Geekbench Results Visualize Possible Link Between iPhone Slowdowns and Degraded Batteries

Primate Labs founder John Poole has plotted the kernel density of Geekbench 4 scores for iPhone 6s models running iOS 10.2, iOS 10.2.1, and iOS 11.2, visualizing an apparent link between lower performance and degraded battery health.


The charts show that on iOS 10.2, the vast majority of iPhone 6s devices benchmarked similarly in performance. However, Poole explains that the distribution of iPhone 6s scores for iOS 10.2.1 appears multimodal, with one large peak around the average and several smaller peaks around lower scores.

In other words, after iOS 10.2.1 was released last January, the performance of a percentage of iPhone 6s devices began to suffer.


In a statement, Apple said it made improvements in iOS 10.2.1 to reduce occurrences of unexpected iPhone shutdowns that a small number of users were experiencing. The shutdowns were reportedly caused by uneven power delivery from older batteries, which could cause an emergency shutdown of the devices.

While at least one report suggested that Apple tweaked its power management system in iPhone 6s devices, the company never disclosed what specific improvements it made to reduce the unexpected shutdowns.

A recent Reddit discussion, however, has reignited speculation that Apple is intentionally slowing down older iPhones to maximize power efficiency and stability when battery capacity has degraded, and reduce voltage-related shutdowns, and the Geekbench charts and Poole himself lend credit to that theory being true.

"The difference between iOS 10.2 and 10.2.1 is too abrupt to be just a function of battery condition," he said. "I believe … that Apple introduced a change to limit performance when battery condition decreases past a certain point."

The charts show that the number of iPhone 6s devices with lower Geekbench scores was even more pronounced on iOS 11.2, which is likely because the software update is around one year newer, which means that the battery capacity of many iPhone 6s devices has likely continued to deplete as expected.


Interestingly, even the iPhone 7 may be starting to be affected. While the distribution of Geekbench scores for the device on iOS 10.2, iOS 10.2.1, and iOS 11.1.2 appear identical, the results change with iOS 11.2 and start to resemble the iPhone 6s' degraded performance starting on iOS 10.2.1.

What this all means is that Apple may be intentionally slowing down older iPhones to maximize power efficiency and stability when battery capacity has degraded, as speculated, seemingly without publicly acknowledging so.

However, it's important to remember that all lithium-ion batteries naturally lose some of their ability to hold a charge over the course of a few years. Given the iPhone 6s was released in September 2015, the device has been available long enough that some users should consider replacing their battery regardless.

Apple charges an out-of-warranty fee of $79 to replace the battery of all eligible iPhone models in the United States. iPhone 6s users can contact AppleCare or schedule a Genius Bar appointment at a local Apple Store using the Apple Support app.

Apple did not immediately respond to our request for comment about the Geekbench findings. We'll update this article if we hear back.

Buyer's Guide: iPhone 8 (Buy Now)

Discuss this article in our forums

iMac Pros With Custom Xeon Chips Possibly Appear on Geekbench Ahead of December Launch

While the iMac Pro doesn't launch for another six weeks or so, possible benchmarks for the computer may have already surfaced on Geekbench. The results provide us with an early look at just how powerful Apple's $4,999-and-up desktop workstation will be when it is released in December.


Interestingly, the iMac Pro models benchmarked appear to have custom, downclocked Xeon chips that Intel hasn't publicly announced yet. There is a benchmark result for a model with a 3.2GHz 8-core Xeon W-2140B processor, while a third listing exists for a model with a 3.0GHz 10-core Xeon W-2150B chip.

All of the models are identified as "AAPJ1371,1," and unlike other Xeon chips, the processors have a "B" suffix. A few of the benchmark results are from late August, while the rest are from October.


MacRumors spoke with Geekbench founder John Poole, who speculated that the iMac Pro may require chips with lower thermal design power, and thus lower frequencies, due to its all-in-one form factor. He noted that the other chips in the Xeon Processor W family have relatively high TDPs of up to 140W.

The multi-core Geekbench score for the 8-core model averages out to 23,536, which is the highest performance of any iMac ever. It's nearly 22 percent faster than the latest 5K iMac equipped with a maxed-out 4.2GHz quad-core Core i7 processor, which has an average multi-core score of 19,336.

The higher-end 10-core iMac Pro has a multi-core score of 35,917, which is roughly 41 percent faster than the latest Mac Pro maxed out with a 2.7GHz 12-core Xeon E5 processor. Even its single-core score of 5,345 is faster than all but the highest-end 5K iMac released earlier this year.

All in all, the benchmarks point to the iMac Pro being unsurprisingly powerful from top to bottom. And that's not even looking at the 18-core iMac Pro, which hasn't been benchmarked yet and will surely blow every other Mac out of the water—at least until the modular Mac Pro is ready.

Apple said the iMac Pro will also feature top-of-the-line Radeon Pro Vega graphics, up to 4TB of SSD storage, and up to 128GB of ECC RAM. The computer will share the same design as the standard iMac, but with an all-flash architecture, a new thermal design, and four Thunderbolt 3 ports.

Related Roundup: iMac Pro

Discuss this article in our forums

New 27-Inch iMac Has Up to 80% Faster Graphics at Compute Tasks Compared to Previous Model

Earlier this month, Apple launched new iMac models with Intel's seventh-generation Kaby Lake processors and improved AMD Radeon Pro discrete graphics options at its annual Worldwide Developers Conference.


Early benchmark results for the new 27-inch iMac have already surfaced on Geekbench that suggest the 2017 models are up to 15% faster in multi-core CPU performance compared to last-generation models.

Apple's new high-end 27-inch iMac stock configuration with a 3.8GHz quad-core Core i5 processor has an average multi-core score of 14,886, for example, compared to 12,953 for the equivalent 2015 model.


John Poole of Primate Labs, the creators of Geekbench, said the new 27-inch iMac also has up to 80% improved graphics performance compared to the equivalent 2015 models at compute tasks such as image processing.

Geekbench's new GPU Compute Benchmark measures the performance of GPUs at performing compute tasks such as image processing, computer vision, and physics simulations, rather than rendering tasks.


Poole said compute performance is becoming more important as more applications, such as Photoshop, take advantage of GPU compute.

The built-to-order 27-inch iMac with a 4.2GHz quad-core Core i7 processor is Apple's fastest Mac ever in single-core CPU performance, according to the Geekbench results, continuing a trend set with the late 2014 model. The 2013 Mac Pro remains Apple's fastest Mac in multi-core CPU performance on Geekbench.

That will undoubtedly change in December when Apple launches the iMac Pro with workstation-class tech specs, including up to an 18-core Intel Xeon processor, top-of-the-line Radeon Pro Vega graphics, up to 4TB of SSD storage, and up to 128GB of ECC RAM. iMac Pro will start at $4,999 in the United States.

The new 27-inch iMac starts at $1,799. New 21.5-inch models are also available from $1,099. Prices vary by country.

Related Roundup: iMac
Tags: Geekbench, benchmarks
Buyer's Guide: iMac (Buy Now)

Discuss this article in our forums

2017 MacBook Pro is Up to 20% Faster Than Last Year’s Model in Benchmarks

Apple this week refreshed its MacBook Pro lineup with Intel's seventh-generation Kaby Lake processors, and early benchmarks for the notebooks suggest the 2017 models are up to 20 percent faster than the equivalent 2016 models equipped with Intel's sixth-generation Skylake processors.


Specifically, the new 15-inch MacBook Pro configured with a 2.9GHz Core i7 processor has average single-core and multi-core scores of 4,632 and 15,747 respectively based on nearly a dozen Geekbench 4 results so far.

By comparison, last year's 15-inch MacBook Pro configured with a sixth-generation 2.7GHz Core i7 processor, which was the equivalent high-end stock configuration, has average single-core and multi-core scores of 4,098 and 13,155 respectively based on over 4,800 Geekbench 4 results.

On a model-vs-model basis, the benchmark results suggest the 2017 MacBook Pro with a 2.9GHz processor is up to 13 percent faster in single-core performance, and up to 19.7 percent faster in multi-core performance, than the equivalent 2016 MacBook Pro model. Its price remains unchanged at $2,799.


There's only one Geekbench result for the new 15-inch MacBook Pro's base configuration with a 2.8GHz Core i7 processor, but the benchmarks suggest that model is up to 9.5 percent faster than the equivalent 2016 MacBook Pro equipped with a sixth-generation 2.6GHz Core i7 processor.

There are no Geekbench results yet for the new 15-inch MacBook Pro's highest-end built-to-order configuration with a seventh-generation 3.1GHz Core i7 processor, so its performance cannot be compared to the equivalent built-to-order 2016 MacBook Pro with a sixth-generation 2.9GHz Core i7 processor.

Related Roundup: MacBook Pro
Tags: Geekbench, benchmarks, Kaby Lake
Buyer's Guide: MacBook Pro (Buy Now)

Discuss this article in our forums