Whenever a high end smartphone is released, techies and wall street pundits clamor to inspect, dissect, detect and finally interject their opinions into the fray. Among them are perennial fan boys and bashers on both sides. One thing, however, is pretty clear; the iPhone is regarded as being significantly more responsive than Android and Windows. By responsive I mean the time it takes for a touch, or change in position of a touch, to be reflected by a response by the system.
You don't have to be a hardcore Apple fan to notice this. There are numerous references to Android being "laggy" even on Android boards, and have been for a long time. For example, nickdalzell wrote in 2012 in Android Forums:
I don't get It and it makes me sick. … it becomes so frustrating as the OS fails to keep up with my 80+ WPM touch-type rate that it lags or produces often misspelled results or poor performance… I find myself in such instances whipping my ipad from the closet finding the response in comparison rather snappy, my only question is, why? Why is android, even today, with today's specs so slow?
Why does it matter?
It matters very much. The iPhone has consistently won highest customer satisfaction ratings by groups such as J. D. Powers (although a recent ACSI report does show the Samsung Galaxy edging them out). This customer satisfaction is a prime force in maintaining the view of Apple as a premium brand and justifying the current price points and hence margins - and margins are certainly of great interest to the investor.
This sensation of "snappiness" or responsiveness is something that has always set the iPhone above the competition, a key factor in customer satisfaction, especially if an iPhone owner uses a competing product. It is only after you listen to a good stereo system that you notice the lack of quality in a mediocre one. The same holds true here to a certain degree.
Still, the reality of an iOS lead has been mostly subjective. Until now.
Ad technology company Agawi has just released a report TouchMarks I: Smartphone Touchscreen Latencies that provides "the first quantitative and objective benchmark of app response times…" The report goes to "measure and reveal the minimum response times of flagship smartphones from top manufacturers." They described their process:
Using a combination of high frame rate cameras capturing at 240fps and custom Agawi hardware (pictured above), we can accurately measure the App Response Time (ART), which we define as the latency from the time the user feels that they've touched the device's display to the time the user sees a response on the screen… this benchmark defines the Minimum App Response Time (MART) a user could experience on a mobile app on each device.
The results are unequivocal: Apple iPhone 5 had less then half the latency of any of the other competitors, 55 milliseconds vs. 114 milliseconds for the Samsung Galaxy S4, and the rest are up from there. (The chart is shown at the bottom.)
Hey, 114 ms that is not much - you might say. Actually, 100 ms is one-tenth of a second and this is definitely noticeable. (Say "one-thousand-one", the "th" is roughly 1/10 sec.) This means that on the best of the Android systems, the lag from touch to response is over 1/10 of a second, and that on an iPhone it is half that.
There is a video from Microsoft's Applied Sciences Group that shows exactly what this means. (Video here.) Oddly, the researcher, Paul Dietz, chose to demonstrate almost precisely the delay values at issue (100 ms vs. 50 ms) so you can get a realistic view of the difference as he moves his finger around the screen. There is a huge difference. Dietz concludes that while 50 ms is a great improvement over 100, it is at 1 ms that you get a perceptual change, that you are really manipulating a physical object.
Still, the improved responsiveness of the iPhone seems to be very noticeable, particularly on tasks such as list scrolling.
If the Android systems have processors running even faster than those on the iPhones, then why the increased delay?
In order to understand this, we need to understand how such a system works. Computing devices work in a series of interacting layers - from hardware to computing units to software and back again. I have illustrated this in the following drawing. (Note that it is simplified.)
[Drawing: J M Manness]
Each of these systems contains its own inherent time delays. The speed of the CPU, therefore, is only one portion of the equation. Still, if we assume that Samsung has the best touchscreens available (they do make them after all), then the top three layers ought to be at least as fast as those in Apple's devices. Therefore, the issues must be in the code.
Software engineer Andrew Munn, who once interned at Google, provides some insight to this issue in his post on "Android graphics true facts." It is a rather technical argument, but at the bottom of it all, he argues that there were design decisions made in the early development of Android that negatively affect the ability of the system to respond. These have to do with the sequence and priority of response events when the hardware detects a touch.
Android UI will never be completely smooth because of the design constraints I discussed at the beginning:
- UI rendering occurs on the main thread of an app
- UI rendering has normal priority
The original Android prototype wasn't a touch screen device. Android's rendering trade-offs make sense for a keyboard and trackball device. When the iPhone came out, the Android team rushed to release a competitor product, but unfortunately it was too late to rewrite the UI framework.
Munn argues that this would be very difficult for Google to fix, and so they have an inherent disadvantage to iOS in this area.
Then why is this bad news for Apple?
Engineers exist to solve problems. However, they generally like to have the problems quantified before they address them. Prior to TouchMarks study, there was no quantification of the difference, it was all subjective. Now we have a real - and public - disclosure of the problem. Now the Android developers have greater motivation to address the problem, and specific measurements by which to judge their success. This means that in a few years, Apple may have to give up its advantage.
The performance of the iPhone and iPad are important issues for the investor, as they indicate that Apple really does have a significant technological edge over the competition, and this directly counters some of the current fears in the investment community that Apple is no longer a leader.
The responsiveness of iOS devices has long been one of the advantages that leads to the great user experience that is so prized by the customers and in which Apple takes great pride. Until now, this has been only a feeling by users, observed perhaps, but never quantified. Now it has been quantified.
The benefit to Apple is that now this advantage is incontrovertible. The disadvantage is that now their opponents can better focus on diminishing it.
[Source: Agawi ]
Related Article: Apple Shares Buyback - the numbers talk.