From what we know about A5 from iPad 2, the embedded GPU is quiet powerful and beats dual core phones like Samsung Galaxy SII. SGS2 sports Exonys processor,which is not as powerful as Nvidia Tegra2. So, its pretty much to be seen if iPhone 4s’s A5 can beat Tegra2 GPU. Having played Tegra 2 games on my Galaxy Tab 10.1, I definitely feel Tegra 2 is quiet powerful with games that are designed exclusively for the Tegra platform.

However, Linpack benchmark gives us a totally different story: A5 on iPad 2 beats the hell out of Tegra 2 on Galaxy Tab & Xoom. However, in the real world Gaming on Tegra2 has that PC level feel to them, we hope to see the same in smaller form factor A5 device, iPhone 4S.

image credit: ars

In both CPU and GPU performance Apple’s A5 processor left NVIDIA’s Tegra 2 in the dust, performing at least 2x faster and sometimes even 5x faster. So, the answer to the initial question – Apple is not only able to compete with other chipmakers, but as of now they have the best ARM dual-core processor on the market.

Firemint is teasing us with their latest game designed for iPhone 4s: Real Racing 2. Other than superb graphics, it can streaming not just one device to your big screen, but four – four-player streaming over AirPlay. The apps’ upcoming “Party Play” feature lets an iPad 2 or iPhone 4S play host to up to three additional iOS devices running the game, piping the stream for all four to your boob tube via Apple TV. The update also promises to make the most of the iPad 2 and iPhone 4S’A5 processor, kicking the graphics up a notch.

Apple demoed Infinity Blade 2 at the launch event, the graphics look breathtaking. the dynamic lighting is one of the best we’ve seen on phones, so far.

Update: Asus Eee Pad Transformer Prime Tegra 3 vs A5 iPad2.

We write about Latest in tech, Google,  iPhone, Gadgets, Open Source, Programming. Grab them all @taranfx on Twitter or below:

Qualcomm is claiming that its next generation of Snapdragon chips would beat current generation of processors used in gaming consoles like PS3. That’s a big claim, but noting the facts laid out by Sony PS Vita, these look real.

S4 codenamed ‘Krait’ devices would launched by end of the year. S4 would be first 28nm quad core processors and run Adreno 225 GPU – Quad core cpu + Adreno 225 aim to blow away the fluidity of PS3 graphics.

Qualcomm S

There are four new tiers of processor, named S1 to S4. That “S” stands for “System”, with the number referring to how powerful the chip is.

S1 includes standard handsets such as the HTC Droid Incredible and Sony Ericsson X10i, S2 is featured in more powerful phones like the HTC Thunderbolt and LG Revolution, and S3 powers uberphones and tablets such as the HTC Sensation and HP TouchPad.

Portable Gaming & Future

Portable gaming is going to pave the way to future. Already Android and iPhone has together eaten the major percentage of mobile market and they would start interfering with consoles next year. Imagine this – plug your mobile to a TV and use it like a console or wirelessly stream video. If not in a year, you would see this in practice 2 years from now.

2012 is the year of Quad cores. Nvidia Tegra 3, Snapdragon S4, and Intel is jumping in too. Apple isn’t holding back either, iPad 3 (or iPad 4) and iPhone 4 are going to be A6 quad core too.

We write latest and greatest in Tech Guides, Apple, iPhone, Tablets, Android,  Open Source, Latest in Tech, subscribe to us@geeknizeron Twitter OR on Facebook Fanpage:

The upgrade beings significant jump in gaming performance making iMacs capable of handling all the lastest games smoothly. The 21.5” iMac comes in AMD 6750M and 6770M, while the 27” model adds a 6970M GPU. 27” iMac can push out 2560×1440 resolution display.

Here are some benchmarks which would give you some idea of the capabilities of these Apple desktops. Checkout raw specs of the AMD Radeon 6750M, 6770M and 6970M GPUs.

iMac 2011 Benchmark

Gaming Performance

If you are a hard core Gamer, don’t look out for 6750M or 6770M.  But the Radeon 6970M with 2GB of DDR5 RAM, handles almost any games quite easily. However, if you play them in 2560×1400 resolution, performance would suffer. If you stick the upper limit to 1920×1080, you would be happy with the performance even with high detail in just about any modern game.

Radeon 6750M

The weakest of these Radeon GPUs is best suited to running games with basic OpenGL games. If you plan to play latest games, you would have to stick with low detail settings. At 1280×800, StarCraft II and FIFA 2011 can give out 100 fps. With High settings most games deliver below 30fps at 1920×1080 with 16x anti-aliasing. Crysis 2 and Metro 2033 each slow to a 12fps crawl on ultra settings.

Radeon 6770M

With the higher clock speeds, it can deliver better performance than 6750M. At a resolution of 1360×768, several games like  StarCraft II, DIRT 2 and Call of Duty: Modern Warfare are playable at near 120fps.

Radeon 6970M

This is the one you want if you are serious about graphics. The 6970M sports double GDDR5 RAM at 2GB and stream processors at 960. At 1920×1080, Ultra settings, it was capable of give out 35fps in Crysis 2, 70fps in Black Ops and 58fps in StarCraft II. All these are decentnumber for a Core i7 Sandybridge.

Video Encoding Benchmark iMac 2011 vs 2010

We write latest and greatest in Tech Guides, Apple, iPhone, Tablets, Android,  Open Source, Latest in Tech, subscribe to us @geeknizer on Twitter OR on Facebook Fanpage:

The AMD HD 6980 is the world’s fastest dual-chip graphics card for screaming-fast gamer PCs, which can be added to a computer and power as many as five displays at full resolution.

AMD had recently launched few new APU (application processor unit) which is a FUSION of CPU & GPU on a single chip.

AMD claims that the new dual-card graphics card can run 1.4 to 1.8 times faster than a card with an Nvidia GeForce GTX 580 graphics chip. AMD claims you can achieve resolution upto 7680×1600 and still get outstanding speeds. And if you want more, put two of these graphics card ( equals four chips) you can double the performance.

There is a difference in strategy among the two companies: Nvidia typically has the fastest desktop graphics chip, while AMD can take the crown for the fastest graphics card with two chips in it. Its hard to say which one is a better strategy but gamers do get a choice.

AMD HD6990 graphics card, code-name Antilles, has two settings to make overclocking a child’s play. The card drains 450 watts of power, which is substantial but as far as the heat generation is concerned, the design enables each chip to dissipate the same amount of heat.

The total performance of HD 6990  exceeds 5.4 teraflops with each graphics chip having 2.64 billion transistors. It has 4 GB of GDDR5 memory and runs at 830 Mhz (overclockable to 880 megahertz). It has 3,072 stream processors.

AMD is also offering updates to its Gaming Evolved program, which supports open standards for gaming technologies, out of which main one is “Open Stereo 3D“,

AMD has now shipped more than 35 million graphics cards that are compatible with Microsoft’s DX11 graphics standard.

Dirt 2, a DX11 Game:

For latest Google, Open Source, iPhone, Android, Tech news @taranfxon Twitter or subscribe to us below:

Nvidia has plans of leaping the current gen smartphone processors Tegra 2. Within few months, the graphics giant will launch Tegra 2 3D, a 3D version of Tegra 2 with high performance 1.2Ghz dual core processor and 3D display.

At CES 2011 we learned that existing Tegra 2 chips will power flagship Android devices like the LG Optimus 2X and Motorola Atrix 4G smartphones, as well as Motorola’s Xoom tablet.

That’s just a taste of things to come in the near future. Think new breed of portable devices incorporating glass-less 3D displays, Blu-ray support and enhanced graphics, paving the way for stereoscopic 3D gaming, exciting user interfaces and other perks.

Going further, there will be Tegra 3 Quad core Cortex A9 1.5 Ghz CPU with 3x faster graphics, BluRay video, 1900×1200 display. Something of that config is a true desktop replacement.

The ULP designation on this listing stands for Ultra Low Power in NVIDIA parlance, which would indicate an aggressively tuned power management system — the only way we can envision a quad-core anything operating within a tablet. Fall 2011 is when we should know for sure.

For latest Google, Open Source, iPhone, Android, Tech news @taranfxon Twitter or subscribe to us below:

Flash Player 10.2 is available for Windows, Mac, and Linux in adobe labs.

Stage Video API — full hardware acceleration for videos

Stage Video API delivers best-in-class, high performance video playback across platforms. Flash Player 10.1 had support for hardware decoding of H.264 content, arguably the most taxing part of playing back video.
The new API uses the GPU rather than the CPU for even more stages (hence the name) of the video rendering pipeline, including color conversion, scaling and blitting. So how much CPU will it use now for playing 1080p HD videos? NIL. Yes, for most cases, when available is fairly good, you will have almost no processing on CPU.

However, to take the full advantage of the Stage API, Content providers have to update their video players. Just because Adobe has good partnership with Google, expect them to be rolling out soon on Youtube and Android. From what we’ve heard it is already available on Google TVs and under test with Youtube.

IE9 hardware acceleration support and multi-monitor support
The new beta also includes Internet Explorer 9 hardware acceleration support , enhanced text rendering, a native custom mouse cursors API and support for full screen playback with multiple monitors.
Flash Player will also use the GPU to render other graphic elements leading to a performance boost of up to 35 percent in some cases. Apart from that, users will be able to watch a full-screen flash content on multiple monitors.

Native custom mouse cursors

Flash developers looking to add more personality to their apps, or just to annoy their users, can now create native custom mouse cursors. This means that they’ll be able to create custom mouse themes which will then be handled natively by the operating system, a much better and faster way to the alternative, rendering a custom cursor inside Flash.

Sub-pixel Text Rendering:

S-pixel text rendering enhances the Font type faces, making them better looking regardless of the fonts or local scripts used. It makes the typography beautiful on just any device.

Flash Player 10.2 Beta for Windows, Linux, Mac is available for download here.

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

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:

Although you can Build yourself Supercomputers, it does cost millions when its built as a nation’s pride.

The Tianhe-1A, meaning Milky Way, was designed by the National University of Defense Technology (NUDT) in China. It is the work of 200 scientists,  that makes groundbreaking performance  record of 2.507 petaflops, or quatrillion calculations per second, making it the fastest system in the world today. Ordinarily the computer runs at 563.1 teraflops, or trillion calculations per second.

This giant comfortably costed over $88 million.

Compare this: as of June 2010, the 500 fastest supercomputers in the world combine for 32.4 petaflops of computing power and Chinese new supercomputer does 2.5 petaflops alone.

The whole system weighs 155-ton and is cooled by 103 refrigerator-like cabinets lined up on an area of about 1,000 square meters. The Tianhe-1A will be employed in calculations of seismic data for oil exploration, conduct bio-medical computing and help design aerospace vehicles.

Tianhe-1A is powered by 7,168 Nvidia Tesla M2050 GPUs and 14,336 Intel Xeon CPUs. But the design is intuitive enough to beat 50,000 CPUs & half the floor space.

The second best, Cray XT5 Jaguar system has more than 19000 computers and 224,000 CPUs based on AMDs.

The performance scale was definitely not possible without GPUs. GPUs are redefining high performance computing. With the Tianhe-1A, GPUs now power two of the top three fastest computers in the world today. These GPU supercomputers are essential tools for scientists looking to turbocharge their rate of discovery.

Power Consumption:

Its not worlds Greenest Supercomputer, but Tianhe-1A is energy efficient, saving enough power to provide electricity to over 5,000 homes for a year, Tianhe-1A consumes only 4.04 megawatts, making it 3 times more power efficient compared to a 2.507 petaflop system built entirely with CPUs that would consume more than 12 megawatts. However, affording its yearly bill still means $2.7 million hole.

The battle to build the fastest supercomputer has become a point of national pride as these high performance computers are used in a areas like defense, energy, finance, science. They are also used for drug discovery, hurricane and tsunami modeling, cancer research, car design, and studying the formation of galaxies.

Of the world’s fastest 500 supercomputers, the United States has 291, while Europe has 145 and Asia 49.

UPDATE: World’s Fastest Supercomputer – IBM’s Mira US owns the crown, again.

We write about Latest in tech, Google, Green Tech, Gadgets, Open Source, Programming. Grab them all@taranfx on Twitter or below:

The latest Chrome Dev builds now ”upgrade” from the last Chromium 7.0.548.0 to Chromium 8.0.549.0 fixes just one bug with the sync feature.

Chrome 7 is targeted to all under the hood enhancements and almost nothing for th user appeal. Chrome 8 will change this. In addition, it will continue to improve the performance with  full hardware acceleration and support for native 3D graphics via the proposed WebGL standard, bringing IE9′s hardware performance out of the context.

WebGL promises hardware accelerated 3D graphics which go deep down the Operating system level (native API) to enable web developers to bridge the gap between desktop and the web when it comes to gaming. So what this means is that Games on the web would be as capable as native Games. All in all, good news for all Gamers, its going to get cheaper, more portable than ever.

Chrome 8 also adds new features to Chrome Labs. Some come especially handy for Chrome OS, like the Background WebApps feature, Remoting and, to a lesser degree, the Cloud Print feature, Chrome Instant – Google Instant in chrome.

Chrome 8 dev is just the preview of some basic features, more stuff is coming as it goes stable. Its going to prove-out to be very interesting release for a couple more reasons, both the Chrome Web Store and Chromium OS are scheduled to launch in the same time period.

Download Chrome Dev 8.0.549.0

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

Google is now introducing GPU hardware acceleration from Chrome. This enhancement would speed up the rendering of complex pages by offloading a lot of the processing to a computer’s graphics card.

The strategy is inline with other Popular browsers Microsoft’s IE9, Firefox hardware D2D acceleration.

IE9 is scheduled for first public beta next month, and Chrome seems to have the early basic preview already on the horizon. At this moment, Chrome only uses hardware acceleration for displaying some content and going further, they will have “even more of the rendering from the CPU to the GPU to achieve impressive speedups.”

Going further, they will offer support for accelerated 3D graphics in the browser. to match that of IE9.

How to get Hardware GPU Acceleration in Chrome

To try it out, get the Chromium builds. Once installed, you need to run the application with the –enable-accelerated-compositing flag. To do so, you can either run the program from the terminal and set this switch by hand, or – in Windows – check the properties for the executable and append the flag to the target in the properties dialog.

As of now the differences in page rendering may not be huge but as far as my tests are concerned, they were noticeable for heavy, complex pages. Try it out.

Related: IE9 vs. Chrome vs. Firefox vs. Opera Performance

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

Top of the line codec performance is showcased by CoreAVC, that lets you play 1080p smoothly, even on low-end machines (1.5Ghz Dual core, low-end GPU).

The new VLC version i.e. 1.1 seems to address these issues. Here are the key features:

GPU-Based Decoding
This is the most coolest one. Technologies like CUDA are closed and limited to NVIDIA only, but VLC opens up the gateway to General Purpose processing using GPU with DxVA (DirectX Video Acceleration API) on Windows Vista and Windows 7, and VAAPI on Linux.

It works for few GPUs and reduces CPU usage pretty well, but on others might not. You can check the GPU compatibility list here. This should improve as we get near to the final version.

Extensions
WITH VLC 1.1, users will be able to write their own plugins for VLC, codenamed Extensions. The difference with other VLC modules consists in the language used to write these Extensions: Lua, a simple a lightweight scripting language, embedded inside VLC media player.

How to use Extensions:
Download the VLC and Lua scripts and copy them to your VLC scripts folder:

• Linux: ~/.local/share/vlc/lua/extensions/
• Windows: %APPDATA%\VLC\lua\extensions

Open the Plugins & Extensions dialog to reload the list of extensions.

New Media library
Till now, VLC was just a player with no library. This is going to change with v1.1. It will now have a searchable music/video database (based on SqlLite) for easy access and management.

New Playlists, MIDI Support
VLC 1.1 has added support for WPL (WMP playlist), ZPL playlists and MIDI sequences.

Windows 7 Integration
Starting from VLC 1.0.0.3, VLC started to feature Windows 7 Integration. VLC 1.1 plans to bring all the cool stuff that is available to Windows Media Player: Preview Thumbnails, recent media, jump lists, etc.

All New User Interface
The New UI will become a lot more prettier and intuitive.

New Visualizations
Milkdrop visualizations come to VLC using Opensource project projectM.

Desktop Wallpaper Overlay
Remember the cool Feature of WinAMP that enabled you to put videos on the desktop behind icons? Same thing is now coming to VLC.

Download VLC 1.1 Download

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

Today, Nvidia announced new Mobile GPU GeforceGTX 480M – The Fastest Graphics on Laptops/Mobile.

the new graphics for mobile is a re-spin of its extravagantly overpowered Fermi desktop counter-parts. The chip, scheduled for release mid-next month, will bring whooping 352 CUDA cores and a 256-bit memory interface to up to 2GB of GDDR5 RAM.

In contrast with Nvidia’s current top of the line: 360M, this is a huge step ahead. However, to save on battery juice, the graphics processor speed is nearly halved from the desktop GTX 480, @ 850MHz, the RAM runs on 1,200MHz, and the graphics run at 425MHz. Normally, good GPUs are meant to be >500Mhz.

The Nvidia 480M has great tessellation performance of 897 gigaflops, stepping more than double from Nvidia 360M having only 470 gigaflops.

Nvidia 480M Benchmark

Its hard to give you an idea of the performance at this time, but as per the estimates, it should run games like Just Cause 2 in 1920×1080 Medium-High detail and anti-aliasing OFF around 55-65FPS.

Nvidia 480M GTX Specs

Here are the detailed specs:

We write latest and the Greatest in Hardware, Tech, iPhone, Android, Google and more @taranfx and you can subscribe to us below:

AMD Fusion was planned for 2011, but as per today’s announcement, they seem pretty confident that they would be able to deliver Fusion products to system builders within this year. No doubt, AMD has been talking about Fusion for years, and Intel has already taken the lead (only in time) to AMD by launching its Clarkdale Core i3 and Core i5 CPUs with integrated graphics, even though these use separate CPU and GPU dies combined in a single CPU package.

Related: Indepth Tech analysis of AMD’s CPU+GPU Fusion: Llano

AMD however, is determined to combine everything into one piece of silicon that does both General purpose CPU and Graphics. On being asked, How well is this going to be, AMD’s representative Bob Grim replies:

“I don’t think there’s a simple answer to that, if you look at the history of AMD, when we came out with dual-core processors, we built a true dual-core processor. When we came out with quad-cores, we built a true quad-core processor. What our competitors did was an MCM solution – taking two chips and gluing them together.

There was a fair bit of engineering work involved too, but we just have a tradition of building a piece of silicon from the ground up, in fact the only MCM (Multi-Chip Module) solution I’m aware of that we’ve ever done is on the server side with our 12-core product (the Opteron 6174). [Regardless], what we’ll be launching with Fusion is definitely all on one die.”

AMD believes (and which is logically true) that building a CPU/GPU hybrid chip on a single piece of silicon would yield significant speed advantages if not Tremendous. Apart from that, it should have power and performance advantages because it is done with latest 32nm manufacturing process. The Future is Fusion.

Read more about AMD Fusion

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

Well, here is the solution- Get best of both worlds.

Asus’s upcoming AMD-based CrossHair IV Extreme motherboard would satisfy senses. CrossHair IV is the first motherboard to feature Lucid Hydra on Both Intel and AMD platforms and lets users mix and match their graphics cards from either Nvidia or ATI and shoot up the performance.

ATI’s CrossFire and Nvidia’s SLI, do respective multi-GPU systems to boost their graphics-processing power. The current generation of CrossFire, can handle up to four GPUs at once, mixing and matching both single-GPU cards like Radeon HD 4890 and double-GPU cards like the Radeon HD 4870 X2. On the other hand, SLI can support up to three compatible cards at once, or four GPUs with two dual-GPU cards like the GeForce 7950 GX2 or GTX 295.

Crossfire vs. SLI is always an un-solvable argument, no one has ever won the argument.

Lucids Hydra chip will try to solve this debate by supporting different combination of both Radeon and GeForce cards. Last year’s test system used a Radeon HD 4890 and a GeForce 260GTX card working together pretty nicely, the cards consistently performed better than either single card, and often slightly better than one of the dedicated dual-card setups.

MSI launched the first Lucid Hydra motherboard (called Big bang) for Intel chips in January and Asus now brings Lucid Hydra to both platforms with CrossHair IV things would start to look better.

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

ANGLE or Almost Native Graphics Layer Engine, is an Open Source Project to make it possible to run WebGL, a driverless API to DirectX 9. The project is still-developing cross-platform Web standard for accessing low-level 3D graphics hardware based on the OpenGL ES 2.0 API that can be implemented directly in a browser without a plugin.

WebGL is implemented in many browsers but has direct dependency on OpenGL drivers on each Windows, Linux and Mac OS X. On the other hand, the competing graphics API is Microsoft’s Direct3D, which is part of the company’s DirectX graphics technologies. Microsoft’s DirectX technologies have increasingly become dominant in PC gaming and have almost ruled out OpenGL on windows.

The challenge with WebGL is that many Windows machines  can’t (Mac, Linux can) render WebGL content because the OpenGL drivers aren’t installed, even though the computer has powerful graphics hardware (GPU to be precise).

ANGLE will allow Windows users to overcome this problem and run WebGL content without having to find and install new drivers for their system. Because ANGLE aims to use most of the OpenGL ES 2.0 API, it may help developers working on mobile and embedded devices as per Google.

“ANGLE should make it simpler to prototype these applications on Windows and also gives developers new options for deploying production versions of their code to the desktop”

Access ANGLE on Google Code

We write Latest Tech updates in Open Source, Programming, Java, Tech News find us on Twitter@taranfx or subscribe below:

With GTX 480, NVIDIA is stressing that GTX 480 has a  superior tesselation benchmarked performance over the ATI HD 5870, but it was very levelled for most of the other tasks. Tesselation is all about the extra detail you can embed using DirectX 11 APIs and GTX 480 handles extreme amounts of tesselation pretty well.

Nvidia’s claims makes all sense but we will have to wait for the Real Game Benchmarks to conclude who’s better. But enticing  part in Nvidia‘s tech is the 3D vision, Which makes you jump into the 3D Reality world, which still needs time to mature on ATI platform.

More about: Nvidia Fermi Tesla GPU Insights

We write Latest in Hardware: Intel, AMD, Nvidia, Research,Gadgets: Android, iPhone, Apple, Google andTech Guides, Tech News @taranfx on Twitter and:

Mozilla had started working on “JagerMonkey”, a new “baseline” method JIT compiler for SpiderMonkey, Firefox.The JagerMonkey method JIT will provide a much better performance baseline, and tracing will continue to speed up.

Right now, the work is pretty much in progress, JagerMonkey  is 18% faster interpreter and it’s a long way to go, the scope is huge.

Another amazing development is the DirectWrite & Direct2D landing in the nightly builds.

DirectWrite and Direct2D are Windows Vista and Windows 7 based DirectX 11 APIs for text and 2-D graphics that can be hardware accelerated. To test this feature, download the latest nightly build and do:

• Enter ‘about:config’
• Enter gfx.font in the ‘Filter’ box, Edit  ’gfx.font_rendering.directwrite.enabled’ to set it to true
• Below this, right click and select New > Integer to add a pref setting and Enter ‘mozilla.widget.render-mode’ for the preference name, 6 for the value
• Restart

Known Issues: Some extensions like stylish and adblock+ may break the new features.The major benefit can be seen in  Rendering Heavy pages, and also the fonts look really great with DirectWrite enabled. Also, difference can be seen in 2d apps like Google Maps, very similar to what is coming in IE9.

So, if your GPU is sitting idle, it’s time to put it to good use.

JagerMonkey: Under the Hood

As David Madelin  claims on Mozilla blog, TraceMonkey’s tracing JIT is very fast, it is 9x faster at certain operations such as SunSpider’s “math-cordic” benchmark.

But it can’t really trace a benchmark like date-format-tofte, which calls eval in its main loop, so tracing only yields a 5% speedup on that program. David Anderson calls it “TraceMonkey has rocket boosters, so it runs really fast when the boosters are on, but the boosters can’t always be turned on”. There are many factors that can prevent the rockets from turning on, so there’s really no short description of the programs that don’t trace, but most of them fall into a few categories:

Programs with very branchy control flow.

Programs with many type combinations. Because TraceMonkey generates type-specialized code, it must generate a separate trace for every type combination (mapping of variables to types) the program generates. If there are 1000 type combinations, we have the same problems we get with 1000 paths.

In nutshell, type-specializing trace JIT generates really fast code, but can’t generate native code for certain situations. On the other hand, an optionally specializing method JIT generates moderately fast code, but can generate native code for any JavaScript program. When both of these are used collectively and intelligently, they can complement each-others drawbacks and produce extremely fast execution.

If all goes well, Firefox can compete with chrome’s current performance, but who knows how fast Opera and chrome would be by that time.

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

Well, not really. They have been around since years but they haven’t lived up to user expectations. Nvidia will change that with their new Optimus technology (hopefully).

We are not talking about PowerMizer which only changes the clock (and that too is absent in Windows 7). It’s rather about switching between integrated and discrete graphics. Switching between integrated and discrete graphics started with physical toggles, which needed restarts, followed by Windows In-Tray utility that blanked your screen for a while.

Nvidia Optimus promises to make it all easier fro the end-user. When the system detects a 3D application, game or video, the GPU simply turns on. When its done, it goes back to the integrated GPU. There are no flips, banks or noticeable stoppages on the video.

You must be thinking, was it that tough to make this truly transparent all these years? Read on.

Previous graphics had multiplexers (MUXes) connecting both the integrated and discrete graphics to the display. This was costly, and required software to change the system’s rendering between the two graphics systems.

Optimus is a lot cleaner. When discrete GPU needs to switch, it powers up in a split second and copies the results of its frame-buffer to the frame-buffer of the integrated graphics chip over the PCIe bus. Because the system is technically always displaying the contents of the integrated graphics chip’s frame buffer, there is no flicker or blanking of the screen, and no series of MUXes required to connect them both to the display.

The other part of it is the software, which will detect the need and usage that relies on Nvidia’s drivers having application profiles for the games or applications. The technology will ultimately makes its way to other Nvidia driver features that require profiles, like SLI and 3D Vision.

Currently, Optimus supports pretty much all Intel notebooks: Core 2 Duo, Atom N4xx Pine trail, Core i3, i5, and i7. Optimus-enabled GPUs will be found in the GeForce 200M, 300M, Ion, and future-generation notebook GPUs.

Obviously, this not a revolution. This is something we expected to see already, but was absent. Ironically,  the average consumer assumes that switchable graphics already works the way Optimus does, so here we have a technology that finally delivers on customer expectations.

We write Latest in Hardware: Intel, AMD, Nvidia, Research,Gadgets: Android, iPhone, Apple, Google and Tech Guides, Tech News @taranfx on Twitter and:

AMD has announced a new chip is called the “Llano” processor which is not just a  CPU or GPU, instead, it’s a hybrid design that the chip company calls as an “Application Processor Unit,” or APU.

Intel’s current offering for the laptops is pretty limited in power they offer for graphics. They offer very-basic level of HD video playback, with DirectX 9 graphics to notebooks. AMD’s Llano will be combining a full DX11-compatible GPU with four CPU cores on a single 32nm processor die. This can literally swipe away all the Intel Laptop market.

At this time not much is known about the GPU, what we know is that the GPU offered by this APU combo is highly parallel vector. For the CPU, AMD has taken the “STARS” core that’s used in their current 45nm processor, shrunk them into a new 32nm SOI high-K process. AMD further added new power gating and dynamic power optimization capabilities to it. The result: 4Cores, each core has approx. 35 million transistors, and a 1MB L2 cache. It is planned to be clocked under 3Ghz and the power consumption would range between 2.5 to 25 watts(peek).

AMD will introduce another offering for the mobile called “Bobcat” which is more portable across a range of processes and configurations, and features less custom work that in Llano.

Llano will see some appearances in the later half of this year, and OEMs will be shipped around early 2011.

Power optimization
The most amazing part of this chip is it’s unique approach to dynamic power optimization. It is a step ahead from the traditional analog power management, towards digital, which makes it more accurate, efficient.

How Power optimization worked in current-gen. Processors: Normally, Power module takes analog input from a set of diodes placed throughout the die which act as thermal sensors, informing the module when the die heats up in an area due to increased compute activity.Temperature is the means for calculating power consumption.

Now, there is a problem with this approach: Ambient (surrounding) temperature changes (even minor ones) can change the values, make unwise decisions in setting the right clock speed and hence inefficient use of energy. And if the temperature changes are huge, like moving a laptop from A/C cabin to outdoors, the calculations go upto 40% incorrect.

However, Llano does it smartly. It uses a set of 95 bits of signals from different parts of the chip (pre-identified) are monitored, which have strong correlation to power consumption. This includes monitoring  integer traffic, cache misses, or branch mis-predicts via low-frequency sampling, which yields a very-high accuracy in determining the power usage.

AMD Fusion and future:

Clearly, AMD + ATI Fusion is now the way to go on notebooks. The combo/Fusion will reduce the Bus latency between CPU-GPU as they are now on the same Chip, making graphics faster and GPGPU apps more powerful.

Intel barely has any integrated GPU for 2011. Unless Intel buys NVIDIA, I don’t see them leading this segment.
On the other side,  NVIDIA’s upcoming x86 CPU + GPU combination would be interesting but we’ve to see how they perform in the CPU segment and not to forget they have some pending DMI licensing disputes.

Altogether, 2011 is going to be interesting for the notebook segment and It would make alot more sense even for general purpose apps to run on GPU with DxVA (DirectX Video Acceleration API) on Windows, VAAPI on Linux.

[via Inquirer]

We write Latest in Hardware: Intel, AMD, Research,Gadgets: Android, iPhone, Apple, Google and Tech Guides, Tech News @taranfx on Twitter and:

The Evolution and Features that can mark Success for Windows 7 – Direct X 11 from dx10 and dx9

No doubt, Windows is the best platform for Gaming. Other platforms, due to lack of proper driver support and optimizations (Linux, Mac OS); never lived upto the competition. What we mean by Best platform is: If you have a good Graphics GPU and a great Processor, you can beat the performance of PS3 and Xbox 360.

Note: This article is part of Why Windows 7 series.

Prevalently, DirectX nowadays is a mark of success over the other standards. But, we tend to forget that a decade ago, war waged between Microsoft and Silicon Graphics in the field of 3D APIs has lived for years.  Part of it was to do with the success of the Quake engine, which paved the way to solid support for OpenGL for the makers of 3D cards in their drivers.
Microsoft was starting from scratch, and the learning curve was exponential. So, for several years, Direct3D’s capabilities were below the curve, with an API that many programmers found a lot more confusing than OpenGL’s. With each new version of Direct3D, it gradually began to catch up with OpenGL.

In Direct X8, for the first time, Microsoft’s API did more than just copy from SGI. It actually introduced innovations of its own like support for vertex and pixel shaders. With DirectX 9, Microsoft managed to strike a decisive victory, imposing its API on developers.
There’s a good chance that Direct3D 11 will prove to be a more important page in the history of the API than version 10 was. While Direct3D 10 was a complete revision, brining incompatibility,  Microsoft has now put enough distance between it and this new version to correct the problems raised by the first major overhaul of its API. So you could call Direct3D 11 a major update with potential fixes to new features in Dx10. It re-uses all the concepts that were introduced with Direct3D 10, and is compatible with the preceding version and with the preceding generation’s hardware. It’ll be available not only on Windows 7, but also on Vista. This widens the gaming horizon.

A typical game’s development phase is between 2 and 4 years. So by the time a game that is just now starting its development phase is released, Direct3D 11 will be already well established for PCs, since it’ll run on all PCs shipped with Windows 7 and work on the great majority of PCs running Vista.

The Difference is here:

The best way to praise Directx 11 is to compare it with previous versions. Here is a screenshot that gives you slight idea:

Direct x - dx9 vs. dx11

You might have to concentrate to see what’s changed between the two versions. Hint: It’s not about better shape and natural colors.

Dx10 vs. dx11

In that first shot with the swine flu guy, it seems the one on the right is a bit nicer looking, as in his head and mask don’t look as polygonal as the one on the left. The two vents in his mask are actually round instead of octagonal, and there are more details all around. and the one on right has more Natural colors than amplified ones on left.
The second shot, has more to do with Digital Vibrancy and detail. The right image gives a better lighting to the far off land across the gate.
In the stills, the difference will be minor, but it becomes much more noticeable while you are in the game.

Let’s have a brief look on the Core changes to DirectX 11:

Multi-threaded Rendering

One may ask, “We’ve had multi-core CPUs for several years now and developers have learned to use them. So multi-threading their rendering engines is nothing new with Direct3D 11“. Well, this may come as a surprise to most, but current engines still use only a single thread model for 3D rendering. The other threads are used for sound, decompression of resources, physics, etc. But rendering is a heavy user of CPU time, so why not thread it, too? There are a several reasons, some of them related to the way GPUs operate and others to the 3D API. So Microsoft set about solving the latter and working around the former.

Multithreaded GPU -via TomsHardware

After all, there’s only one GPU (even when several of them are connected via SLI or CrossFire, they create one virtual GPU) and consequently only one command buffer. When a single resource is shared by several threads, mutual exclusion (mutex) is used to prevent several threads from writing commands simultaneously and stepping on each others’ feet. That means that all the advantages of using several threads are canceled out by the critical section, which serializes code.

No API, today, can solve this problem—it’s inherent in the way the CPU and GPU communicate.
But Microsoft is offering an API that can try to work around it, if not solve directly. Direct3D 11 introduces secondary command buffers that can be saved and used later.

Direct 3D 11

So, each thread has a deferred context, where the commands written are recorded in a display list that can then be inserted into the main processing stream. Obviously, when a display list is called by the main thread (the “Execute” in the “Multi-threaded Submission” diagram below) it has to be ascertained that its thread has finished filling it. So there’s still synchronization, but this execution model at least allows some of the rendering work to be parallelized, even if the resulting acceleration won’t be ideal.

As one would anticipate, a large share of the workload was still on the main thread, which was already overloaded. That doesn’t ensure good balance, needed for good execution times. So, Microsoft has introduced a new interface with Direct3D 11: a programmer can create one Device object per thread, which will be used to load resources. Synchronization within the functions of a Device is more finely managed than in Direct3D 10 and is much more economical with CPU time.

Thanks to tessellation, the potential is enormous. It, now,  becomes possible to do without the normal map and implement a level of detail directly on the GPU, allowing the use of very detailed models (several million polygons instead of 10,000 or so with current games)—at least in theory.  Can Direct3D 11 cards avoid these pitfalls practically? It’s too early to say, but in any event not everybody is convinced, and id Software is working on solving the same geometry problem with a completely different approach based on ray casting with voxels.

Nvidia’s CUDA marked an evolution. Microsoft wasn’t about to let the GPGPU market get away and now has its own language for using the GPU. The model they chose, like OpenCL, appears to be quite similar to CUDA, confirming the clarity of Nvidia’s vision. The advantage over the Nvidia solution lies in portability—a Compute Shader will work on an Nvidia or ATI GPU and on the future Larrabee, plus feature better integration with Direct3D, even if CUDA does already have a certain amount of support. But we won’t spend any more time on this subject, even if it is a huge one. Instead, we’ll look at all this in more detail in a few months with a story on OpenCL and Compute Shaders.

Graphics will mark a new step with Direct X 11. Are you in ?

credits: Tomshardware, Gamefest, gizmodo

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

Intel claims that Nvidia Ion is overkill for Netbooks battery. But practically, Intel’s Pine Trail Atom n450 chips are pretty bad in almost everything. Their power management is not as good as Nvidia’s and when it comes to performance, there’s no comparison.

It had been long predicted, that this day would come when Browsers will learn to leverage GPU for rendering webpages. After Microsoft’s announcement of GPU acceleration in it’s upcoming browser IE9, Firefox also started to get into it seriously.  Microsoft had published new API with Windows 7 called Direct2D. Direct2D is designed to be  replacement for GDI and takes the task of vector graphics rendering engine, using GPU acceleration to give large performance boosts to the webpage rendering.

Update: Chrome gets Hardware GPU Acceleration, Chrome 7 to get Hardware Acceleration, 60x Faster

Firefox gets GPU Hardware Acceleration, JagerMonkey Boost

Firefox gets GPU accelerated OpenGL Push with WebGL [3D Graphics]

The support for D2D is being seen on Firefox’s Latest alpha release Minefield 3.7 a1(nightly)
Direct2D has been implemented in Firefox as a Cairo backend. Firefox uses Direct3D textures as backing store for all surfaces. This allows Firefox to implement operations not supported by Direct2D using Direct3D

Since we are offloading the CPU task to GPU, there is a delay in between the data transfers from GPU to CPU  before the CPU can read it. Readbacks have significant performance penalties because of GPU-CPU synchronization being required.

However, on Direct3D10+ hardware  (Windows Vista, Windows 7) this should not negatively impact performance, it does mean it is harder to implement effective D2D software fallback.

Benchmarks
Bas managed to get a Benchmark of the browsers on a  ATI Radeon HD4850 GPU:

If we alook at the pattern from the graph, complexly structured websites the performance advantages are significantly less, and the error margin in the measurements can be seen to be larger (i.e. different rendering runs of the same site deviated more strongly).

We can see that D2D provides a much better experience on our daily browsing. All the Graphic intensive work like Google Maps, Scrolling through webpages,  Image rendering, animation with javascript will have much smoother framerate on Direct2D.

Download MineField 3.7a1

These test were done on a Direct X 10 card, but older PCs with pre-D3D10 graphics cards and WDDM 1.0 drivers will not show significant improvements, moreover, It may or may not work on D3D9 graphic cards, depending on exact graphics card specifications.

Let’s hope the support goes better for Direct X 9 cards as well, may be Google has a better solution on this, we will have to wait and see.

At this year’s SC09 supercomputing trade show in Portland, Oregon, Nvidia unveils its plan for the next-gen Chip ‘Fermi’. Overall, the GPU co-processors aimed at personal supers and massive clusters were the stars of the show. This is something much more powerful than AMD’s Fusion.

The video card versions of the Fermi chips will be called the GeForce 300 M line. Nvidia  is keeping the Tesla brand for its next generation GPU co-processors for workstations and servers. The Fermi chips will be sold under the Tesla 20 brand.

Last month El Reg talked about the architecture.

The SM design contains 32 basic Cuda cores – four times as many found in previous generations of SM – each comprising one integer and one floating-point maths unit. It is able to schedule two groups of 32 threads – a group Nvidia calls a “warp” – at once. The networked cores connect to 64KB of shared L1 cache, also used by four Special Function Units (SFUs) which handle complex maths formulae such as sines and cosines.

One of the 32-core Stream Multiprocessors

image credit

Flavors and Specs:

As per Andy Keane, GM at Tesla supercomputing (Nvidia), the Tesla 20 cards will come in 2 flavors. Nvidia will sell co-processor systems that can plug right into HPC clusters and link to servers through PCI-Express 2.0 links – and at around 130 watts. Keane bristles at anyone who claims that a fully burdened heat budget for a server – not just a microprocessor, but its memory controller (if it is not integrated), its chipset, and its memory – will be any lower.

With the Fermi family of GPUs, Nvidia is adding L1 and L2 caches to the co-processors and is putting ECC memory scrubbing on internal GDDR5 video memory on the card as well as accesses to external server memory. This ECC support, as it turns out, is as important as anything else in the chip if you want to sell GPUs to nuke labs.

The Fermi chip has 512 cores, which is more than twice the cores of the first Tesla GPUs. The Fermis bundle 32 cores together into a streaming multiprocessor that has 64 KB of shared L1 cache. All 512 cores have access to a shared 768 KB L2 cache, and they support the IEEE 754-2008 double precision floating point standard.

In theory, the Fermi chip can address up to 1 TB of memory, but the Tesla C2050 GPU co-processor has 3 GB of GDDR5 memory and double precision floating point performance of 520 gigaflops, costs $2,499. The Tesla C2070 GPU has 6 GB of GDDR5 memory and is rated at 630 gigaflops, costs $3,999. The bang for the buck is best with the smaller unit, which weighs in at $4.81 per gigaflops compared to the $6.35 per gigaflops of the faster GPU.

On the higher Front, a 1U appliance with four of the faster C2070 GPUs delivers 2.52 teraflops of double-precision floating point performance and costs $18,995, or $7.54 per gigaflops.

GPU and Programming

There are two secret stuff in the Fermi GPUs that are going to get HPC freaks thinking about using GPUs.

First one, they will now support Nvidia’s C++ compiler (not just C).

Secondly, a set of new InfiniBand and Tesla drivers that InfiniBand chip maker Mellanox and Nvidia have got together to streamline the movement of data from the InfiniBand ports, to the CPU’s main memory, and then down through the PCI-Express bus to the GPU card.

The way it works now: Data comes in over InfiniBand, works its way into main memory and is copied; before it is moved down to the GPU, it is copied again and that copy is what is moved. The driver changes allow for the data moved into memory to be moved down to the GPU in one fell swoop. The improvements? They ahve been able to demonstrate a 30 percent speedup.

Finally, Nvidia this week released the beta of the CUDA Toolkit 3.0, which exploits the Fermi GPU’s features.

Availability

The Tesla 20 GPU co-processors & the appliances based on them will be available in the 2nd quarter of 2010. The GeForce graphics cards based on the same GPU chips will be seen around first quarter.

Need more Tech? Subscribe below:

AMD is claiming that this graphics card breaks several older records. Besides being the fastest, it’s the one of the first to provide support for Microsoft DirectX 11 and Eyefinity multi-display (that’s three displays at once) at a max resolution of 7680×1600. Talking about the Horsepower, it delivers approximately 5 TeraFLOPS of computer power (and up to 10TeraFlops when combined with ATI CrossFireX).

Inside DirectX 11 – Dx11

The HD 5970, by default, is capable of being Overclocked. And there is no need to worry much about heating, the advance onboard fan does cool the card pretty nicely due to the new vapor chamber technologies and a vented exhaust.

Radeon HD 5970 specifications

GPUs runs at 725MHz, that the graphics card has 3200 stream processors (2×1600) and that it has 2GB GDDR5 memory.
The rest of specification includes 160 texture units, 116 Gtexel/second texture fill rate, compute performance of 4.64 Tflops as well as 64 ROPs. Pixel fillrate is 46.4Gpixels/s, Z / stencil is at 185.6 Gsamples/s and the memory data rate is 4.0 Gbps.

Maximum board power is 294W, while the idle power is 42W. Let’s not forget that the card has two DVI ports and DisplayPort, something that barely anyone needs.

In celebration of their fancy new card and their 40th birthday, AMD is also sponsoring a contest, starting today, where participants could win one of 10 Microsoft Xbox 360s or one of 10 Wiis from Nintendo. You can enter by becoming a fan of AMD on Facebook.

Update: AMD has added a video to demonstrate the power:

The Radeon HD 5970 is now available for sale with a hefty price tag of $599. Gamers, game on.

ATi Radeon HD 5750 5770 brings Gaming on Budget