Archivi Blog
AMD launches its next-gen FirePro graphics card lineup, we go hands-on at SIGGRAPH (video)
Just as you’ve cozied up with “Tahiti” and “Cape Verde,” AMD has returned to grow its “Southern Islands” family of graphics cards with four fresh FirePros, offering up to four teraflops of graphics computing power. That spec can be found in the company’s new W9000, which is capable of four TFLOPs single precision and one TFLOP double precision with a price tag just shy of $4,000. That behemoth of a card offers 6GB of GDDR5 RAM and requires 274 watts of power. More humble members of the fam include the W8000, which has the same form-factor as the higher-end W9000, but eases back on the specs, consuming 189 watts of power and carrying a $1,599 price tag.
We had a chance to take a closer look at both cards at SIGGRAPH, and while they packed a significant amount of heft, you’ll likely never take a second look once they’re buried away in your tower rig. Fans of smaller housings (and price tags) may take notice of the W7000 and W5000, which are both considerably more compact and require less power to boot, with pricing set at $899 and $599, respectively. Those cards were also on hand for our demo, and can be seen along with the top two configs in our gallery below. You can also sneak a closer peek in the hands-on video after the break, and glance at the full specs over at our news post from earlier today.
TI demos Windows RT on OMAP 4470, we go hands-on (video)
It seems like every ARM chip manufacturer wants a piece of Windows 8 here at Computex 2012 — and for good reason. Hot on the heels of Asus’ Tegra 3-equipped Tablet 600 and Qualcomm’s Snapdragon S4-based development tablet, Texas Instruments is showing Windows RT on its very own OMAP 4470-based system. The 1.5GHz dual-core SoC features a PowerVR SGX544 GPU and leads the competition with a dual-channel memory interface. We chatted with Bill Crean, Product Manager of the OMAP Processor Business Unit who showed us Microsoft’s latest OS running on TI’s development tablet. The demo looked snappy enough, providing some insight about what to expect from some of Toshiba’s upcoming devices. No word yet on a quad-core version. Enjoy our hands-on gallery below and take a peek after the break for our demo video.
Kepler comes of age: NVIDIA unveils GeForce GTX 680 desktop GPU, 600M series for laptops
NVIDIA’s next-gen GPUs sure took their sweet time arriving, but the first of the Kepler crew is finally available in stores and its 28nm silicon is just itching to show off what it can do. You may be wondering what the 2GB GeForce GTX 680 brings to the gaming table, and whether it’ll put an end to AMD’s free run at the top of the food chain. Well, NVIDIA now claims it has “the fastest GPU in the world”, with both lower power consumption and a 10-40 percent performance advantage over AMD’s single-GPU rival, the Radeon HD 7970, at 1920 x 1080. How can it back up such a boast? Ultimately, everything hinges on independent benchmarks (coming soon in our review round-up), but in the meantime we need to look at NVIDIA’s new architecture for clues. Intrigued? Then head on past the break.
MSI Outs Three New AMD R7950 GPU Video Cards
Today marks the launch of three new video cards for gaming enthusiasts and overclockers from MSI. All through the video cards use the new AMD R7950 GPU inside and have different cooling solutions and presumably different clock speeds. The coolest of three cards uses MSI’s custom Twin Frozr cooling solution.
That custom cooling solution promises 10°C lower temperatures and 13.7 dB less noise than traditional cooling systems. Better cooling means higher clock speeds to overclockers. MSI claims that you can overclock up to 37.5% more with the custom cooling solution.
It’s not clear what the clock speeds are on any of the new video cards. We do know that the video cards use military class III components for durability and stability. They also have special Hi-c capacitors with eight times the lifespan of normal solid-state capacitors. That means that the cards can be overclocked and still last longer than some other similar products.
MSC Nastran 2012 si affida alla potenza delle GPU NVIDIA
NVIDIA ha annunciato che MSC Software Corporation, azienda leader nelle soluzioni software di simulazione multi-disciplinare, ha lanciato una versione accelerata da GPU di MSC Nastran 2012, un’applicazione FEA (Finite Element Analysis) utilizzata in una vasta gamma di compiti di simulazione tecnica.
La nuova versione accelerata dalle GPU, MSC Nastran 2012, è disponibile per sistemi Windows e Linux e offre prestazioni sino a cinque volte superiori rispetto all’edizione precedente. Gli utenti potranno sfruttare modelli di maggiore realismo e simulazioni di qualità superiore, aprendo la porta alla progettazione di aeromobili e navicelle spaziali davvero rivoluzionari, oltre che a diversi altri sistemi mission-critical.
AMD announces next-gen Radeon HD 7970 for $549, says it ‘soundly beats’ rivals
A fresh contender for your blow-out 2012 Olympic gaming rig: AMD’s first 28nm GPU, the Radeon HD 7970. It’s scheduled to arrive on January 9th, priced at $549 — nearly $200 more than its direct ancestor, the 6970. Then again, this newcomer packs some supremely athletic specs, including a 925MHz engine clock that can be readily OC’d to 1.1GHz, 2,048 stream processors and an uncommonly muscular 384-bit memory bus serving 3GB of GDDR5. At the same time, AMD hopes to make the card more practical than the dual-processor 6990 by bringing the card’s power consumption down to less than 300W under load and a mere 3W in ‘long idle’ mode, and promising quieter cooling thanks to improved airflow and a bigger fan. We’ll have to wait for benchmarks in January before we hand out any medals, but in the meantime NVIDIA’s forthcoming 28nm Kepler GPU might want to step up its training schedule.
New Samsung chip has two of everything: two cores, 2GHz, 2560 x 1600 graphics
Sammy’s current Cortex A9-based chips are hardly slackers — the Galaxy Note already proved that to any lingering doubters. Nevertheless, the next-gen Exynos 5250 SoC promises to double that sort of performance, by harnessing two Cortex-A15 chips clocked at 2GHz each, along with a GPU that can output resolutions of up to 2560 x 1600 (WQXGA). It’s like big.LITTLE computing, except without the LITTLE. Samsung reckons it’ll start mass producing the 5250 for use in high-end tablets by the second quarter of next year, which should be just in time to stop NVIDIAfrom getting too cocky.
NVIDIA GeForce 290.36 dispobili per il download
Nvidia rilascia la prima versione beta dei driver release 290, pacchetto che supporta le GPU GeForce serie 6, 7, 8, 9, 100, 200, 300, 400 e 500-serie desktop GPU e GPU desktop ION.
Eccovi la lista degli aggiornamenti effettuati e i link per il download:
NVIDIA PhysX
- Updates PhysX System Software to version 9.11.11.07 for the best PhysX experience in Batman: Arkham City.
NVIDIA SLI
- Added or updated SLI profiles for Crysis 2, Heroes and Generals, Inversion, Stronghold 3, and Syndicate.
HD Audio
- Updates HD Audio driver to version 1.3.9.0.
Some Key Bug Fixes
- Fixes random flickering as Windows boot logo is loading or fading away.
- Fixes corruption in Crysis 2 with SLI and lower quality shadow settings.
- Fixes corruption seen in Settlers 7 with 275.33 drivers.
Other Details
- Supports single GPU and NVIDIA SLI technology* on DirectX 9 and OpenGL.
*Note: The following SLI features are only supported on Windows Vista and Windows 7: Quad SLI technology, 3-way SLI technology, Hybrid SLI, and SLI multi-monitor support.
NVIDIA GeForce 290.36 Win 7 Vista 64-bit
NVIDIA GeForce 290.36 Win 7 Vista 32-bit
NVIDIA GeForce 290.36 XP 32-bit
Samsung signs up PowerVR SGX MP GPUs for future devices, keeps its options open
Even though it’s already listed on the dance card for ARM’s upcoming MALI-T658 mobile GPU, Samsung is also licensing tech from Imagination Technologies. The new agreement will allow it to include Imagination’s PowerVR SGX multiprocessor GPU (a.k.a Series 5XT a form of which already resides in the A5 chip used by Apple’s iPad 2 and iPhone 4S as well as the PlayStation Vita) in its upcoming devices, but doesn’t specify how many cores or what configuration may be used. MobileTechWorld also speculates this could be in preparation for SoCs built to run Windows 8, but until we actually get a peek inside whatever devices are up Samsung’s sleeve it’s impossible to know for sure.
NVIDIA says Tegra 3 is a ‘PC-class CPU,’ has screenshots to prove it
Asus can’t be absorbing all those limelight photons today. Not when its freshly detailed Transformer Prime depends so heavily on NVIDIA’s special sauce. Admittedly, we already know a lot about Tegra 3 from its Kal-El days, but we haven’t seen much in the way of real-world performance claims. Until now, that is. Below you’ll see newly released screenshots of Android games that have been souped-up to capitalize on the imminent Asus Eee Pad as well as other Tegra 3-powered devices — including smartphones — that are expected early next year. NVIDIA has also put out slides containing in-house benchmarks and head-to-head comparisons with the Tegra 2, which you’ll find right after the break.
Benchmarks clock iPhone 4S’ A5 CPU at 800MHz, show major GPU upgrade over iPhone 4
Maingear EPIC X2: Dissipatore a liquido Dual-GPU
Maingear annuncia il nuovo EPIC X2, un dissipatore a liquido in grado di dissipare il calore di due GPU contemporanamente; il cooler supporta la GeForce GTX590 (aka EPIC 590 X2), GeForce GTX580 in SLI (EPIC 580 X2) e a breve per la Radeon HD6990 (aka EPIC 6990 X2). Il Maingear EPIC è più performante delle soluzioni di dissipazione tradizionali delle schede video e presenta inoltre un design completamente aggiornabile dall’utente senza dover rivoluzionare tutto l’impianto.
Purtroppo non abbiamo altre informazioni, Vi suggeriamo però di visitare questo indirizzo per ulteriori dettagli.
Switchable graphics solutions duel it out, AMD ends up looking like a chump
At this point, NVIDIA’s Optimus switchable graphics are old hat, but AMD is relatively new to the game of packing dual GPUs in a single laptop. AnandTech decided to pit the two solutions against each other and, well, lets just say AMD doesn’t come out looking so great. The biggest problem appears to be stability. While performance is acceptable (though, not quite as good as it should be) the site had trouble getting four of the supposedly 16 supported games to switch between the integrated and discrete GPUs as advertised. Regular driver updates, not to mention a bit more testing, could solve the issues, but for now NVIDIA’s Optimus simply outclasses its AMD counterpart. Check out the video after the break and don’t forget to hit up the source link for all the details.
NVIDIA CEO sees major growth in mobile processing, quad-core tablets coming this year
During a sitdown with reporters yesterday, NVIDIA Chief Executive Jen-Hsun Huang discussed his company’s near- and long-term financial outlook, while providing some insight into the chipmaker’s quad-core future. According to Huang, NVIDIA expects to rake in between $4.7 and $5 billion in revenue during fiscal year 2013, with revenue from its mobile chip unit projected to mushroom tenfold by 2015, to a whopping $20 billion. Huang acknowledged that these predictions could be affected by external factors, including the ongoing patent wars between tablet and smartphone manufacturers, but didn’t seem too concerned about their immediate impact. “At this point, it looks like it’s much ado about nothing,” he said. In fact, Huang foresees rather robust growth in the mobile processing sector, estimating that there are about 100 million devices that will need chips this year — a figure that could soon rise to one billion, on the strength of more affordable handsets, efficient ARM processors and the rise of ultra-thin notebooks. And, despite his recent disappointment, Huang expects Android tablets to comprise a full 50 percent of the market in the near future, claiming that NVIDIA’s Tegra chips can currently be found in 70 percent of all slates running Google’s OS, and about half of all Android-based smartphones.
In the short-term, meanwhile, NVIDIA is busy developing its quad-core mobile processors — which, according to the exec, should appear in tablets during the third or fourth quarter of this year (quad-core smartphones, however, may be further down the road). Huang also sees room to develop wireless-enabled, Snapdragon-like processors, thanks to NVIDIA’s recent acquisition of Icera, but he hasn’t given up on GPUs, either, predicting that demand for graphics performance will remain stable. The loquacious CEO went on to divine that Windows 8 will support apps designed for Windows 7 (implying, perhaps, that Microsoft’s Silverlight platform will play a major role in future cloud-based developments), while contending that smaller, “clamshell devices” with keyboards will ultimately win out of over the Ultrabook strategy that Intel has been pursuing. For the moment, though, Huang seems pretty comfortable with NVIDIA’s position in the mobile processing market, citing only Qualcommas primary competition. “We’re the only people seriously on the dance floor with Qualcomm,” he argued, adding that companies without a solid mobile strategy are “in deep turd.” You can find more of Huang’s insights at the source links below.
Sony's ultraslim 13-inch VAIO Z laptop revealed in Europe, packs external GPU for power on demand
Turns out those leaked shots we saw of Sony’s new VAIO Z laptop were right on the money as the company showed it off officially today for the European press. The specs reveal a 13.1-inch “ultramobile” notebook that comes in at under 1.2kg with a 2.7GHz Intel Core i7 processor, 256GB SSD, 8GB RAM, 1600×900 screen and sheet battery borrowed from the earlier VAIO S for up to 7 hours of computing. Onboard it features only Intel’s HD Graphics 3000 solution but the VAIO Z beats other ultralights with its Power Media Dock, which contributes the power of an AMD Radeon 6650M GPU with 1GB of dedicated memory connected via “the architecture codenamed Light Peak” — Sony can’t call it Thunderbolt — when more polygons have to be pushed. The dock sports one USB 3.0 hookup plus additional USB, VGA and HDMI ports, and a slot for either a DVD or Blu-ray drive. , but it is promised to ship by the end of July in Europe so if the full specs (included after the break) are appealing then you don’t have much time to save up.
Update: Head over to the Sony UK site to configure one yourself — pricing starts at £1,434 ($2,294) with a Core i5 CPU, 4GB of RAM and no PMD. The dock is a £400 ($640) option with no optical drive included, while upgrading to a 1080p 13.1-inch LCD is a mere £40 extra.
Microsoft decides to pass on WebGL over security concerns
Well, it looks like Microsoft is taking those warnings about WebGL pretty seriously. The company has decided not to support the web-based 3D standard because it wouldn’t be able to pass security muster. Highest on the list of concerns is that WebGL opens up a direct line from the internet to a system’s GPU. To make matters worse, holes and bugs may crop up that are platform or video card specific, turning attempts to plug holes in its defense into a game of whack-a-mole — with many players of varying reliability. Lastly Microsoft, like security firm Context, has found current solutions for protecting against DoS attacks rather unsatisfying. Lack of support in Internet Explorer won’t necessarily kill WebGL and, as it matures, Microsoft may change its tune — but it’s still a pretty big blow for all us of hoping the next edition of Crysiswould be browser-based.
ASUS Mars II and Matrix GTX580 Platinum eyes-on
If you thought the original Mars graphics card from ASUS was a little bit ridiculous, get ready to see what a lot of ridiculous looks like. The company’s Mars II that was recently teased alongside a fresh new Matrix GTX580 Platinum card, squeezes two GeForce GTX 580 chips on the same board and overclocks them for good measure. In order to achieve such great feats, the card requires no less than three 8-pin auxiliary power connectors and takes up the space of three (2.6, to be precise) PCI slots with its ginormous dual-fan cooler. Heatpipes are also employed to keep the raging fires within in check, and — for situations where all else fails — ASUS has installed a special red button that sends the fan into full speed when depressed. ASUS hasn’t yet finalized how far above the default engine clock speeds the Mars II will reach, but it has a bit of time to figure that out as this extremely limited edition card is coming sometime in July. Buyers in the US, Europe and Asia-Pacific region will have to be quick on their credit card trigger, as only 1,000 Mars IIs will ever be produced. Oh, and if you’re wondering how much power a dual-GTX 580 graphics card might consume, the answer is 600W. All by itself.
NVIDIA's quad-core Kal-El used to demo next-gen mobile graphics, blow minds (video)
You might think yourself too grown-up to be wowed by shiny, glittery things, but we doubt many will be able to watch NVIDIA’s new Glow Ball tech demo without a smidgen of childlike glee. Built to run on the company’s quad-core Kal-El processor, it shows us the first example of true dynamic lighting on mobile devices and also throws in some impressive physics calculations like fully modeled cloth motion. Instead of the pre-canned, static lights that we see on mobile games today, NVIDIA’s new hardware will make it possible to create lighting that moves, fluctuates in intensity, and responds realistically to its environment — all rendered in real time. The titular glow ball can be skinned with different textures, each one allowing a different amount and hue of illumination to escape to surrounding objects, and is directed around the screen using the accelerometer in your tablet or smartphone.
NVIDIA demoed the new goodness on a Honeycomb slate with 1280 x 800 resolution and the frame rates remained smooth throughout. In order to emphasize the generational leap that we can expect with Kal-El, the company switched off two of the four cores momentarily, which plunged performance down to less than 10fps. That means the simulations we’re watching require a full quartet of processing cores on top of the 12-core GPU NVIDIA has in Kal-El. Mind-boggling stuff. Glow Ball will be available as a game on Android tablets once this crazy new chip makes its way into retail devices — which are still expected in the latter half of this year, August if everything goes perfectly to plan. One final note if you’re still feeling jaded: NVIDIA promises the production chip will be 25 to 30 percent faster than the one on display today. Full video demo follows after the break.
NVIDIA refreshes notebook graphics with GeForce GTX 560M, attracts ASUS, MSI, Toshiba and Alienware
If you’ve enjoyed NVIDIA’s fine tradition of merely bumping along its GPUs time and again and affixing a new badge, you’ll like the GeForce GTX 560M — it’s much like last year’s GTX 460M, but with more bang for the buck than ever. ASUS, MSI, Alienware, Toshiba and Clevo have all committed to new notebooks bearing the graphics processor in light of the potent performance NVIDIA claims it will bring: Namely, those same 192 CUDA cores (now clocked at 1550MHz) and up to 3GB of GDDR5 memory (now clocked at 1250MHz, with a 192-bit bus) should enable the latest games to run at playable framerates on a 1080p screen with maximum detail — save antialiasing. Of course, that assumes you’ve also got a recent quad-core Sandy Bridge processor and gobs upon gobs of RAM, but NVIDIA also says that with the built-in Optimus switchable graphics, those same potent laptops should be able to manage five hours of battery life while idling.
If you’re looking for some inexpensive discrete graphics, however, NVIDIA’s also got a refresh there, as the new GeForce GT 520MX bumps up all the clock speeds of the GT 520M. When can you expect a mobile GPU to knock the GTX 485M off its silicon throne, though? Glad you asked: a chart shows a “Next-gen GTX” coming late this year. Meanwhile, see what NVIDIA says the GTX 560M’s capable of in the gallery below and a video after the break.
