Firefox 7 is released and its back to re-capture those early days when it used to be among the leaders. Lets see how they bechmark against each other for various tests:

Test system: Tests were conducted on Core i7 Macbook 6gb DDR3 RAM, Windows 7 & OSX Lion.

Matrix: A Relative matrix has been used to make graphs easy to read.

Benchmark Tests:

• SunSpider Tests
• Generic Javascript operations
• DOM / CSS operations
• Startup time

Performance Results:

Overall performance: Chrome 14 vs Opera 11.5 vs Firefox 7 IE9

Firefox seems to have recovered from the times where it used to be a low performer. Chrome is still holds the top position overall, on Windows & Mac. However, Chrome 14 was slower than Opera on Windows.

We write about Google, Twitter, Security, Open Source, Programming, Web, Apple, iPhone, Android and latest in Tech @geeknizer on Twitter or by subscribing below:

IE9 vs Chrome 10 vs Firefox 4 Benchmark

Platform: Windows 7 32bit on a Intel i7 Macbook Pro, 4GB RAM.

We are not testing 64bit since IE9 has poor performance on 64bit as it still uses older javascript engine, while IE 9 32-bit is using the newer, more efficient Chakra JIT.

Benchmark Results:

V8 Benchmark results

Sunspider Benchmark Results:

Kraken Benchmark Results

Peacekeeper Benchmark:

Chrome 10 wins in 2 tests and Firefox 4 does in one, and IE & Chrome tie in sunspider test.

Browser Verdict

I really don’t think that JavaScript performance is an issue any more, its fairly fast enough to be handled by users. What matters more is better support for modern technologies like HTML5. When it comes to the support for modern technologies, Firefox & Chrome beat IE9 out of the scene. There is a huge set of features tha IE9 needs to learn from Chrome 10 & Firefox 4:

Application Cache (offline), Web Workers (threads in JavaScript), HTML5 Forms (validation mechanism, CSS3 selectors), JavaScript Strict Mode , ForeignObject (embed external content in SVG), SMIL Animations (SVG animations), File API, WebGL (3D), CSS3 Transitions (for animations), CSS3 Text Shadow, CSS3 Gradients, CSS3 Border Image, CSS3 Flex box model, ClassList APIs, FormData, HTML5 History API, Drag’n Drop from Desktop.

We write latest and greatest in Browsers: Chrome, Opera, Firefox, IE, and latest in Tech @taranfx and Facebook:

Mozilla has let loose the JaegerMonkey engine, enabled hardware graphics acceleration on Windows and Mac. Whats more is the exciting new integrated Sync with the Panorama/”Tab Candy” that looks great.

Mozilla says their JaegerMonkey compiler, combined with their other improvements, brings new bar of performance to Firefox 4 being powered by a great  page rendering engine. Their chart of three benchmark tests, shows what looks like some serious improvements over Firefox 3.x:

In addition to the raw code and horsepower additions, Mozilla also notes the inclusion of hardware-based graphics acceleration in this beta, along with more obvious integration of previous experiments like Sync, Pannorama, and App Tabs.

Watch the performance improvments in the video:

Firefox 4 is available for Windows, Mac, Linux, go ahead and download it.

For latest Google, Open Source, iPhone, Android, Tech news @taranfx on Twitter or subscribe below:

The new 12-core Opteron is code-named “Magny-Cours” would fall under AMD Opteron 6100 Series.

The multi-socket processor features a dual-die each featuring 6 cores with total of 12 MB L3 cache, built using 45 nm fabrication technologies.

The Technology

AMD has pumped up a lot of new features including 4 memory channels, HyperTransport technology 3.0, a fourth HyperTransport technology link (for better processor-to-processor communication in 4P servers), and new power management features that enables AMD to double the cores, and achieve up to double the performance, without having to double the power or the cost.

Also, AMD claims that along with the increase of chip count, they have upgraded their direct connect architecture to improve CPU-to-CPU chip communication speeds by 33%.

Performance Benchmark

Instead of using another tool, AMD used the same tool as Intel for benchmarking their Processors– SPECint_rate2006 — for showcasing performance improvements.

Integer Performance: AMD Opteron Model 6174 processor vs. 2P Xeon X5680

Floating Point performance: AMD Opteron Model 6174 processor vs. 2P Xeon X5680

The new Opteron is 21% faster than Intel’s.

And these tests were not with “Highest Performing” AMD Opteron chips, but rather cost-effective ones.

Power consumption

AMD has done some magic around power efficiency. They have doubled cores and stayed in the same power/thermal range compared to previous generations, and therefore they beat Intel’s 130watt  by 80W ACP part

Pricing

AMD’s medium-high notch (read cost-effective) processor (which performs way better than Intel) is 42% cheaper than Intel’s. ($1663 vs. $1165)

[via AMD blog]

We write on Hardware, Open Source, Programming, Tech news, gadgets,  Get them @taranfx on Twitter

But it’s time to change it all. Microsoft has built the new browser ground up and lead innovation to let the industry to follow. IE9 is the First browser to support full GPU acceleration (Direct 2D) Rendering which means your graphics card is now put to good use. Earlier to this Firefox 3.7 nightly showed some GPU rendering which is pretty limited.

Our early impressions of IE9 were good, but it has gone better.

Rendering the web page in most browsers including IE8 is performed on the CPU. Direct2D (a new Windows API in Windows 7) uses hardware Direct3D acceleration to accelerate 2D graphics. This would be available as a patch for Windows Vista. Through the use of Direct2D, IE9 performs all graphics rendering on the GPU, providing quicker page rendering, faster, smoother animation, and high quality image scaling, resulting in richer experience.

Download IE9

Performance Benchmark

The New engine used sounds exactly like V8 engine used in Chrome. On the Inside, the new IE9 JavaScript engine will compile JavaScript into native code, just like in V8, and the technique it will use to speed up the object-oriented nature of JavaScript also sounded similar to the approach V8 takes.

We don’t go by the words of the manufacturer, we like to test it on our own. Today Microsoft announced a Preview of IE9, we got our handson and reviewed it. We used SunSpider benchmark to get Real world-alike Javascript performance results and Some intense graphics Based applications like Bing maps, google Maps and some of the demos hosted by Microsoft itself.

SunSpider Benchmark

Results:  My eyes popped out, IE9 seems to have some great improvements:

Clearly, IE9 has come very close to Safari  and beats all versions of Firefox. It would be some time till it touches chrome’s mark (may be in the final version) but the results are brilliant so far.

Graphics Framerate (FPS)

Era of Rich graphics on web just started over last few years with beginning of Maps applications and some fantastic animations in various webapps. We took some of the most common ones: google Maps, Bing Maps and few demos hosted by Microsoft for IE9 and then average framerate over all of these and results were eye-opener:

IE9 sweeps of all browsers in this area. The GPU acceleration is way to mature than in Firefox 3.7. And Opera 10.5 does some decent job thanks to it’s Vega graphics and Carkan engines.

Update: IE9 vs. Chrome 10 vs. Firefox 4 Benchmarks - on Core i7 machine

We write latest and greatest in Browsers: Chrome, Opera, Firefox, IE, and latest in Tech @taranfx and below:

Opera claims it’s latest offering, Opera 10.5 Stable,  to be the fastest Browser on Earth for Windows. But the same is true for every platform.

Update: Opera 10.5 Stable for Windows Released at Opera.com

Here is the summary of stuff that’s new with Opera 10.5. Official benchmarks Opera revealed a while ago:

But Opera didn’t Compare the SunSpider results to Chrome 5, which is the fastest browser. So we decided to do  our own benchmark, and the results were surprising. And the Winner is:

TEST                   COMPARISON            FROM                 TO             DETAILS
=============================================================================
** TOTAL **:           1.29x as fast     1649.8ms +/- 2.1%   1276.2ms +/- 2.3%     significant
=============================================================================
  3d:                  1.58x as fast      259.8ms +/- 4.7%    164.0ms +/- 6.6%     significant
    cube:              1.86x as fast       93.6ms +/- 6.3%     50.4ms +/- 6.4%     significant
    morph:             1.77x as fast       97.6ms +/- 4.6%     55.2ms +/- 2.5%     significant
    raytrace:          -                   68.6ms +/- 13.8%     58.4ms +/- 15.1%
  access:              1.35x as fast      191.0ms +/- 4.1%    141.2ms +/- 2.6%     significant
    binary-trees:      *1.38x as slow*     12.2ms +/- 36.3%     16.8ms +/- 3.3%     significant
    fannkuch:          1.12x as fast       76.8ms +/- 7.4%     68.6ms +/- 1.6%     significant
    nbody:             2.07x as fast       77.8ms +/- 6.5%     37.6ms +/- 12.3%     significant
    nsieve:            1.33x as fast       24.2ms +/- 12.3%     18.2ms +/- 3.1%     significant
  bitops:              3.51x as fast      175.6ms +/- 7.1%     50.0ms +/- 9.1%     significant
    3bit-bits-in-byte: 3.12x as fast       20.6ms +/- 6.9%      6.6ms +/- 10.3%     significant
    bits-in-byte:      4.38x as fast       36.8ms +/- 4.4%      8.4ms +/- 8.1%     significant
    bitwise-and:       7.23x as fast       56.4ms +/- 11.9%      7.8ms +/- 7.1%     significant
 --Truncated for Brevity ---

The Result: Opera 10.5 is blazing fast. It’s 29% Faster than Google Chrome 5.

Update: Opera 10.5 Stable on Windows is 7% faster than Beta. Damn, Chrome is way behind

What’s New:

1. Performance — Vega Graphics,  Carkan & Presto: The 3 secret power components.

Opera’s SVG library now becomes throttled engine for Opera’s  graphics throughout. This would bring Graphics acceleration based on GPU like IE9 (Internet explorer 9), in future.

The Speed difference can be felt easily. In benchmarks, it’s alot faster than Firefox and even beats that of Chrome 5.

The faster rendering secret is the newer rendering engine (Presto 2.5.x). Long anticipated Carakan is here to do it’s fast Javascript execution. It’s more than 10x faster in SunSpider than Opera 10.10 with Futhark on Windows (Mac optimization is not as far along).

2. Better search

Whether it’s web search or searching inside the page, its better than before.

3. Better UI – More Glass

Inbuilt page search, web search, URL bar, tab stacking sidebar,  have been modified for easier and snappier navigation.

4. Full Windows 7 Integration
Opera 10.5 takes full advantage ofWindows 7’s Superbar. By default, the jump list displays your Speed Dial pages. What’s more is that you can pin pages as well, though right now it only works with Speed Dial items — so pinning isn’t all that useful just yet.

Per-tab thumbnails, on the other hand, work beautifully.

5. HTML5: After Firefox and Chrome, Opera 10.5 finally supports HTML5 natively. It uses an updated rendering engine known as Presto. Presto 2.5 brings new HTML5 features including support for CSS3 Transitions.

6. Per-tab private browsing: It’s pretty similar to what we’ve seen with other browsers: browsing and download history are not saved. Everything which happened in the tab is forgotten about as soon as you close it. One big plus is that you don’t need to open an additional window, making Opera’s implementation just a little bit more inconspicuous.

No matter they occupy a small share of Web browsers, but their innovation is sky-high. Waiting on Opera on iPhone.

Download: Opera 10.5 Stable for Windows, Beta for Mac OS,  Linux

We write latest and greatest in Firefox, Chrome, Tech Guides, Open Source, Apple, iPhone, Android, Open Source, Latest in Tech @taranfx on Twitter.

The list of changes so far is fairly small for PC users. But the javascript performance improvement is again noticeable. On the contrary, Mac users will notice a bit more of a difference, with fixes addressing plugin stability and crashing while dragging tabs, a slightly improved cookies manager, and minor tweaks to Chrome’s task manager.

On the performance ( the thing that matters most) Chrome 5 speeds-up Javascript execution by upto 22% as per the Sun Spider Benchmark done on Windows 7:

This Makes Chrome 5 dev “World’s Fastest Browser”.

Detailed Results:

TEST                   COMPARISON            FROM                 TO             DETAILS

=============================================================================
** TOTAL **:           1.22x as fast     1116.6ms +/- 1.6%   914.2ms +/- 2.3%     significant
=============================================================================
  3d:                  1.14x as fast      169.8ms +/- 8.9%   149.0ms +/- 9.9%     significant    cube: +/- 15.7%    54.8ms +/- 19.9%     morph:             -                   60.4ms +/- 18.9%    55.6ms +/- 21.8%     raytrace:          1.34x as fast       51.6ms +/- 5.3%    38.6ms +/- 14.5%     significant
  access:              -                   98.6ms +/- 1.9%    95.4ms +/- 6.5%     binary-trees:      -                    4.6ms +/- 36.3%     4.6ms +/- 52.7%     fannkuch:          -                   40.0ms +/- 7.9%    39.0ms +/- 9.3%     nbody:             -                   42.8ms +/- 13.2%    41.0ms +/- 5.3%     nsieve:            -                   11.2ms +/- 24.1%    10.8ms +/- 9.6%
  bitops:              1.18x as fast       95.8ms +/- 13.1%    81.2ms +/- 12.6%     significant  3bit-bits-in-byte: ??                   7.0ms +/- 12.6%     8.0ms +/- 45.3%     not conclusive: might be *1.14x as slow*    bits-in-byte:      -                   16.8ms +/- 9.6%    16.0ms +/- 11.0%     bitwise-and:       -                   29.8ms +/- 23.9%    27.4ms +/- 20.2%     nsieve-bits:       1.42x as fast       42.2ms +/- 20.3%    29.8ms +/- 13.0%     significant
  controlflow:         -                    6.4ms +/- 17.4%     6.0ms +/- 41.4%     recursive:       -                    6.4ms +/- 17.4%     6.0ms +/- 41.4%
  crypto:              1.24x as fast       68.6ms +/- 5.4%    55.4ms +/- 7.9%     significant    aes:               -                   23.6ms +/- 12.7%    21.8ms +/- 14.8%     md5:               1.28x as fast       23.0ms +/- 13.2%    18.0ms +/- 15.4%     significant    sha1:              1.41x as fast       22.0ms +/- 13.3%    15.6ms +/- 21.5%     significant
  date:                1.27x as fast      125.4ms +/- 4.1%    99.0ms +/- 6.8%     significant    format-tofte:      1.34x as fast       61.6ms +/- 8.9%    45.8ms +/- 18.7%     significant    format-xparb:      1.20x as fast       63.8ms +/- 5.4%    53.2ms +/- 7.3%     significant
  math:                -                  113.4ms +/- 10.5%   101.6ms +/- 8.9%     cordic:    1.25x as fast       45.4ms +/- 15.7%    36.4ms +/- 7.1%     significant    partial-sums:      -                   49.2ms +/- 12.9%    47.4ms +/- 16.7%     spectral-norm:     -                   18.8ms +/- 3.0%    17.8ms +/- 24.9%
  regexp:              ??                  37.2ms +/- 9.3%    40.0ms +/- 15.4%     not conclusive: might be *1.08x as slow*    dna:               ??                  37.2ms +/- 9.3%    40.0ms +/- 15.4%     not conclusive: might be *1.08x as slow*
  string:              1.40x as fast      401.4ms +/- 3.3%   286.6ms +/- 5.7%     significant    base64:            1.45x as fast       46.4ms +/- 8.6%    32.0ms +/- 15.8%     significant    fasta:             1.52x as fast       62.0ms +/- 7.4%    40.8ms +/- 18.7%     significant    tagcloud:          1.21x as fast       89.8ms +/- 6.2%    74.0ms +/- 9.8%     significant    unpack-code:       1.28x as fast      130.4ms +/- 16.4%   101.8ms +/- 21.0%     significant    validate-input:    1.92x as fast       72.8ms +/- 13.8%    38.0ms +/- 10.6%     significant

Apart from performance on Windows, there are two key updates. First, Chrome will now use your default downloads folder (i.e. Username\Downloads on Vista and Windows 7)

Now user has more control over the content Chrome displays and executes. User settings for cookies, pop-ups, plugins, scripts, and images are being moved to their own panel. In addition, user can choose to hide Extension buttons from the bar.

You can download the dev channel build .

Update: Opera 10.5 Beta is the NEW Fastest Browser on Earth

We write latest and greatest in Firefox, Chrome, Tech Guides, Apple, iPhone, Android, Open Source, Latest in Tech @taranfx on Twitter.

What is HTML 5 and FAQ

Things are going to change, Microsoft at  PDC 09 told the world that they want to change the face IE had been living with. With IE9 Microsoft has promised to take care of JavaScript, web standards, and graphics technology.

Update: IE9 Preview released: IE9 vs. Chrome vs. Firefox vs. Opera Performance Benchmark

IE9 features a new,  faster JavaScript engine, which will have richer support for web standards like CSS 3, and will use the new Direct2D and DirectWrite technology for its graphics and text rendering.

The most interesting part is that Internet Explorer 9 will move graphics and text rendering to the graphics chip (GPU), using DirectX‘s Direct2D. In one of the demonstrations, which showed the difference between non-GPU and GPU rendering: Bing Maps did about 14fps without using the GPU, and up to 60fps while using the GPU. A similar difference will of course be seen for Google Maps.

Inside DirectX 11

Rendering the web page in IE8 is performed on the CPU. Direct2D (a new Windows API in Windows 7) uses hardware Direct3D acceleration to accelerate 2D graphics. This would be available as a patch for Windows Vista. Through the use of Direct2D, IE9 will perform all graphics rendering on the GPU, providing quicker page rendering, faster, smoother animation, and high quality image scaling, resulting in richer experience.

Benchmark

The New engine used sounds exactly like V8 engine used in Chrome. On the Inside, the new IE9 JavaScript engine will compile JavaScript into native code, just like in V8, and the technique it will use to speed up the object-oriented nature of JavaScript also sounded similar to the approach V8 takes.

We don’t go by the words of the manufacturer, we like to test it on our own. We got our handson on a developer preview of IE9. We used SunSpider benchmark to get Real world-alike results.

Results:  IE9 is whoopingly 2.5x times faster than IE8, and 13x times than IE7 but still lags behind ALL other browsers.

Browser Time IE 5500 IE9 1620 Firefox 3.6 Beta 3 1498 Chrome 4 (Nightly) 590 Safari (Nightly) 605

The Results claim that the current IE9 performs nearly as good as betas of Firefox, Safari, and Chrome. IE is still the slowest, but the difference now ranges from 9% slower compared to Firefox to 260% slower compared to Chrome, rather than 800% slower, which was the case with IE8.

Since this is not even a beta version, we can expect better performance in upcoming versions ,and I won’t be surprised if it beats Firefox 3.6. And as far as graphics rendering is concerned, we need some real world FPS test, but Running ChromeExperiments was no where near to Chrome.

Compliance and standards

IE was, once, leader in this area. But when Firefox took on the Browser wars, followed by Chrome and Safari, IE started looking outdated, which was the case with IE6. IE7 made some small improvements, and IE8 did some homework by giving out at least HTML 4.01 and CSS 2.1 specifications.

Today, there is Huge demand (among developers) for new features: HTML5, CSS 3, SVG, & Canvas. Many of these standards are still themselves in draft, But, their craving features like HTML 5′s native support for embedded video, CSS 3′s rounded corners, Canvas’s extensive graphical capabilities are in demand.

IE9 will surely incorporate HTML5, CSS, however there is no word on SVG, canvas. What they did mention was that the company’s focus would be on providing the features that were actually most useful to developers, and ensuring that those features had robust test suites. So that means Microsoft finally wants to work with web developers, hand-in-hand.

Vulnerabilities in HTML 5 and Future of Web Apps [GMail]

IE9 is still in its early days, and there is no hint of a date at which a beta might become available. Many questions, especially about standards support, remain unanswered. The switch to Direct2D may also mean an end to XP support (Direct2D is unavailable on that OS).

Nonetheless, the announcements give a strong indication that Microsoft is taking its browser seriously, and wants it to be a viable alternative to its competitors.

Update: IE9 vs Chrome 10 vs Firefox 4 Benchmarks

More Benchmarks:

• Windows 7 vs Ubuntu 9.10 ,
• Chrome for Mac – Browser Benchmark ,
• Browser Performance Benchmark – Firefox 3.6: 23 Percent Faster

I’ve thought long and hard about usage comparison, Windows proves out to be higher on ease of use. On the other side I have years of Windows experience and a lot less with Ubuntu but I still can work with both without any trouble.

Software:

Windows 7

Bunded software is poor, however, lots of free software are available. But most professional (and popular ones) have to be bought at a cost. (Check Windows 7 review for details)

Ubuntu 9.10

Comes complete with an excellent array of software that you will use in day-to-day life. Access to, and installing, new software is a snap using synaptics package manager and my favorite. In fact this is the same technology that is used on Jailbroken iPhones Cydiaapp, so you know how easy it is. However, on the bad side, if you are a windows user willing to migrate, you can run most windows apps but not all using Wine.

Media Support

Both Platforms have good support for Media Formats and more codecs can be added with Ease. Playing FullHD 1080p content on both gives smooth 30fps.

Hardware support

Windows 7

On the whole, Windows 7 offers excellent support for modern hardware right out of the box. You might be out of luck with old hardware

Ubuntu 9.10

Overall, hardware support is good, and getting better. On really low-end hardware you can substitute Ubuntu for Xubuntu. There are no guarantees. Newer hardware may find incompatibilities.

No matter what we try to compare, everything on both platforms, they are more or less equal. So what could be the distinguishing point? It’s Performance and that’s what we talk about in detail.

Test System

My new Sony VAIO CW16 Notebook:

Hardware:

• CPU: Intel Core 2 Duo 2.53 Ghz @ 1066Mhz FSB
• RAM: 4GB DDR3,
• Graphics: Nvidia GEFORCE GT 230M 512MB (16 stream processors) GPU

This is a great gaming notebook on budget, ideal for Desktop use.

Results:

1. Basic Tests- Winner: Ubuntu 9.1

1. Installation Time (Trivial)
2. Diskspace Used
3. Boot Time
4. Shutdown time

2. I/O Performance Tests – Winner: Ubuntu

1. File Copy from External Drive to Internal 1GB in Size
2. Bulk File copy each 10kb in size
3. 1GB File Transfer average speed over Ethernet Network 100Mbps

3. Gaming Performance, Winner: Windows 7

• Unreal Tournament (UT) 1280×720
• Call of Duty 4 (COD4) 1280×720
• Need For Speed Most Wanted (NFSMW) 1280×720

Update: Compared Native Unreal Tournament  on Ubuntu to the one using Wine, FPS increased by 6-7%, the difference between Win7 & Ubuntu is still significant. Since other games are not available natively, we’ve to compare via Wine.

Clearly, except for Gaming and Networking performance, Win7 loses everywhere. There is a huge difference in frames per second (FPS) you get on Windows 7 because of Direct X 11.

Inside DirectX 11 – The Technology in Detail

Direct X11 is Microsoft proprietary and much more mature than any other graphics acceleration standard. Hence, Gaming on Windows will always be superior.

VERDICT:

All said and done, if you need performance, Ubuntu is for you. If you are looking on Gaming, and a default good Look and feel, you have no other choice than Windows.

We write latest in Tech: Google, Open Source, Ubuntu, Windows 7, get them via Twitter updates, or RSS or below:

Recently, we have seen pre-beta builds (chromium) appearing for Linux and Mac. I must admit, I was less than impressed.  Chromium builds have been a unstable and unrealiable — UI failures, bugs, and Blue Tab of Deaths “Awww Snap!”

Even Google co-founder Sergey Brin expressed his displeasure on Web 2.0 summit.

By chance, I tried Developer beta for Chrome. Chrome (4.0.223.11 for Mac), overall, feels much more stable. I`ve been using it for around 10 days, it’s robust – A Chome 4.0 Windows kind feel.

Stability, and Overall Feel:

Chrome builds, besides being more stable than chromium, offer some additional perks, such as being able to import bookmarks, something which the Chromium builds turned off months ago for some unknown reason. Though, this only works when you first load up Chrome. The Bookmark Manager itself still does not work yet.

During the last week, I received one update, and overall it crashed only twice, despite being a power user. At all times I have like 20+ tabs opened, and system is never shutdown, max goes to sleep. So running  a browser continously for 10 days with lots of tabs, all time, is a good test for this browser. If I have to compare stability with Firefox, it’s almost the same. Firefox 3.5.3 has gives me similar no. of crashes, but Firefox dies with 30 tabs, Chrome for Mac beats this.

Themes: Chrome Themes have also appeared in this release and are visible from the Thumbnails page (in the lower right corner) just as they are in the Windows version.

Performance: It’s faster than chromium builds, but still not quite as fast as it is on Windows, but it already seems comparable in speed to both Safari, Camino.

Performance tests were done on Snow Leopard with:

Hardware: The benchmarks were done on a Intel Core 2 duo 2.4Ghz with Nvidia 9 Series and 4GB of RAM on Snow Leopard

Reference: Firefox 3.0 is held as the reference point for all tests.

Tests: Variety of benchmarking tools are employed to nail down the performance.

• NonTroppo: CSS Rendering, Page load test
• SunSpider: Javascript execution test

Results from all tests were combined to prepare Comprehensive Relative Performance Index, CRPI, Values.

Here are the CRPI scores:

Browser Benchmark on Mac - Snow Leopard

Among all the tests, Chrome did well for Graphics rendering. Even practical benchmark of ChromeExperiments does well on Chrome than Safari or any other browser.

All said and Done:

If you are starving on Mac for a snappy browser, switch to Chrome dev build. At this point it seems to have all the Pros.

Google has made it public that Chrome for Mac (stable) will be out before the end of the year. Chrome Dev does a great job, and we could be nearing beta, soon.

You can download the developer build of Chrome for Mac here.

Subscribe to Twitter updates, or RSS, join Facebook fanpage for more Tech updates.

Mozilla , next week, is rolling out first Firefox 3.6 public Beta. The Release is supposed to be the first one, with big fixes for Firefox 3.5.

My experience with Firefox 3.5 had been the worst ever. The crashes were more frequent than any other previous versions, Tab restore was almost useless, worked rarely. With large no. of tabs opened (of the order of 20), it can eat 500MB of RAM and taking your CPU to adventures of peeks.

I won’t lie, But Firefox was jealous that I prefer Chrome 4 for reliability, and for the heavy web 2.0 apps. I use Firefox for causal general purpose surfing. All-in-all, it’s the only one that can fulfill taste-buds with so many addons.

Finally, I decided to take the latest nightly build of Firefox 3.5, 3.6, 3.7 to benchmarking adventures. Since, we are pretty close to planned code freeze, the results of Firefox 3.6 nightly will be 99% similar to public beta.

The Result was something that makes me glad: 3.6 is faster at JavaScript execution and much faster at page rendering.

Benchmark:

Hardware: The benchmarks were done on a core 2 duo 2.4Ghz with Nvidia 8 Series and 4GB of RAM on Windows 7. Windows 7 was chosen because it nears in performance to XP., and has continued to grow even before the release.

Reference: Of course, like other benchmarks, IE7 is held as the reference point for all tests.

Tests: Variety of benchmarking tools are employed to nail down the performance.

• NonTroppo: CSS Rendering, Page load test
• SunSpider: Javascript execution test

Results from all tests were combined to prepare Comprehensive Relative Performance Index, CRPI, Values.

Here are the CRPI scores:

Overall, The Winner is  Chrome 4, again. Though, on Windows XP Chrome 3 takes the lead. Firefox 3.6 comes with 23% or 1/4times faster than it’s predecessor Firefox 3.5.3

The pace of speed boost for Firefox is still lower than Chrome or Safari. Safari 3 to 4 was a giant leap in performance. On the other hand, Firefox speed boost ad been slow and steady. But, what is more important is that it has continued to improve, overall.

So today, Chrome 4 is almost 15x times faster than IE7, and Firefox 3.6 is 9x times.

Chrome 4 will get performance boosts as it comes out of Dev to the mainstream and with plugins on Chrome 4, it’s gaining popularity.

Firefox 3.6 can max jump from Firefox 3.5 by 25%. But what is more awaited is the bug fixes, better stability which makes it more attractive among it’s fans.

Subscribe to Twitter updates, or RSS, join Facebook for more Tech updates.

Misconception: I don’t need caching

For a typical theme in wordpress, it needs to execute about 35 SQLs and 1000 of lines of PHP logic to yield a single page. Sure, modern servers are fast enough to do all this in split seconds. But when it comes to handling large traffics, like from Digg, Reddit,etc. it could easily prove out to be an overkill for the CPU. Put a Xeon or Core i7, just nothing would be able to handle 1000 simultaneous hits, without delays.

WP SuperCache

It’s among the bet known caching plugins out there. It caches post content, plugins and theme’s php executions firly well. Also, it let’s you compress the plain HTML in gzip for optimal bandwidth usage and faster load times. Gzip is so important that, just by enabling it, you can do wonders on saving your bandwidth. The difference is mostly of the order of 5x times or higher.

On the darker side, there are lots of pitfalls in wordpress, the most annoying one: sometimes it just won’t work. Recently, I changed my permalinks structure, and since then it started utilizing CPU higher than ever and also reduced page load times. It increased from 1 second to 10seconds – horrible.

Tried debugging several aspects, read several guides on fixing it, but nothing helped. It was time to look for alternatives.

Hyper Cache

Hyper Cache, basically, uses similar concept and does all that without touching your .htaccess.This could be good for people who have long .htaccess with large no. of mod_rewrite, other operations.

The performance was similar. With SuperCache and Hyper Cache (with gzip enabled on both) loading times were 0.6s and 0.77 seconds, respectively.

There are pitfalls too. Unlike SuperCache, which let’s you choose whether to serve cached or DB based pages to logged-in users, hyper cache defaults it to fetch from DB. On the upside, if you keep on modifying theme, like me, it always serves you the latest page without the need of clearing the cache.

On the good sides of things, It has one  additional feature: Realtime Statistics. It shows you the kind of benefit you avail on installing Hyper Cache. It draws simple graphs to represent count of how many hits did you receive and out of them how many were actually served via Cache vs. Database. It also covers the clients that didn’t support gzip compression and plain html was served for them.

Overall, Hyper Cache is great for people hosting their own WordPress, especially if you get a lot of traffic or you are on a shared hosting plan.
DB Cache Reloaded

This plugin is modified version of it’s parent: DB Cache. DB cache had some inherent bugs with tags that never got resolved. DB Cache takes a different approach to caching. DB Cache caches database queries For instance page x took 35 queries, it caches them all for the next page atleast 15 would be similar (which belong to header, sidebar, footer, plugins). Your recent posts, your tag clouds and category lists, recent comments etc, depending on what widgets and theme you use. These generally don’t change from page to page but for a normal caching plugin on a new page, they still need to be called so that the full html page can be saved.  This way you can reduce no.of hits to the database and improve speed.  DB Cache reloaded makes it even better with bug fixes.

On one side it helps w.r.t Search Engines. A normal caching plugin outperforms when a lot of visitors access one specific page. On the other side, it actually harms when one crawler accesses a lot of pages, most of them being outdated. DB cache reloaded, by caching common Database queries, can fill exactly that hole which significantly reducing the CPU overhead used to serve all the bots crawling all the pages.

You also significantly reduce the load when a bot does his daily mass crawl. While DB cache will not give you the speed hyper cache will on a single page load, it will certainly reduce  server’s resources utilization.
The Deadly Combo

First, when I thought of combining the two, it sounded crazy. I thought it might conflict the other.  But, the distinct way these two worked, encouraged me to give it a try on my local setup. In actual, they wouldn’t really conflict and can actually complement the pitfalls quite well: One is designed for max performance for users and other one worries about Google bots.

Why both still work, together.

Practically, when you run them collectively, DB Cache Reloaded takes precedence, the first time page is loaded. On subsequent page loads, it always has at least few queries cached by DB Cache Reloaded. Proof is, the cached queries increase with each time new page is reloaded. This, as a result, allows a new non-hypercached page to load quicker, since we already have few cached queries fro the new page, which is then saved into hyper cache for further visitors, Making it non-conflicting combo.

I saw the improvement in Cache Misses pages on Statistics. The missed pages were alot faster than before. (order of 4x times)

Overall, This combo has given the fastest load times ever. It enabled me achieve best of the both worlds, without comprimisng on any aspect, beyond what WP SuperCache has to offer.

Try the combo, and let me know what you think.

A lot  changed over 1.4 to 1.5. There were critical compile-time improvements, Runtime smartness and with Java 6 more performance optimizations; Java 7 – the modularity.

Let’s start discussing each aspect of Java 7 which makes it go Loud -

• Modularization – JSR 294 or Project Jigsaw
• JVM Support for dynamic languages
• More New I/O APIs which are nearly finished, includes true asynchronous I/O  and finally a real file system API – JSR 203
• The native language support for XML. (probable)
• Safe rethrow – Allows a broad catch clause, with the compiler being smarter on what you’re allowed to rethrow based on what is thrown from the try block. (I had not seen this before but it looks nice)
• Null dereference expressions – Null checks with ‘?’ syntax similar to Groovy… lettign developers avoid a nest of null checks.
• Better type inference – Example around generics instantiations, but it was not clear how far the inference would be taken (the more the better in my opinion).
• Multi-catch - Allows a comma separated list of disjunctive exception types in catch clause.
• JSR 296 – Swing application framework – It still needs to be easier to create Swing apps.

Update: As per Sun Blogs

After much discussion it’s become clear that theSwing Application Framework API as it is today hasn’t reached consensus and we feel still needs further design work done.

Since the SAF API was committed to milestone 5 of JDK7 and that time is already here, this date is now impossible, and we need to decommit SAF from any specific JDK 7 milestone.

The “Small” Sun changes are:

• Upgrade class loader architecture – Work started in Java 5 and continues to evolve. There are some deadlock issues today in classloader delegation that will be addressed.
• XRender pipeline for Java 2D – This was an Open JDK Integrators Challenge project,and is an analog to the OpenGL pipeline but much more portable across x11.
• Swing Updates – JXLayer, DatePicker, CSS styling (maybe) that Ethan Nicholaus (sp?) has been working on
• JavaFX

Update: Few new Small changes of  “Project coin” are available here.

The “Fast” changes from Sun (This refers to performance improvements):
Hotspot run-time compiler enhancements

• A couple of concurrency (JSR 166) tweaks (Better support for Multicore)
• G1 Garbage collector- Leads to much smaller pause times and hopes to replace CMS (Concurrent mark sweep) GC
• Compressed pointer 64 bit VM
• MVM-lite – Multiple Virtual Machines will help run isolated applications and allow a kill -9 on a Java application. Mark said it is not clear what problem would be solved, and original project was extremely ambitious, but desire to drag apps out of browser plugin presents a good usage and need for MVM. (This could be moved to a later release)

The Theory and Practical

As far as what changes you’ll see in your day-to-day work, my guess is that the major impact will be stuff like JSR 203 which overhauls the file system API. If JSR 310 is included, then it would also have a major impact on how you interact with any aspect of the date and time APIs. Many of the other JSRs will only impact you if you happen to already do something in that particular area (JMX – JSR 255, concurrency – JSR 166, etc).

The biggest thing most people will notice may be performance. This is my favorite. And that is exactly what I thought about writing, and it expanded the scope to features aswell. As usual, each JDK brings a whole new set of performance optimizations. We’ve already seen some very encouraging results in String performance, array performance, and a new concurrent garbage collector (G1). I suspect many people will find that their existing code will work and run noticeably faster than it did in the past.

Performance

I saw this one about the new features in Java 7:

http://www.ibm.com/developerworks/java/library/j-jtp03048.html

They use MergeSort as an example of how to exploit multiple CPUs for sorting. Java 7 has the nice feature, that it can now decide at runtime, how many threads should be used to solve a particular problem (see the coInvoke part).

However, there is this tricky constant, SEQUENTIAL_THRESHOLD, which is used to decide whether to enforce sequential processing or not. How do you set this value? Well, you set it at design time, even though the example was meant to show how Java adapts at runtime…

The next thing is that the whole array is passed as parameter. No matter what programming language you use, this is a bad design. If Java doesn’t copy the memory, you may have 2 CPUs looking at the same RAM area. If Java has a runtime optimization that detects that 2 CPUs are looking at the same area, and decides to copy the data, it will copy too much data…

I’m not sure this example would perform better on a 4-CPU machine than on a single-CPU machine with the same CPUs…

The basic problem in all this is, that it is extremely hard to find real world examples of parallelization of algorithms that can be optimized to any kind of parallel hardware. Good multi-threading must be done on a functionality level, not on the algorithm level.

Also, every time we add multi-threading to code, we make it more complex. In other words, it comes at a cost. I predict that some of the future performance gains don’t come from making algorithms more threaded, but from changing data structures, reducing memory footprint and simple optimizations. As the price of more performance increases, efforts will be spent where most speed can be gained at the lowest price.

> java -version
java version "1.7.0-ea"
Java(TM) SE Runtime Environment (build 1.7.0-ea-b66)
Java HotSpot(TM) Client VM (build 16.0-b06, mixed mode, sharing)

And the Result The Winner is … Java 7

Java 5 <=== 18% faster=== < Java 6 < ===46% faster===< Java 7

Note – This was totally based upon my tests, doesn’t necessarily means it’s for overall Java. The results may vary for different kinds of computations.

Alright, When is Java 7 coming ?

Optimus Prime has returned to save the World from The Fallen!

Microsoft lost it’s long built impression with Windows Vista. Vista is the best example of a Software development failure. It showed the world, no matter how good your hardware is, your software is crappy, you cant do anything, it will fail to perform.

Windows 7 RTM [Final] Review. Benchmark vs. XP, Vista

Microsoft learned lessons, and this time before coming up with Launch of Windows 7, they had their Home Work done properly. What does it mean for you? You will get a better OS than what you had previously. I don’t even want to compare 7 with Vista. The fact is that most corporates and home user never moved to vista or upgraded back to XP. The real question is will XP users migrate to windows 7 ? Read on to find out.

It’s stable, smooth, and well-cooked, brings-out new graphical features — a new taskbar that is inspired from Mac OS X dock, and  much improved security enhancements and device management that make it both easier to use with inherited security.

The important aspect – it won’t require the hardware upgrades unlike Vista. As buggy and irritating as Vista was, Windows 7 isn’t. Instead, it’s the brave, mature successor to Windows XP, and finally places it on competition with the best operating systems like OS X and Linux.

Windows 7 Flavors – Microsoft, like in Vista, is offering six versions of Windows 7: Starter, Home Premium, Professional, Ultimate, OEM, and Enterprise.

Our Guide: Choosing the Right Windows 7 Version [Flavor]

Platform - Windows 7 will run both on 32-bit and 64-bit systems.

Hardware Minimum Requirements –  The minimum requirements for 32bit are: 1 GHz processor, 1 GB RAM, 14 GB available hard-disk space, and a DirectX 9 graphics device. 64-bit – Atleast 1 GHz processor, 2GB RAM, 18.5 GB of free space , and a DirectX 9 graphics device with WDDM driver.

Compare that with that of Vista and XP. It’s pretty near to XP. So most PCs that can run XP, with minor or no upgrades, can run windows 7. Some users have claimed to have limited success running the Windows 7 with less than 1 GB of RAM, but that’s not recommended, though it will run. If you’re not sure if your current computer can run Windows 7, you can download and run the Microsoft Security Essentials from Microsoft.

Optional – A Touch-screen monitor is required to take advantage of the native touch features.

Installation – New and Upgrades
You can buy a new computer with the operating system already installed, upgrade from Windows XP or Vista, or do a clean install on a computer the user already owns. A Typical installation takes around 30 minutes, but that will vary depending on your Hardware.

Features:

Enhanced Taskbar and Aero Peek UI
Now this is the first thing that might attract you – All new Look – Mac OS X like feel. This is a full replacement OS, and more than just “Vista done right.” From driver support to multi-touch groundwork for the future, from better battery management to the most easy-to-use interface Microsoft has ever had, Windows 7 is surely full-baked.

The first thing that you would probably find eye-candy is the new taskbar. This is one of the best improvements. It features pinned programs using large, easy-to-see icons. Mouse over one and all windows associated with that program appear in preview. Mouse over one of those preview panes to reveal an X to close the window. Hover over the preview to show a full-size preview of the program, or click on the window to bring it to the front. Because of the button size, people with touch screens should find it especially easy to use. Though I prefer changing this to conventional takbar items where it’s wider than the default button and Grouping set off. I never liked Grouping in XP too, it makes things slower. It’s totally a personal choice.

Jump lists is another new taskbar improvement that make recently opened documents easier to get to. Right-click or left-click and drag on any program icon pinned to the taskbar to see a list of files that you’ve recently used in that program. In Media player, it shows recently played media – Music, video, photos, playlists. For IE, it shows recently visited Web sites, although it doesn’t seem to work in Firefox, yet.

Show Desktop icon got a new place- the right corner. Mouse over to the right corner. Hovering over the Show Desktop box reveals the desktop, and then hides it when you mouse away. Click on the box to minimize all windows.

Windows 7 comes with bundled theme packages, which makes it much faster to change. From the Desktop Right Click >,select the theme under Appearance and Personalization. You can customize the themes to get different colors for taskbar, etc, wallpaper rotation, sound packs. You can also download them.

Windows Media Player
Windows 7 comes bundled with WMP 12. I didn’t like it much from UI though. It’s not as easy to use as was WMP11. Aprat from that, you can now stream media files from one Windows 7 computer to another, across the Internet and out of network. Even better, the setup procedure is dead simple. A New definition to Media sharing/streaming.

Streaming - In Windows Media Player, there’s a new Stream option. On toolbar, Click Stream, and you’re presented with two choices. Both require you to associate your computer with your free Windows Live ID. When you’ve associated a second Windows 7′s WMP with that same ID, you can remotely access the media on the host computer. Windows Media Player’s mini mode looks much slicker, emphasizing the album art–sometimes at the expense of clearly seeing the controls, but it’s a definite improvement.

Another WMP improvement is the new Device Stage makes managing peripherals significantly easier, combining all your installed hardware — printers, phones, and portable media players, USB drives into one window. Device Stage support for older devices makes one of Windows 7′s best features applicable to peripherals and externals that don’t need to be upgraded. One annoying change is that Bluetooth driver support no longer comes baked into the operating system. If you need a Bluetooth driver, you’ll either need the installation disc on hand or you’ll have to go download it.

Search
Search has also improved. The Proof:- Files added to the hard drive were indexed so fast that they were searchable less

than 5 seconds later. Search result snippets now include a longer snippet, and highlight the snippet more clearly. It’s a useful feature for anybody who wants to find files faster. However, the search field is available by default only in the Start menu and in Windows Explorer, and cannot be easily added to the taskbar.

Touch screens

Touch-screen features worked surprisingly well. The hardware sometimes misread some of the multitouch gestures, occasionally confusing rotating an image, for example, with zooming in or out of the image. Overall, though, there were few difficulties in performing the basic series of gestures that Microsoft promotes, and this places Windows 7 in an excellent position for the future, as more and more computers are released with multitouch abilities.

XP mode

Probably if you are upgrading to 7, your business and home applications should run on 7. Windows 7 should run most of you apps as it is. If they fail to, 7 has a special feature to tackle this – XP Mode. XP Mode creates a virtual environment within Windows 7 that should assuage any fears of upgrading without backwards compatibility.

Security – How Secure is Windows 7 ?
The crappy troublesome User Account Control, or UAC, is back in Windows 7. But the good news is, Microsoft has tweaked the feature so that it’s less intrusive, but it’s not clear whether that means you’re actually more or less secure than you were in Vista. UAC was one of the biggest changes in Vista. It tightened program access, but did it in such a way as to frustrate many owners of single-user computers. Windows 7 provides more options for user customization of UAC.

The default setting is to notify users only when programs try to make changes to the computer, one step below the most restrictive setting of Always Notify. Under Always Notify, anytime a program tries to access the Internet, or you try to make changes to the computer, Windows 7 will require user confirmation. The second-least restrictive option doesn’t dim the desktop when UAC is activated, and will only notify the user when programs try to make changes to the computer. When the desktop dims, Windows 7 is locking it down and preventing access. Never Notify is the most relaxed option, and is only recommended by Microsoft for programs that aren’t compatible with UAC.

UAC also displays a blue banner when confronted with a program from a known publisher versus a yellow banner and exclamation point when the program is from an unknown publisher. The number of clicks it should take to use UAC safely has been reduced, However, it’s important to note that it’s a less-aggressive default posture by UAC.

A less glitzy, but no less important, change to how removable drives are handled also can impact your media. Unlike Windows XP and Windows Vista, Windows 7 will no longer AutoRun external hard drives and USB keys when they’re connected. This kills off a risky vector for malware infections that has been the bane of many security experts.

It’s possible that Windows 7 will eventually ship with additional programs, including the revamped Live OneCare antivirus and antimalware program now called Microsoft Security Essentials, but Microsoft has given no indication if that will happen. For now, users will have to download a third-party antivirus and antimalware program, although the Windows Firewall remains intact. As with many features in Windows 7 that have been carried over from Windows Vista, people will notice there’s far more granular settings control than before. Features like filtering outbound traffic, which were available in Vista but not exposed, are easier to access in Windows 7.

Performance

Windows 7 feels faster than Windows XP and Vista, but it turns out that’s not always the case, sometimes, it’s the slowest out of the three operating systems. CNET Labs, ZDnet tested: Windows 7 RTM build 7600, Windows Vista with Service Pack 2, and Windows XP SP3.

Hardware used: Core i7 quad core @ 3.2Ghz

Graphics- ATI Radeon HD, 6GB RAM

Memory management and Cache performance

Cache & Memory benchmark - Source: ZDnet

With Vista, Microsoft introduced a new technology called SuperFetch, for caching applications and speeding up boot times. This feature preloads frequently-used applications into memory, so they can be accessed quicker when they’re needed.

For conventional magnetic hard drives, most common, this technology makes sense. But if an SSD is used for mass storage, it’s better to turn SuperFetch off. The superior access times of SSDs mean they launch applications much faster than magnetic drives, so SuperFetch makes little difference.
In the RTM version (7600.16385), only Defrag is in fact inactive for SSDs — SuperFetch and ReadyBoost start just as they would with a magnetic disk.
The SuperFetch feature in Windows 7 differs significantly in approach and cache usage from its counterpart in Vista. Under Vista, the caching of applications starts immediately at launch. As the graph below shows, after three minutes just over 1GB of memory has been allocated. In Windows 7, SuperFetch starts after five minutes and after 10 minutes a little more than 600MB has been allocated. By that point, Vista’s SuperFetch has allocated more than 1.5GB.
Even without SuperFetch turned on, Windows 7 makes fewer demands on cache. For operating system-related functions, it uses 333MB, while Vista without SuperFetch uses 519MB of cache.
Windows 7 clearly makes do with fewer resources, so its cache usage is significantly lower than Vista’s.

Overall Performance: PCMark Vantage Benchmark

Benchmarking the applications that are integrated into Vista and Windows 7.

PCMark score: longer bars are better source: zdnet

The benchmark is based on several usage scenarios, with the default PCMark Suite simulating everyday PC usage. Here’s a key areas of the tests:

Memory utilization, TV, Movies, Gaming, Music, Communication, Productivity, HDD

More information about PCMark Vantage benchmark is available in this PCMark white paper.

Roughly, every application will work 20 percent faster under Windows 7 on average, bare minimum.

Windows 7 performs better than Vista and is faster than XP. Although XP is still more capable for devices with limited/outdated CPU, memory and graphics.

Faster system startup and shutdown, improved parallel processing(specially for Multi-core), and faster loading of drivers and operating system components.

Enterprise users will observe the faster login to a domain.
Windows 7 is more cache-efficient because of improved display drivers. Even with multiple open windows, the memory usage of the Desktop Window Manager, DWM remains same.  This is so because the video card’s memory taking on the load of opening of additional windows. Fact is, under Windows 7 this load is half of Vista. Direct2D, further speeds up 2D graphics rendering. For graphics cards containing ATI and Nvidia chips, this is not an issue. However, these drivers are not yet available for older Intel graphics chipsets and pretty helpful for them.

Users can employ the built-in Windows 7 DirectX 11 interface to access the graphics power of the GPU using appropriate graphics hardware – useful to those who provide video-encoding tools.
In Windows 7, Microsoft has succeeded in providing an OS that’s likely to meet the performance requirements of consumers and business users alike. The early signs are that Windows 7 will enjoy a much better take-up than Vista.

Windows 7 looks like the operating system that both Microsoft and its consumers have been waiting for. By fixing most of the perceived and real problems in Vista, Microsoft has laid the groundwork for the future of where Windows will go. Windows 7 presents a stable platform that can compete comfortably with OS X, while reassuring the world that Microsoft can still turn out a strong, useful operating system.

For Pricing, follow here.

Subscribe to Twitter updates, or RSS, join Facebook fanpage for more Tech updates.

It is a Cross Platform – Windows/Mac/Linux/Embedded Linux/FreeBSD – Lightweight web browser uses the same rendering engine as Google Chrome and Safari, but works on almost any platform. Arora uses the QtWebKit port of the fully standards-compliant WebKit layout engine.

Apart from the must-have features, Arora also Features that browsers like Firefox and Chrome do:

• blazing fast startups
• smart location bar , session management
• privacy mode
• download manager
• Set of tools for web developers – Debugging
• 30 Regional Languages

The browser offers most of the features of mainstream browsers, including private browsing, session management, and a smart location bar—all in a lightweight browser that opens nearly instantly. It’s unlikely that Lifehacker readers will leave their precious Firefox behind for this, but it’s a nice, functional, and fast browser that’s definitely worth a look—especially for Linux users interested in a functional WebKit browser with a private browsing mode.

In my testing on Windows, Mac, Linux the browser worked without a problem, blazing fast startups, and rendered pages extremely fast (just like Chrome 3).

Tot test the real world scenario, I did some Benchmarks for Chrome 3 (Fastest browser yet on Windows) vs. Arora.

The benchmark was based on FutureMark PeaceKeeper -

Higher is better – Epiphany, Midori runs only on  Linux.

Results: Midori is fastest on Linux, and Chrome 3.0 is fastest on Windows followed by Arora.

Arora kicks Firefox’s ass badly. The REAL Fact is - Gecko cannot hold a candle to WebKit when it comes to performance. And mercifully, WebKit brings decent browser performance to those of us who have been suffering on Linux under the growing weight of Firefox. Though I love my Firefox extensions, I will happily bid them farewell in favor of the speed Midori and Epiphany are currently delivering.

I was delighted to see the results, Arora.

Site:http://code.google.com/p/arora/

Download

JavaFX is built on top of a mature Runtime that gives developer’s amazing features. Java being in market since several years, they now have a more sophisticated API.

Update: This Blog post is Winner of Java Blogging Contest.

So what is JavaFX Hype all about?

I first got to hear about JavaFX more than a year back when I attended Sun Tech days. It wasn’t very mature at that time and failed to catch my attention. It was this year in Feb, when I saw a momentum and maturity in this platform, which encouraged me to take my steps further.

JavaFX gives developers an amazing feature-Rich API which will help them build scalable, fully integrated, Rich Graphics, Animated applications for the web. The beauty of JavaFX lies in the fact that its base (JVM) is already installed on all computers(90%).

When I first saw the API, I was amazed that it had so much pre-built into the API. e.g. All the common transitions like fading, scrolling, perspective animations are built-in. All you need is the idea and you can be all set to make the most powerful web application.

Being a Java at heart, you can always integrate it with existing java applications. Imagine a Live graphic Rich market-watch client talks directly to a EJB server to fetch live quotes and prepare animated graphs of statistical data and analytics to help you invest for future.

The feature I loved the most was, “applications are draggable to desktop”. Since apps run in JVM of local machine, though they open in browser, they can be dragged and dropped to desktop without interrupting any operation or performance.

Here is What I intend to Cover:

1. Example: How to Code Web Gallery in Javafx

i. Basic Overview

ii. The Code

a. Brief on animation Terminologies

b. Basic Code.

c. Transitions.

2. Benchmarking

i. Hardware used.

ii. Invocation and Startup

a. Observations

b. JavaFX vs. Flash

c. Conclusion

iii. Animated Transitions and Full Image load:

a. Key Observations

b. Conclusion

3. Verdict

i. Pros and Cons.

Let’s start with an example.

1. Example: How to Code Web Gallery in JavaFX

Before we start, you can have a Quick look at what we are going to achieve over here. It’s a decent web gallery inspired by one of them demoed on Javafx site. I took that gallery and re-wrote most of it to achieve the following:

1. Redesign UI to match my taste and site’s theme.

2. Optimize the code to max possible.

I did the latter to Benchmark JavaFX’s capabilities, as described in a later section. For now let’s look at the basics of JavaFX application design. Before we start, Let’s look at what did I make.

The Gallery is a JavaFX Frontend GUI with a backend supported by java Flickr API. On Load, Flickr is queried with REST API as http://api.flickr.com/services/rest/?method=flickr.people.getPublicPhotos&user_id=24847938@N07&api_key={apiKey}&per_page={thunbLayout.count}

HereI am sending request to Flickr to return me all public Photos for the user_id specified and per page show me {thunbLayout.count} many results per page using my unique API key.

This metaData when loaded, loads the 1st page of thumbnails. On selection via keyboard or mouse click on a pic, it opens for medium size and fizes the aspect ratio on load. While image is loaded, it is animated with Fade transition effect built into JavaFx. Let’s look at the code and things will get more clearer.

ii. The Code

Before we start, lets get you familiar with Jargons.

a. Basic Animation Terminologies:

Like Flash, JavaFX uses the generalized concept of Animations by using standard animation industry jargons. The First thing to learn is using Stages and Scenes. Stage is the main/core Placeholder for your JavaFX application. Scene is an independent widget which fits into Stage. Multiple Scenes can be put inside stage but there will be only one stage per application.

b. The Basic Code:

JavaFX made it really easy to code. Here is the code syntax:

[code lang="javascript"]
var stage = Stage {
title: "Taranfx Gallery"
width: 840
height: 520
visible: false
style: StageStyle.TRANSPARENT
scene: bind scene
}

function run() {
initUI();
stage.visible = true;
// Set focus to background by default
bgRect.requestFocus();
}

[/code]

Here, run() Method is the default main method inside Main.fx class. So setting stage to visible, and getting the focus, does the job!

Now Let’s create scene for our thumbnails:

[code lang="javascript"]

var scene : Scene = Scene {
width: 840
height: 520
content: Group {
content: bind stageContent //defined below inside initUI()
clip: Rectangle {
width: bind layout.width
height: bind layout.height
arcWidth: 20
arcHeight: 20
}
}
fill: Color.TRANSPARENT
}; [/code]

And Instantiate the UI components:

[code lang="javascript"]

function initFrontUI() {

descTxt = Text {
x: bind layout.thumbX
y: bind layout.descTxtY
font: bind layout.descFont
wrappingWidth: bind (layout.thumbGroupW - 10)
content: bind description
fill: Color.WHITE
smooth: true
textOrigin: TextOrigin.TOP
clip: Rectangle { // calculations for layout
x: bind descText.x
y: bind (descText.y - 2)
width: bind (layout.width - 20)
height: bind ( layout.thumbGroupY - descTxt.y - 17)
}
}

stageContent = [
backRect, titleBar, nxtButton, bckButton, exitButton,
descTxt, thumbGroup, pageBttnGroup,
fullView, progressBar,fiv.saveBttn //names are self explanatory, last one is save button inside Scene Full Image view.
];

getImageData(); // Call Flickr REST API using standard Java Flickr API. Loads list of thumb image src, description, etc.
}[/code]

c. Transitions:

Now look at how Transitions are created. To achieve the animation effect, you have to play with the javafx.animation.Timeline, javafx.animation.KeyFrame. Again, these are generic words used in animation industry (same with Flash). If you are new to animation, let me define it for you. Timeline is the time scale of the animation which consists of number of frames through which animation will move. KeyFrame is the frame with concrete data i.e. Initial or final state of the animation and in-between the frames happens the transition that creates animated illusion. The animation used, when you select a thumbnail to view fullImage, is demonstrated below:

[code lang="javascript"]

var tL:Timeline = Timeline {
rate: bind tLRate with inverse
keyFrames: [
KeyFrame {
time: 1s
values: [ opacity => 1.0 tween Interpolator.LINEAR ]
canSkip: true
}
]
};

...

function fadeEffect() {

if(visible) { // javafx.scene.Node.visible, if current stage is visible, its set true.
tL.time = 0s;
tLRate = 1.0;
opacity = 0.2;
visible = true;
} else {
tL.time = 2s;
tLRate = -2.0;
opacity = 1.0;
}
tL.play();
}

[/code]

Animations are so easy to code. I just loved it.

I don’t trust any hypes, statistics till I test them on my own. Proceed with the next section and you will know what convinced me here.

.

2. Benchmarking the JavaFX (Gallery example):

i. Hardware used:

Notebook: Dell XPS

Processor: Intel Core 2 Duo 2.2Ghz @800Mhz FSB

RAM: 2.5GB

Graphics: Standard Intel extreme

(I have a Nvidia notebook as well but I wanted to test it on the onboard graphics to check how bad can it get.)

Tools: Netbeans 6.5, VisualVM 1.1

ii. Invocation & Startup:

JavaFX startup is pretty heavy for initial 10 seconds. During this interval, It is supposed to load a large number of underlying classes.

The main loader is com.sun.javafx.runtime.main.Main.class, packaged inside the distributable Jar. In a web browser, actual invocation starts from the javascript from URL “http://dl.javafx.com/1.2/dtfx.js”.
This Javascript looks for “Javafx” method inside the HTML <script> tag. In this tag, information about the “JAR to load” is present. JAVAFX runtime Main.class takes over the control and loads the Projectname_browser.jnlp which invokes the actual jar user classes i.e. com.taranfx.* in this case.
Before loading your Main class i.e. com.taranfx.Main.fx, it loads all the base underlying classes to support the Javafx features.

a. Observations:

My application had import for around 200 classes and 20 classes were defined under com.taranfx.*. The Total classes loaded after the start-up was nearing 4000.
If you look at the graph above, loaded classes for first peek was 2500. This is the point where JavaFX started reading my com.taranfx.* package. This means, at all times for JavaFX to start empty, it will need 2500 runtime classes. During this time, (almost 6 seconds), CPU utilization is max. On my Dual core it was able to eat upto 70% CPU though it was using only 7mb of heap.
If you look at the CPU graph, the second peek was incurred when my classes started executing. This was the time when Request for Flickr REST API was being generated. During this, CPU utilization of 27%, Heap size of 12mb was observed. The third CPU peek came when response from Flickr was received, and images were starting to appear. Loading of 50 images (10in a row x 5 in a column) took peek utilization of 25.5%. As clear from Memory utilization graph, Images were being loaded and heap size was increasing in steps till 20mb of which roughly 13mb was being used. Immediately after this, a substantially GC call reduced the memory overhead. (after this watever memory increase you see is because of my further operations in GUI, we will consider them in next section)

.

b. JavaFX vs. Flash
Now, I had a similar application in Flash ofcourse pretty different in layout, etc but similar functionality. Let’s compare those statistics with JavaFX.

It was roughly 40% at first peak with RAM usage of 4mb, and then 20% second peek with 10mb, and 3rd peek 29% with 19mb RAM.

c . Conclusion:
All the performance statistics were pretty close to Adobe Flash but were mostly on the higher side which is but obvious because its VM

.

iii. Animated Transitions and Full Image load:
So after loading of thumbnails is over, next comes the main application. Navigating quickly between images as it prepares requests for each of the image to be downloaded for preview. Let’s discuss the Performance of the application during power user usage. Since the application has lots of animations specially Fade transitions, We will compare how well are those can run on our machines.

Please note, I have chosen animation mainly as fade as it needs alot of CPU juice to run smoothly. Infact, smooth fading was far from reality in Swing. So let’s see how well JavaFX handles this in new refined JVM.

Let’s now look at the profiling:

a. Key Observations:

Refer to the CPU, the first wave peek you see is at approx. 15 seconds from when all thumbs were loaded. This peek was incurred due to my first click, which triggered the very first transitioning of fading all 50 thumbnails screen to Full Image preview (class ). Fading was very smooth and what a surprise it took just 19% of my CPU. Are you kidding me JavaFX ? Fading involves a number of interpolated frames during the transition time (set to 2 seconds) and frame rate of decent 20 fps was observed (I used a screen capture tool that auto-detects frame rate of the screen, it gave 20.3 fps)

Moving on, I became a power user (clicking madly on all images and then returning back) at full pace. Clicked images were being loaded in background threads while new images being licked were having fading transitions, one by one being put to backround thread queue to load images from flickr. As expected, it resulted in good CPU utilization, though transitions still remained smooth as before. CPU wasn’t kicked much it ranged between 30 to 60% which is pretty expected, no surprises. But one thing impressed me, Heap Usage. It didn’t eat much of the Heap. At start, Heap size was 20MB (15mb used), it then raised only to 25mb after 40 seconds of power user activity. GC wasn’t kicked more than once and it reduced Used heap from 20 to 15mb making it less hungry for more Heap. I must admit, memory management was pretty good.

Now let’s look at the other part, the most active Threads.

Java.util.concurrent.FutureTask.run() ate most of the CPU. Because of high number of requests, large no of threads got created, and for obvious reasons, this had to be the highest usage. Second highest is javax.swing.SystemEventQueueUtitlities, which is a old class from the generation of swings exposing some generic UI utilities. This involves the fading, and other graphical operations. Third was sun.rmi.transport.tcp.TCPTransport, this was for TCP connection to Flickr. The CPU usage for this thread is during translation of TCP requests from Java to right to your OS protocol stack.

b. Conclusion

I’m still left astonished if this was possible inside VM! The Ability of Graphics (atleast for 2D) is brilliant. Later on when I took the same test to Nvidia platform, I did see noticeable differences. The Fade effect had risen from bare 20fps to 28fps.

3. The VERDICT!

The Pros:

- Solid API with bundled inbuilt features for almost anything you will need.

- Brilliant Graphics Rendering (2D)

- Low on CPU usage than previous generation of JavaVMs.

- Comparble performance for generic RIA.

- Excellent memory management. (it’s not just standard VM memory management, there’s more to it.)

- OpenSource: A More open approach to RIA in the industry as compared to Flash, Silverlight like AJAX.

- Universal Language for Desktop, mobiles, other portables.

The Cons:

Nothing in the world is without Cons, even JavaFX has it:

- Graphics and multimedia are powerful on JavaFX, but it runs in JVM and not the real system, hence graphics can never be as powerful as native applications like Adobe Flash. The high complexity advanced 2D, 3D graphics will suffer significant performance hits inside VM. Such aplications will have more floating point calculations and More frequent memory allocation/de-allocations which will choke VM. (Though, for low complexity 2D, jaavaFX might beat Flash)

- Another -ve is for Video and High definition: The mediaplayer (via API) in JavaFX is able to attain a decent framerate of 24/25fps and CPU utilization goes to peak during this playback over SD (Standard definition). Now if you are looking for HD content to go via JavaFX, it’s not possible at this time. I used the same hardware as in the benchamark. CPU utilization goes to peak to achieve that framerate for SD.

For Playing 720p and 1080p HD content, VM came to the knees, failed to provide any tolerable frame rates. Definitely, needs lot of improvement in this sector. May be Sun can develop native APIs to interact for each Nvidia CUDA, ATI graphics Kernel for each platform to accelerate this. This could encourage 3d gaming in JavaFX. (though this is very hard to do and maintain keepin platform independence in mind.).

- A big challenge awaits javafx to reduce Load times. To load startup classes, it takes 10-20 seconds depending upon the application and CPU you are using. Flash is much faster at this.

Definitely I found my effort worth while after days of playing with JavaFX. I can proudly say it will stand as a clear challenger to Flash (provided few things improve, specially load times). From the past trend of Sun, it has innovated in every field of Java and will continue to do so. And now that It’s with Oracle and Larry Ellison, I don’t have a doubt about its future.

Future could be JavaFX. Are you Ready?

Get Started www.Javafx.com

I’m always suspicious of all claims that are along the lines of “world’s blankiest blank,” especially when those claims come from Apple.

So, is Safari 4 the fastest browser? Let’s test out Apple’s claim by pitting it against Google Chrome.

I’m using my standard test bed – QX9770 Core 2 Extreme running at 3.2GHz, with 2GB or RAM on a fully up-to-date Windows Vista Ultimate 32-bit platform.

I’m putting Safari 4 against Google Chrome because that’s currently the fastest browser on the block. If Safari 4 can beat that, it earns the title of “world’s fastest browser.” If it can’t beat Chrome, then the claim is puff.

As usual, two tests – SunSpider JavaScript, and V8 benchmarks.

The Results:

While it’s clear that Safari 4 is fast, and it’s almost caught up with Chrome in the V8 benchmark, Safari 4 is clearly NOT the world’s fastest browser.

Benchmark Safari 4 vs. Firefox 3.5 vs. chrome 2.0 vs. Chrome 3.0

Google Chrome still outperforms Safari 4.  The Firefox 3.5 was as bad as 1.6x Slower than Chrome 2.0.

Also, I did some tests with chrome 3.0 with firefox 3.5 and Safari 4.

Ahh, Chrome 3.0 is slick, don’t even look at other browsers Its 1.8x times faster than Firefox 3.5 and 1.35x times faster than Safari 4.