Sign in with
Sign up | Sign in

RoboHornet Pro: Microsoft Snubs Google, Mozilla Concurs!

RoboHornet Pro: Microsoft Snubs Google, Mozilla Concurs!
By

Last week we broke the story of RoboHornet, a new Web browser performance benchmark. Today, we chronicle the fallout of what is perhaps the worst-received benchmark launch in history, plus add the results of Microsoft's counter-move: RoboHornet Pro.

Holy train wreck, Batman! The end of September was one weird, wild week for Web browser news.

Despite RoboHornet being an independent GitHub project on paper, Microsoft and others aren't buying it. In possibly the fastest turnaround in development history, Microsoft has taken RoboHornet from first alpha release to "Pro" in less than 24 hours. In another surprising turn, Mozilla formally concurs with Redmond that RoboHornet's technical merits are questionable. Perhaps arch-rivals Microsoft and Mozilla finally realize that Google is their real enemy. After all, while the browser war veterans were busy battling each other, Chrome just waltzed through the front gates and usurped the throne.

Let's go through the wacky events of last week as they occurred:

Monday: We broke the RoboHornet story. The new Web browser performance test is actually a suite of micro-benchmarks in the areas of HTML, CSS, JavaScript, DOM, and SVG. In our testing, Internet Explorer 10 for Windows 8 claimed a decisive victory, more than doubling the average Web browser performance on the test creator's reference MacBook Pro (late 2011).

Tuesday: Microsoft says "Thanks, but no thanks" to RoboHornet, dismissing the new test as a meaningless micro-benchmark that does not reflect real-world performance. Below is an excerpt from the IE Blog, and from the language used in the opening sentence, you can tell that Microsoft isn't buying the "independent benchmark" part:

Yesterday Google released its latest micro-benchmark, RoboHornet, in which Internet Explorer 10 scores rather well.  While we appreciate the gesture, members of our engineering team took a look at the benchmark and found that RoboHornet isn’t all that representative of the performance users might encounter on real-world sites. Like all micro-benchmarks, RoboHornet is a lab test that only focuses on specific aspects of browser performance.

We decided to take the RoboHornet micro-benchmark and run it in the context of a real-world scenario. Using modern browser capabilities like CSS3 Animations, CSS3 Transforms, CSS3 Text Shadows, custom WOFF fonts, Unicode, Touch, and more, we created a site that looks a little bit like the familiar Matrix. We then ran the RoboHornet micro-benchmark in the context of this real website. While running both the Matrix and RoboHornet micro-benchmark at the same time, Chrome slows to a crawl and stops animating the screen, because it wasn’t designed to handle a benchmark load in the context of a real-world scenario. Meanwhile Internet Explorer 10 remains responsive, continues animating the screen, and finishes the test in less than half the time that Chrome does… We have made RoboHornet Pro available on IE Test Drive, so you can check it out for yourself.

Later that day Mozilla's Justin Lebar opened a bug on RoboHornet's GitHub page entitled Eliminate and outlaw micro-benchmarks where he states:

If you guys want us (in my case, Mozilla) to take robohornet seriously, I strongly recommend you write some macrobenchmarks and eliminate the microbenchmarks from your test suite.

Wednesday: Microsoft-affiliated steward, John David Dalton, drops all mention of Microsoft from his RoboHornet stewardship. Meanwhile, the Mozilla steward, Daniel Buchner, leaves the committee entirely.

Today: We have Mozilla's official response to the RoboHornet debacle:

There are a lot of benchmarks out there and different benchmarks for the same particular task can behave very differently. Many benchmarks are self-serving, in that the creators will typically pick out a set of programs that they think are worthwhile to get faster on and then only after turning these workloads share them with the wider community. However, what developers and browser implementers really need to have here is good benchmarks that allow us to better see holistic performance. We'd like to see more benchmarks created that focus on the entire consumer experience--for example, benchmarks that focus on interaction with the browser, popular web applications or sites, and common tasks like panning and zooming on mobile.

Micro-benchmarks, like RoboHornet, do not accurately reflect the user experience on the Web. RoboHornet aims to measure real performance, but it falls short. RoboHornet is currently 17 micro-benchmarks, each of which measures one thing that a website can do. But real websites do hundreds and thousands of things, so almost all of them are entirely unaccounted for by RoboHornet. RoboHornet lists some things that are currently slow in browsers and points them out for attention from vendors. That's a useful service, but it's not the same as accurately representing real performance.

It is still unclear what exactly Mozilla thinks about Microsoft's RoboHornet Pro, but now that results are published, we'll ask for a statement and update the story accordingly.

Of the remaining top five browser vendors, Opera refuses to comment, and Apple cannot be reached.

RoboHornet Redux

RoboHornet Pro is already off to a better start than RoboHornet because it works with so many more browsers than the original test. We are now able to include results from the Windows 8 Metro versions of Chrome and IE10, as well as Maxthon and Sleipnir on both Windows and OS X. iOS browsers can now run the test as well. We've included Yahoo! Axis, Google Chrome, Dolphin, Mozilla Firefox, Maxthon, Apple Safari, and Sleipnir. While Android loses its stock browser, it retains Chrome and gains Opera Mobile.

Let's quickly recap the test setup, and then see how all these browsers fare on RoboHornet Pro.

Display 27 Comments.
This thread is closed for comments
Top Comments
  • 11 Hide
    techcurious , October 2, 2012 10:17 PM
    ... nevermind.. I would rate myself down if I could.. cause after further thought, I realized that anyone that chose to use OSX over Windows will never choose to use IE over Safari or Firefox.. Something I am sure Microsoft also realizes :) 
Other Comments
  • -8 Hide
    techcurious , October 2, 2012 10:09 PM
    A little off topic, but something just occurred to me.. that we had Safari on Windows, but not IE on OSX..
    Now that OSX market share is increasing, I wonder if Microsoft will ever release Internet Explorer for OSX, and I wonder if Apple will allow it..
  • 11 Hide
    techcurious , October 2, 2012 10:17 PM
    ... nevermind.. I would rate myself down if I could.. cause after further thought, I realized that anyone that chose to use OSX over Windows will never choose to use IE over Safari or Firefox.. Something I am sure Microsoft also realizes :) 
  • 6 Hide
    dalethepcman , October 2, 2012 10:21 PM
    My one complaint about this article, was the contrast between the massive amount of iOS browsers, and the utter lack of Android browsers. You stated yourself.
    Quote:
    Each of these third-party iOS browsers are essentially just a new GUI and additional functionality added to Safari


    Why not also test firefox, dolphin, skyfire and stock browsers on android?
  • 1 Hide
    dalethepcman , October 2, 2012 10:28 PM
    techcurious... nevermind.. I would rate myself down if I could.. cause after further thought, I realized that anyone that chose to use OSX over Windows will never choose to use IE over Safari or Firefox.. Something I am sure Microsoft also realizes


    Actually I have many Mac users that would use IE in OSX just for the convenience of pass through authentication, but since its not available they all have a separately purchased copy of windows to run in a VM or access IE through Citrix for domain resources.
  • -1 Hide
    adamovera , October 2, 2012 10:51 PM
    dalethepcmanMy one complaint about this article, was the contrast between the massive amount of iOS browsers, and the utter lack of Android browsers. You stated yourself.Why not also test firefox, dolphin, skyfire and stock browsers on android?

    I attempted to, they either wouldn't run the test at all, or they hang indefinitely and are unable to complete it. Stay tuned for the Android Web Browser Grand Prix for the full benchmark results of browsers on that platform.
  • 1 Hide
    tipoo , October 2, 2012 10:53 PM
    I agree with Mozilla, perceived speed > benchmarks for browsers. And ironically that's exactly where they fail hardest. Chrome, Opera, heck sometimes even IE10 now always seem more responsive and stay more responsive than Firefox in my experience. I like its font rendering, I like its smooth scrolling (well, IE10 has those too, I think it has to do with DirectWrite more than the browser) but the small instances of UI lag bug me after using Chrome for so long.
  • 1 Hide
    tipoo , October 2, 2012 11:03 PM
    Once more I'm puzzled why browsers in OSX are consistently and significantly slower than Windows and Ubuntu, even the same browser cross platform.
  • 5 Hide
    adamovera , October 2, 2012 11:25 PM
    tipooOnce more I'm puzzled why browsers in OSX are consistently and significantly slower than Windows and Ubuntu, even the same browser cross platform.

    From page 4: "Our current cross-platform test system provides unusually low Web results under OS X Mountain Lion compared to other operating systems. While the OS X browser scores appear to be accurate in relation to each other, none of the OS X scores should be used to draw conclusions about OS X versus the other desktop environments in this test. Until we can pin down the culprit, please view the OS X results as if they were obtained on an entirely different test system."
    I have been unable to track down the cause of this problem - I tried every single network driver I could find, multiple re-installs, and different DSDT files. This is the only Hackintosh system we've ever used that has this issue - our older Lynnfield-based rig didn't. Hopefully, when I build a totally new Ivy Bridge-based rig this problem will just go away, if not, I guess I need an actual Mac - but that could leave Linux twisting in the wind since Bootcamp is just for Windows - on paper, anyway :( 
  • 0 Hide
    tipoo , October 3, 2012 12:55 AM
    Got ya, I missed that part. Maybe a commenter with a mac can give it a quick run to see how it compares to the hackintosh.
  • 2 Hide
    adamovera , October 3, 2012 1:37 AM
    tipooGot ya, I missed that part. Maybe a commenter with a mac can give it a quick run to see how it compares to the hackintosh.

    They have, and they're all reporting Web-related scores are higher on Mac's with lower-end hardware than our test system. Also, our old Lynnfield-based system shows OS X browsers doing way better in relation to Windows browsers on that Hackintosh versus a genuine MacBook Air - so it's definitely our current Hackintosh configuration and not OS X to blame for the lower scores versus other OSes.
  • 0 Hide
    CaedenV , October 3, 2012 3:25 AM
    adamoveraFrom page 4: "Our current cross-platform test system provides unusually low Web results under OS X Mountain Lion compared to other operating systems. While the OS X browser scores appear to be accurate in relation to each other, none of the OS X scores should be used to draw conclusions about OS X versus the other desktop environments in this test. Until we can pin down the culprit, please view the OS X results as if they were obtained on an entirely different test system."I have been unable to track down the cause of this problem - I tried every single network driver I could find, multiple re-installs, and different DSDT files. This is the only Hackintosh system we've ever used that has this issue - our older Lynnfield-based rig didn't. Hopefully, when I build a totally new Ivy Bridge-based rig this problem will just go away, if not, I guess I need an actual Mac - but that could leave Linux twisting in the wind since Bootcamp is just for Windows - on paper, anyway

    we saw a similar scoring issue with the 'web browser grand prix mac circuit'. If I remember correctly you were running on a Mac, and Win7 in bootcamp that time, and win7 still beat out the OSX scores, granted not by a huge margin.
  • 0 Hide
    adamovera , October 3, 2012 3:41 AM
    CaedenVwe saw a similar scoring issue with the 'web browser grand prix mac circuit'. If I remember correctly you were running on a Mac, and Win7 in bootcamp that time, and win7 still beat out the OSX scores, granted not by a huge margin.

    Yeah, Windows browsers (on a PC or a Mac) have so far trumped OS X browsers, but not by anywhere near the margins we're seeing on this Hackintosh. Gauging from past articles, the actual OS X scores should be somewhere between Windows 7 and Ubuntu (with Safari essentially being the big in-between) - and that order reflects the amount of resources developers allocate to these OSes.
  • 0 Hide
    Cryio , October 3, 2012 1:34 PM
    It breaks my heart when I see in the benchmark listed Opera 12.0x ... since it's basically the WORST release Opera ever had.

    Any, none of this browsers are on an equal foot.

    Internet Explorer Metro in Windows 8 X64 is running the IE10 in x64 mode, opposed to IE10 desktop which works ONLY in x86.

    The future release of Opera 12.10, will be faster on x64 than on x86, because of different coding and different SS2 instruction support. So choose carefully in the future what you benchmark.
  • 0 Hide
    LordConrad , October 3, 2012 6:13 PM
    You include Android as a test bed, but the Included videos won't play on Android Jellybean. Very stupid.
  • 0 Hide
    adamovera , October 3, 2012 10:15 PM
    CryioIt breaks my heart when I see in the benchmark listed Opera 12.0x ... since it's basically the WORST release Opera ever had.Any, none of this browsers are on an equal foot.Internet Explorer Metro in Windows 8 X64 is running the IE10 in x64 mode, opposed to IE10 desktop which works ONLY in x86.The future release of Opera 12.10, will be faster on x64 than on x86, because of different coding and different SS2 instruction support. So choose carefully in the future what you benchmark.

    Opera 12 is perhaps the most disappointing version of Opera ever due to the high expectations it carried, but I'd have to say that the version with a banner ad back in the day was the worst ;)  We benchmark stable browsers only (or RTM in the case of IE) and that's been the policy forever. I realize that IE10 Metro is x64, but in the original RoboHornet test we included both the 64-bit and 32-bit versions of IE9. I'm not sure what you wanted me to do here: exclude Opera, or make an exception for that browser and run a pre-release version?
  • 0 Hide
    meltbox360 , October 4, 2012 4:30 AM
    Hmm. Ran this on my playbook. Super smooth background and finished in 77.107 so... BlackBerry Playbook browser is the fastest mobile browser.

    Edit: According to this benchmark. Although it scores 30.something on the normal robohornet. Pretty robust and quick, has not had an issue running anything yet...
  • 0 Hide
    Cryio , October 4, 2012 11:17 AM
    @Adamovera: You aren't guilty of anything, my friend. It's just Opera version 12 that saddens me. And seeing that you brought it up, since I find v12.0x to be actually THAT awful and unsable (versions 8 to 11.6x tear right through it), a pre-release version here would have helped.

    A last thing.... isn't RTM=gold=stable?
  • 0 Hide
    Cryio , October 4, 2012 11:18 AM
    unusable*
  • 0 Hide
    adamovera , October 5, 2012 12:32 AM
    meltbox360Hmm. Ran this on my playbook. Super smooth background and finished in 77.107 so... BlackBerry Playbook browser is the fastest mobile browser. Edit: According to this benchmark. Although it scores 30.something on the normal robohornet. Pretty robust and quick, has not had an issue running anything yet...

    Whoa, that's interesting. Can you send me a screenshot of the results for that device - just click my name in the byline to email me.

    Also, if anybody out there has one of those HP WebOS pads and has run the tests, I'd be interested in seeing the results for that, too.
  • 1 Hide
    adamovera , October 5, 2012 12:38 AM
    Cryio@Adamovera: You aren't guilty of anything, my friend. It's just Opera version 12 that saddens me. And seeing that you brought it up, since I find v12.0x to be actually THAT awful and unsable (versions 8 to 11.6x tear right through it), a pre-release version here would have helped.A last thing.... isn't RTM=gold=stable?

    I hear ya, man. I held one of the WBGPs to include Opera 12 (I think it was late by over a month), and it finally arrived sans all the good stuff :( 
    Yeah, when it comes to MS/Apple software the RTM/Golden Master are as good as final (but there is always the possibility of massive day one patches which could change results). I wish Linux distros had an equivalent, in my experience the final RC builds just aren't final.
Display more comments