Slashdot Mirror
Apple's A11 Bionic Chip In iPhone 8 and iPhone X Smokes Android Handsets In Early Benchmarks (hothardware.com)
MojoKid writes: Many of the new releases of Apple's iPhone bring with it a new A-series SoC (System on Chip) and Apple is keeping that tradition with the iPhone 8 and iPhone 8 Plus, and iPhone X. Each of those handsets sports a custom ARM-based A11 Bionic processor with six cores -- four high performance cores and two power efficiency cores. The two power efficiency cores will perform the bulk medial chores to maintain battery life, which Apple says will be 2 hours longer than the iPhone 7. However, for heavier workloads, the chip is capable of not only firing up its four high performance cores, but also all six cores simultaneously. If early leaked benchmarks are any indication, the A11 Bionic is going to be a benchmark-busting beast of a chip. A set of just-posted Geekbench scores reinforces that notion. Just prior to Apple announcing its newest iPhone models, Geekbench's database was updated with a new entry for an "iPhone 10,5" which we assume to be the iPhone X. Based on the scores recorded, in this one benchmark at least, the A11 CPU powering the iPhone X appears to be 50 to 70 percent faster than any Android handset on the market currently, even those powered by the new Qualcomm Snapdragon 835.
Too bad its stuck to IOS, the OS made specifically for non power users.
Apple devices tend to destroy everything else on the market in benchmarks, when they first come out. Most of a year later, some Android devices will come out that come close to or even surpass its speed, only for Apple to release a new chip a few months later that leaves them in the dust. One exception: Atom chips (in certain Windows convertible tablets) tend to outperform them, although are generally higher wattage so it's arguably an unfair comparison.
Corruption is convincing someone that the selfless ideal is the same as their selfish ideal.
Android doesn't run it's apps on a java interpreter.
Nerds who can't afford it hate it. Nerds who can afford it buy it and love it.
Hell for that matter apple compiles its apps on llvm, guess what android compiles its apps on nowadays... llvm
Nope. The initial versions of ART used LLVM, but the Google folk couldn't get the resource requirements down enough to run on the phone (no idea why they don't do the compilation on the app store and cache the results, rather than warming the planet), so they moved to a completely different infrastructure. Modern ART has two compilers (that they're trying really hard to unify). There's a JIT that runs when you first start an app. This collects profiling information, but doesn't do much optimisation initially. It will then generate optimised code for some of the hot paths. The profiling information is recorded and overnight (or at any period when the phone is plugged in but not used) the AOT compiler starts in the background and will generate optimised binaries. Unfortunately, the AOT compiler doesn't allow on-stack replacement and so the JIT can occasionally give better code for hot loops (it can perform speculative optimisations that are correct 99.9% of the time and then deoptimise in the case where they're incorrect, whereas the AOT compiler has to do the slower optimisation that's correct 100% of the time).
I think there's also an interpreter for fast start, but I lose track (the ART team keeps changing their mind about whether an interpreter is a good idea).
I am TheRaven on Soylent News
Apple's CPU's belong to the ARM
Do you understand the difference between architecture and implementation? Two people can implement the same architecture and get very different performance.
Intel (used to) love flogging AMD for having the same architecture but significantly worse performance and horrible power. It was a little unfair in that Intel has a state of the art fab that is often the very best in the business, while AMD had to use legacy technology from a less top tier fab on an older process node, but that wasn't the entire source of the discrepancy. In this case, Apple's rivals absolutely do have access to the same top tier processes, originally Apple had its main rival fab its chips! The difference was purely implementation of the architecture.
The best example I can give you is that the architecture defines say, a 32-bit multiply instruction. It defines what the result will be for any given combination of multiplier/multiplicand, including edge cases, and overflow cases, etc. It says nothing about how you implement that instruction, only that you must be able to handle it, and it must produce a given result. You *could* implement it using an adder and a for loop like we learned in grade school. Even running at 1GHz on the very best process, you will have made the slowest multiplier in the industry, and this very much would show on any given benchmark that used multiplies in its tests (all of them).
Figuring out clever ways of implementing that instruction that are fast, low power and consume low silicon geometry is what logic designers/computer engineers spend their days on. It makes a huge difference, and what you should read here is that Apple has beaten its competitors pretty thoroughly on implementation of ARM, and that implementation is entirely owned by Apple and requires significant investment from its competitors to keep up with.
The message to customers is the A11 chip is the best out there, the message that investors won't want to hear is that in-house design is beating the shitty ODM model to small bits. Google and Microsoft are both looking at Apple and thinking they need to design their own chips too, that Samsung, Huawei (Chinese for "state sponsored spies and world-class fuckups") etc. are not invested in their products.