. What do you think? Is there a real “gate” here, or is the problem easily dismissible?
Source: Consumer Reports