When Apple announced its latest iPhone, rumors soon-after hit the Web of leaked benchmarks that placed it well above the competition. With phones finally reaching the hands of fans, though, we’re finally able to get some hard numbers, and as usual, Anand wasted no time to get down and dirty with some benchmarking. The results? Very much in Apple’s favor.
Apple touted the GPU in its A6 SoC as being twice as fast as the previous generation, and according to the benchmarks, it’s true. In GLBenchmark 2.5’s fill-rate test, the iPhone 5 scored 1650, whereas the 4S scored 765.5. This kind of boost should come off as a little concerning to NVIDIA, as its Tegra 3 chip found in HTC’s One X scored a mere 407. Apple’s A6 4x faster than Tegra 3? It sure appears so.
Improvements are not just seen in GPU tests, but that is where the biggest boosts are. Given these numbers, it’s no wonder I’ve heard accounts where people have said that using 3D maps felt even more than 2x faster – the iPhone 5 sure does have some brawn behind its modest frame.
In BrowserMark, the iPhone 5 scored 191K, vs. the 4S’ 106K. It’s clear that non-GPU benchmarks have seen an improvement also, but the differences aren’t quite as stark when comparing to Samsung’s Galaxy S III, which scored 172K. In Google’s V8 benchmark, HTC’s One X came close to the iPhone 5, but this is one of the very few tests where things got quite so close.
Color me impressed. On the desktop, we’re stoked when a CPU gets released that boosts performance by 10%, and at this point in time and given all of their features, it is still impressive. But here, Apple almost doubled all the possible performance points with its A6. The iPhone 5 on the surface might not have seen much of an upgrade, but the CPU under the hood sure has.