07-21-11 03:53 PM
36 12
tools
  1. belfastdispatcher's Avatar
    I think then it is a different comparison, not necessarily the speed of the browser, but how well the OS uses the data it has. (Which RIM has always excelled at)

    For speed tests though, you have to clear everything so that it is the most representative. Sure sometimes you may visit the same site, but you will always have to visit a site for the first time, and the speed that the browser can receive the data and display it, it usually what is being tested.
    To make the browser tests more accurate I would like to see a refresh test after the initial page load.

    Posted from my CrackBerry at wapforums.crackberry.com
    07-20-11 10:28 AM
  2. CanuckSoldier's Avatar
    Yes I agree. Browsers do not exist separate from the mobile platform. Real world performance is what matters to us end users. While I understand testers wanting to create a level test environment for all tests, they should make those tests reflective of how the end users use the phones.

    PC magazines made this mistake in the 90's by using synthetic benchmarks that are ok for comparing one system to the next system in a generic way. But those tests did nothing to tell users how the system would handle game X or how fast would render a photo in App X. Now the tests that mags use are all based on game/app core tests and mean alot more to the end user.

    So yes we all load a page for a first time, and uncached loading is a valid benchmark, but most of us are creatures of habit and do visit the same news sites, forums, etc. From my habits I'd say that probably more than 80% of all the sites I visit are the same every day.

    So cached speed is just as valid a test as uncached speed and testers should design their tests to reflect the way we use our phones not just try and mold our opinions based on "synthetic" tests that don't reflex real world usage.

    So yes Android and iphones are better at loading pages the first time, and BB's excell at data management and therefore are going to be very fast at the sites you visit repeatedly. Both are important real world benchmarks and both should be reported as equally important.

    CS
    07-20-11 01:52 PM
  3. 01itr's Avatar
    Yes I agree. Browsers do not exist separate from the mobile platform. Real world performance is what matters to us end users. While I understand testers wanting to create a level test environment for all tests, they should make those tests reflective of how the end users use the phones.

    PC magazines made this mistake in the 90's by using synthetic benchmarks that are ok for comparing one system to the next system in a generic way. But those tests did nothing to tell users how the system would handle game X or how fast would render a photo in App X. Now the tests that mags use are all based on game/app core tests and mean alot more to the end user.

    So yes we all load a page for a first time, and uncached loading is a valid benchmark, but most of us are creatures of habit and do visit the same news sites, forums, etc. From my habits I'd say that probably more than 80% of all the sites I visit are the same every day.

    So cached speed is just as valid a test as uncached speed and testers should design their tests to reflect the way we use our phones not just try and mold our opinions based on "synthetic" tests that don't reflex real world usage.

    So yes Android and iphones are better at loading pages the first time, and BB's excell at data management and therefore are going to be very fast at the sites you visit repeatedly. Both are important real world benchmarks and both should be reported as equally important.

    CS
    I think it would be hard to test that though. If you are say loading the CB front page, and there is a new post, will the browser load the page brand new since it is different from the last time you viewed it? Or will it recognize that some things are the same and then retrieve those from the cache? In which case it will all depend on each different site and how much of difference there is between what is current and what is in the cache.
    07-20-11 01:58 PM
  4. belfastdispatcher's Avatar
    I think it would be hard to test that though. If you are say loading the CB front page, and there is a new post, will the browser load the page brand new since it is different from the last time you viewed it? Or will it recognize that some things are the same and then retrieve those from the cache? In which case it will all depend on each different site and how much of difference there is between what is current and what is in the cache.
    Yes but if the test is done at exactly the same time it doesn't matter does it? Load a page for the first time on both then refresh the page on both.

    Posted from my CrackBerry at wapforums.crackberry.com
    07-20-11 02:02 PM
  5. CanuckSoldier's Avatar
    Yes it doesn't matter if a BB loads 1 new file on a refresh and an iphone downloads half the site during a refresh. This is infact a very meaningful test of how the entire phones platform is designed, and works in the real world. Exactly what end users want to know!

    Yes to have a fair comparison you should refresh the exact same url within seconds on both phones, but this is not hard to pull off.

    CS
    07-20-11 02:33 PM
  6. belfastdispatcher's Avatar
    Another aspect of browser testing that's never mentioned in tests is the ease of navigation, the keyboard makes this a very very fast experience, W for Tabs,R for refresh, G brings you straight to bookmarks, A creates a bookmark, F for search in page not to mention the usual T,B,Space for page down and more

    Ever seen a touchscreen user frantically scrolling back to the top page?

    Posted from my CrackBerry at wapforums.crackberry.com
    07-20-11 02:40 PM
  7. 01itr's Avatar
    Ever seen a touchscreen user frantically scrolling back to the top page?
    At least on my iPhone 4, if you tap the top of your screen it automatically scrolls all the way to the top.

    However, your other points are valid, but I still find the touch experience faster for navigating than using a physical keyboard. Not comparing iOS to BB, I feel the same with Android phones and when I had my Torch briefly.
    07-20-11 02:52 PM
  8. belfastdispatcher's Avatar
    At least on my iPhone 4, if you tap the top of your screen it automatically scrolls all the way to the top.

    However, your other points are valid, but I still find the touch experience faster for navigating than using a physical keyboard. Not comparing iOS to BB, I feel the same with Android phones and when I had my Torch briefly.
    I suppose depends on how you use it, I like that you can access all the functions from wherever you are withing the browser page, you don't have to get back to the top.
    Then there's some missing features that never get metioned, some videos for example you can't play on the iphone, android can play them yet BB browser can play them or download them.

    This for example: http://www.u.tv/utvplayer/video/138058


    Posted from my CrackBerry at wapforums.crackberry.com
    Last edited by belfastdispatcher; 07-20-11 at 03:11 PM.
    07-20-11 03:08 PM
  9. CanuckSoldier's Avatar
    Another aspect of browser testing that's never mentioned in tests is the ease of navigation, the keyboard makes this a very very fast experience, W for Tabs,R for refresh, G brings you straight to bookmarks, A creates a bookmark, F for search in page not to mention the usual T,B,Space for page down and more

    Ever seen a touchscreen user frantically scrolling back to the top page?

    Posted from my CrackBerry at wapforums.crackberry.com
    Yes human engineering aspects are much more a factor with smartphones than with PC's. So depending on how you use these features, a "slower" phone can still be more efficeint than the phone with the killer specs. Though I think we all agree that in general a faster CPU and more RAM is not a bad thing But high end specs are not their own end of they are not paired with a good UI and good human engineering.

    CS
    07-20-11 03:20 PM
  10. 01itr's Avatar
    Then there's some missing features that never get metioned, some videos for example you can't play on the iphone, android can play them yet BB browser can play them or download them.

    This for example: UTV Player | Watch UTV Live
    Make no mistake, I absolutely hate the iOS Browser. Although it is better (faster/smoother) than the BBOS6 browser, it still pales in comparison to the Playbook/Honeycomb browser. However, it is more reliable. What use is super fast loading and full content webpages if the browser crashes and quits? Give it time, the Playbook browser is going to destroy iOS/Hcomb when(if) it is polished.
    07-20-11 03:25 PM
  11. ADFXPro777's Avatar
    Great posts everyone. I decided to do a test similar to belfastdispatcher's suggestions:

    First, I cleared all caches (except for push content) on both the Blackberry and Bolt browsers. I then visited the Crackberry site and BBC News ten (10) times on each browser. Then, I revisited them another ten (10) times on each browser.

    For Crackberry.com, the Blackberry browser averaged a 5.2 second download time, while the Bolt Browser averaged 7.8 seconds. For BBC News, the Blackberry browser averaged 5.5 seconds while the Bolt browser averaged 5.3 seconds.

    The test does not stop there. I went to 3GSpeed.info to measure the average download speeds from each browser. Out of 10 tries, the Blackberry browser averaged 710 kbps while the Bolt browser averaged 2.3mbps. Rather interesting that the Bolt browser had a faster speed rate than the Blackberry browser, yet posted similar times when visiting webpages.
    07-21-11 03:53 PM
36 12
LINK TO POST COPIED TO CLIPBOARD