Hacker News Reader

Comment by mirsadm | Hacker News Reader

Interestingly enough the GPU is approximately the same speed as the upcoming (or current?) generation of Android phones with the Snapdragon 800 SOC. In the past Apple has generally held a large performance advantage in that area. There is potential that in 6 months when the next Galaxy/HTC phones are released that they will be significantly ahead of Apple for 6 or more months.

View on HN

As Anandtech points out: the iPhone 5s' PowerVR Series 6 moves to a fully scalar architecture (allowing better shaders and GPGPU performance) but loses a lot of raw triangle performance in the process (30-50% slower than iPhone 5 in some cases). On shader heavy scenes (T-Rex HD) the tradeoff is worth it (iPhone 5s is 30% faster than any other GPU).

Apple (and PowerVR) have made a couple tradeoffs in this generation that sacrifice performance on some tests for better shader capabilities.

Final point: it'll be interesting to see what these numbers look like when the bugs in the different graphics tests are resolved (are they compiled using the armv8 instruction set?)

That 30% faster figure is because the iPhone is rendering the scene at a lower resolution than many of the phones with larger, 1080p screens. Look at the next graph where each phone is rendering a 1080p scene to see the difference.

I'm not sure current benchmarks can really measure the A7 correctly, or if they are at all ready to deal with OpenGL ES 3.0.

All new chips support OpenGL ES 3.0 except Tegra 4.

Benchmark applications are not GPUs.

Besides that, GFXBench 2.7 doesn't support ES 3.0 and that's what most of the comparisons with other current-gen GPUs have used.

The Snapdragon 800 tested is a relatively power-hungry tablet part. I _think_ the phone ones will have fewer execution units. If history is a guide, the iPad 5 A7 (A7X?) will have twice the execution units.

LG G2 is a phone.

Reply on news.ycombinator.com