Hacker News Reader

Comment by DustinCalim | Hacker News Reader

Reading through the comments here, I see a lot of reference to GPU speeds. Anand mentions in the article that the new iPhone 5s is more powerful than the 2010 Macbook Air 11" and has more graphics performance than the iPad 4. I used to be one of those guys who would build his own PC and spend hours overclocking everything for the maximum performance gain but I do not understand this;

What's the infatuation with performance in a cell phone?

Snappy is snappy but there are diminishing returns here past a certain point; I think Anand stated it best when he said it is "the most future-proof of any iPhone ever launched"

View on HN

> What's the infatuation with performance in a cell phone?

With better raw CPU and GPU performances, it's easier and faster to write decent code with less time spent on optimisation and well as a great UI — Motion Effects and UIKit Dynamics (supporting parallax effects, physics engine, 30D/layered look) requires a powerful GPU. Games, of course always push the performance of CPU and GPUs. I doubt there'll ever be future-proof hardware in terms of performance, unless the software is dead.

Phones haven't quite reached the point that desktop computers have. With a desktop, a 5 year old machine today is still completely useable even with the latest software. Meanwhile, a 5 year old phone is the original iPhone. This generation of phones might be the beginning of that era, but the increased performance in phones is still enabling new features that weren't possible before (full res burst with image evaluation, 120fps video editing, etc).

If it's true that the iPhone 5s of today matches or exceeds the performance of a laptop from 5 years ago, that could mean we're not long away from full-featured convergence devices (plug your phone into a monitor+keyboard to turn it into a complete PC)

It runs about as fast as a Core2 Duo 2.4GHz from 2010, which was used that year in the Mac Mini.

So, yeah, we're there.

> What's the infatuation with performance in a cell phone?

Consider the case when Apple starts using these (ARM) in a MacBook Air.

I really doubt that will ever happen. If anything I think it will end up being the other way around with x86 CPUs in phones.

I care about cell phone performance because I'm excited for the day when I can realistically start booting a general purpose OS on my phone and call it good enough w/ a monitor and external keyboard.

That will also be the day where you need to 'jailbreak your PC' thanks to Apple.

>> What's the infatuation with performance in a cell phone? Assuming you're talking about GPU performance: Apple cares a lot about GPU performance because iOS uses the GPU heavily for a lot of tasks - all the animations, transitions, blurring and blending effects that you see are executed on the GPU, mostly via CoreAnimation.

Also, there are new games and other applications that are pushing GPU requirements to near desktop-class. For example, the Infinity Blade demo that was shown during the launch.

I think a lot of people care about having enough GPU performance to enable all of these.

Cell phones don't need uber graphic performance but we are a long way from these devices being merely cell phones. They are general purpose computer with sophisticated displays and interfaces. Performance matters here.

While I don't generally get why people get so hung up on specifications ("my phone has more cores than yours"), the performance is actually put to good use. You wouldn't be able to do HD video recording and editing, slo-mo and burst shots and a host of other things had it not been for more powerful and efficient processors. It has nothing to do with snappy, and everything to do with functionality.

One advantage: the faster the hardware, the more time it can spend in idle states at the same load. So, if you manage to speed up your hardware by a factor X while increasing power usage by a factor < X, it is a net win (I don't know whether they achieve this)

That's not really true for the sorts of architectural changes happening here. Because smaller process nodes mean that leakage power for transistors has grown faster than active power there's now a lot to be said for running those transistors fast for short bursts rather than at a reduced frequency. But adding more transistors still increases your power usage linearly, and performance only goes up with something like the square root of the number of transistors you use. There's things that make this not entirely true on the system level, if a better branch predictor or better cache then you might see less bandwidth to main memory and you'll be saving power there. And you can do things with the silicon level process to reduce leakage power at the cost of performance but then make the performance back up with more transistors and still come out ahead. But in general a smaller core will be more power efficient than a larger one.

Reply on news.ycombinator.com