The latest Android smartphone loaded Web pages 52 per cent faster than iPhone 4 running iOS 4.3, according to thousands of independent field tests released today by Blaze Software.
The Web page load times were about a second apart for the two devices in a study that amassed 45,000 load tests in all. For Android 2.3 on the Google Nexus S smartphone using a version of Chrome, the median load time was 2.144 seconds, compared to 3.254 seconds for iPhone 4 on iOS 4.3 running a version of Safari, according to the study.
Blaze used Fortune 1,000 Web sites for the tests, running the Web page loading tests repeatedly over Wi-Fi and 3G wireless connections with nothing else running on the phones at the time. The Android phone was faster than the iPhone in loading 84 per cent of the tested Web sites. “Android wasn’t just faster overall, but rather provided a faster browsing experience four times out of five,” the study said.
Blaze sought to describe its tests as objective, adding it has no association with Google or Apple “in any form,” David Horne, marketing programs manager for Ottawa, Ont.-based Blaze, said in an e-mail. Blaze writes software to automatically accelerate Web site speeds and created a mobile testing tool used in the Android-iPhone study to be able to analyze mobile Web performance and to “discover new optimization to add to our core product,” Horne explained.
While Android came out ahead in the load time comparison, the study noted that both are “generally fast.” However, the study also noted that “browser speed is a big deal” and had been a prominent point when both Apple and Google recently noted their improved JavaScript engines.
“Browser performance is all the rage, and everybody says theirs is faster,” the study added.
The study’s authors said they were surprised by the results.
One surprise came because both iPhone and Android had optimized JavaScript engines in their latest versions, but were not much faster than previous versions also tested, Blaze said. “Both Apple and Google tout great performance improvements [with optimized JavaScript] but those seem to be reserved to JavaScript benchmarks and high-complexity apps,” the study said. “If you expect pages to show up faster after an upgrade, you’ll be sorely disappointed.”
Blaze said part of the problem is that the SunSpider JavaScript benchmark, a kind of custom test used by Apple and Google, and other benchmark tests “are very different than real-world sites and don’t reflect the actual user experience.” Blaze said it measured the load time of Web pages, “mimicking the experience users would get when browsing on their smartphones.”
Other testing groups have only compared a small set of sites manually, Blaze said.
Additionally, of the 1,000 sites tested, 175 were customized for mobile access. The iPhone improved the time to load a mobile Web site by 39 per cent over other standard Web sites, while Android improved the difference in the two types of sites by eight per cent. Still, Android loaded both types faster than the iPhone 4.
Most of the testing Blaze conducted was over Wi-Fi in a home using a high-speed router connected to a fast DSL line, but it also conducted some 3G testing (with nearly 6 Mbit/sec download speed over the Bell Mobility HSPA network) with iPhone 4 running iOS 4.2 this time. In that comparison, it found Wi-Fi was faster in 82 per cent of the cases, but only by half a second.
Blaze’s testing apps — WebView and UIWebView — used each phone’s embedded browser, Chrome for Android and Safari on the iPhone 4, respectively.