OnlyHardwareBlog

General discussion, news & views about Hardware

Posts Tagged ‘Nvidia’

Nvidia to drop GTX460 price

September 7th, 2010

GRAPHICS CHIP DESIGNER Nvidia might be about to lower the price of its GTX460 GPU, shortly before releasing its GTX450 model.

Reports of the price drop first surfaced at Kitguru, where bloggers wrote, “We’d heard that Jay Puri (Sales vice-president) and Jen Hsun Huang (CEO and co-founder of the chip firm) were looking to drop the price of the fastest selling Fermi card ever, by $20.”

“We’re now hearing that the boys in charge don’t think $20 will be enough to stem the tide of the Rise of the Radeons,” it added, reporting that Nvidia might cut the price by another $10 and hinting at an escalating price war with AMD.

We liked the GTX460 when we saw it, and since that was in July it feels like only yesterday, which makes a price cut of $30 seem like more of an aggressive sales tactic than an acknowledgement that it is not selling.

The GTX460 comes in two versions – a cheaper 768MB version with a 192-bit memory interface and 1GB model with a 256-bit memory interface.

In the UK their prices range between £140 and £195, suggesting that some future buyers could get a bargain Green Goblin card.

We asked Nvidia for a comment earlier today, but by press time we’ve not had a response yet.

Source:http://www.theinquirer.net/inquirer/news/1731544/nividia-drop-gtx-460-price

Share

No comments »

Posted in Hardware News

Tags: AMD Hardware Nvidia

Nvidia and AMD ready to launch new GPUs in October

September 7th, 2010

Share

Nvidia is prepared to launch its latest GeForce 400-series GPU, the GeForce GT 430 codenamed GF108, around the middle of October and the GPU will feature a performance in between its GeForce GT 240 and GT 220; meanwhile, AMD is also set to launch its Radeon HD 6000-series around the same time, according to sources from graphics card makers.

AMD will launch Radeon HD 6700-series GPUs for its HD 6000 family initially, targeting the mainstream market.

There are two GPUs codenamed Barts XT and Barts Pro currently known under the 6700-series which will take over the position of the existing Radeon 5770 and 5750, the sources noted.

Nvidia is set to host its GPU Technology Conference from September 20-23 in the US.

The company will discuss its latest technologies as well as the industries that Nvidia has cut in with its GPUs, while partners including Hewlett-Packard (HP), Microsoft, Dell, Samsung Electronics, Acer and Taiwan Semiconductor Manufacturing Company (TSMC) will all attend the show.

Source:http://www.digitimes.com/news/a20100906PD214.html

Share

No comments »

Posted in Hardware News

Tags: AMD Hardware Nvidia

Nvidia creating southbridge with graphics for sandy bridge?

September 2nd, 2010

Share

Nvidia may have devised a strategy to bypass Intel’s northbridge blockade: flank ‘em from the south.

According to unnamed sources in the motherboard industry, the GPU maker plans to stuff a graphics core inside a southbridge, possibly sidestepping the existing legal restrictions.

DigiTimes reports that Nvidia’s product might even be $10 to $15 cheaper than Intel’s offerings.

When Intel developed its Nehalem-based chips, it severed a longstanding license agreement with Nvidia.

The architecture (and subsequent advancements) utilize an embedded memory controller.

Intel claims these types of chips aren’t included in the agreement, effectively preventing Nvidia from making northbridges that interface with Intel’s latest platforms.

However, the DigiTimes report claims that “Nvidia is developing the new chipset to bypass Intel’s new Sandy Bridge architecture.”

What’s more, with the recent investigations by the US FTC and suits for unfair business practices, it would be unwise for chipzilla to do anything that could be seen as anti-competitive.

Source:http://www.techspot.com/news/40143-nvidia-creating-southbridge-with-graphics-for-sandy-bridge.html

Share

Comments Off »

Posted in Hardware News

Tags: Graphics Hardware Nvidia

Winners and losers from Intel’s warning

August 29th, 2010

Share

When the undisputed kingpin of the semiconductor industry, Intel, makes an earnings or demand announcement, the effects ripple far and wide. So when Intel announced Friday that revenue is going to come in below expectations, an already jittery industry took one step closer to the edge.
Join the club, Intel

Intel’s warning focuses on consumer weakness, which isn’t exactly new information. It can join NVIDIA in the blaming-consumers-for-worse-than-expected-results club. NVIDIA closed out last quarter with $811 million in earnings, significantly below its initial guidance of $950 to $970 million. The culprit behind NVIDIA’s stunning revenue miss: weak consumer demand. Oh, and the favorite punching bag of executives blowing earnings: Europe.

Buy this smartphone, not that PC

Beyond the general economic woes that are causing consumers to spend less, the PC industry faces an alarming shift in consumer spending habits. The explosion of smartphones and tablets isn’t causing consumers to spend more on electronics; the Consumer Electronics Association expects consumer electronics sales to be flat in 2010.

Instead, consumers are shifting their electronics dollars away from expensive PCs, where Intel gets higher sales and better margins, and toward tablets and smartphones, where Intel’s processor offerings either aren’t as competitive or don’t exist. The average age of PCs has risen to its highest level in a decade, 4.4 years. Consumers are sticking with old PCs while buying the newest iPhone or Blackberry.

That should spell some future problems for AMD. When Intel reported its last record-shattering quarter, strong enterprise spending was the driving factor. AMD’s recent quarter focused on its strength in the consumer market. Whatever pains Intel is seeing should only be amplified at its smaller rival.

More winners and losers

However, while Intel’s results should have a general downward pull on the IT industry, it’d be wrong to apply that weakness to a wide swath of companies.

For one, Intel affirmed that demand from large companies remains strong. That’s good news for companies targeting sales to large businesses. Outside of semiconductors, the announcement shouldn’t reflect any weakness in companies such as EMC in the booming storage industry or enterprise-focused sellers like STEC. Nor should the announcement have much of an effect on semiconductor companies Qualcomm or Broadcom, which derive significant sales from the smartphone market. Intel’s warning only further points to the continuing success of small, connected devices.
However, the warning is a bad omen for hard-drive makers Western Digital and Seagate. These two companies have higher exposure to PCs, and less to booming markets like smartphones that use solid state drives (SSDs). Neither Western Digital nor Seagate have much exposure to those markets. The warning is also bad news for an obscure software outfit from Redmond named Microsoft, which presumably sells software that are on most PCs.
Bottom line

Intel’s announcement once again demonstrates the significant shift going on in consumer spending. That shift puts all the more pressure on Intel’s new Moorestown processor to enter the market and be a smashing success. The winners of a previous technological generation are rarely the winners of the next. Unless Intel wants to be another example of that axiom, it’d be better off spending less money buying software companies and more money winning the war to power mobile phones and tablets.

Source:http://www.msnbc.msn.com/id/38903304/ns/business-motley_fool/

Share

Comments Off »

Posted in Hardware News

Tags: INTEL Nvidia

Apple mac os x opencl performance vs. linux

August 28th, 2010

Share

Yesterday we looked at the performance of Apple’s “Snow Leopard Graphics Update” for Mac OS X 10.6.4 designed to enhance both the image quality and rendering performance for OpenGL games and applications.

For testing their graphics update we benchmarked Mac OS X 10.6.2, 10.6.3, 10.6.4, and 10.6.4 with the Snow Leopard Graphics Update 1.0 installed and benchmarked the Apple OpenGL performance against Ubuntu Linux.

The results were mixed showing Apple still has room to optimize their OpenGL stack compared to NVIDIA’s Linux implementation and in not all areas did this package update result in performance enhancements.

After we finished that OpenGL comparison, we decided to see how the OpenCL performance compares between Mac OS X 10.6.4 and Ubuntu Linux 10.04.1 LTS. We tested the Open Computing Language on both the Intel Core 2 Duo CPU and on the NVIDIA GPU.

OpenCL was originally developed by Apple in collaboration with the major tier-one hardware vendors and is now maintained by the Khronos Group, which also sees after the OpenGL specification and various other industry APIs.

Apple released Mac OS X 10.6 “Snow Leopard” with a full OpenCL implementation that is capable of running on either the Intel CPU or ATI/NVIDIA GPUs.

Older Mac computers with Intel graphics cannot accelerate OpenCL.

Apple’s OpenCL implementation utilizes the LLVM (Low-Level Virtual Machine) with the Clang compiler, both of which projects Apple also sponsors.

Unfortunately, Linux currently lacks an OpenCL implementation that is provided “out of the box”.

The only way right now to take advantage of OpenCL under Linux is by either installing the ATI Catalyst driver and the AMD Stream SDK or the latest proprietary NVIDIA driver.

When using the AMD Stream SDK you can utilize OpenCL on Radeon/FirePro graphics processors as well as with x86 SSE2/SSE3/SSE4 CPUs, but installing the Stream Software Development Kit requires the Catalyst driver to also be installed.

The latest ATI Stream SDK release, which is version 2.2, supports the latest OpenCL 1.1 specification.

The proprietary NVIDIA Linux driver ships with its OpenCL library and header file within the driver package itself, which is much more convenient than having to install a separate package like is required with ATI graphics.

However, NVIDIA’s OpenCL implementation for Linux is limited to running on the GPU and not any CPU.

The latest stable NVIDIA Linux driver implements the OpenCL 1.0 specification, but their latest development drivers that CUDA/OpenCL developers are able to test do provide OpenCL 1.1 compatibility.

The OpenCL 1.1 specification was released in June adds new data types, enhanced use of events, additional C built-in functions, and improved OpenGL interoperability, among other enhancements since the OpenCL 1.0 specification was ratified in 2008.

Sadly, these are the only two viable options for using OpenCL under Linux at this point.

There is Clover, which is the work of Zack Rusin to bring OpenCL over Mesa and Gallium3D for open-source graphics drivers and from there it could be implemented on the CPU via LLVMpipe.

Clover also uses LLVM and Clang for its OpenCL compilation. While Clover has been around for a year, it has not made much progress and is still far from being usable.

Once there is Clover implemented in Mesa/Gallium3D, we may finally see OpenCL available “out of the box” for Linux distributions or when any other open-source projects come around aiming for just a CPU implementation of OpenCL 1.0/1.1 with LLVM+Clang.

For this testing we once again used the 2009 Mac Mini (Apple Mac-F22C86C8) with an Intel Core 2 Dup P7350 clocked at 2.00GHz, 1GB of system memory, a 120GB Fujitsu MHZ2120B SATA HDD, and NVIDIA GeForce 9400M graphics.

On the Mac OS X side we were running the updated Mac OS X 10.6.4 installation and Xcode 3.2.3 while on the Linux side was Ubuntu 10.04.1 LTS with the Linux 2.6.32 x86_64 kernel, X.Org Server 1.7.6, GCC 4.4.3, and Clang/LLVM 2.7.

Originally, we planned to use the NVIDIA 256.44 Linux driver for our OpenCL testing, but it had not worked with OpenCL on the NVIDIA GeForce 9400M (MCP79) graphics processor.

Instead we had to use the not-publicly-released NVIDIA 258.19 driver with its initial OpenCL 1.1 implementation where the GeForce 9400M had worked fine with OpenCL.

Unfortunately, due to the NVIDIA OpenCL implementation not supporting the Open Computing Language on the CPU and the ATI Stream SDK not working without the Catalyst driver installed, we were limited to OpenCL GPU testing on Linux.

With Apple Mac OS X we ran OpenCL tests both on the Intel Core 2 Duo P7350 and NVIDIA GeForce 9400M.

Unfortunately, this is the only available Apple hardware we had to test.

Testing was done with the JuliaGPU, MandelGPU, SmallPT GPU, and MandelbulbGPU tests via the Phoronix Test Suite.

These tests provide interesting metrics comparing the OpenCL GPU performance with NVIDIA hardware under Mac OS X and Ubuntu Linux along with a look at the CPU versus GPU OpenCL performance under Mac OS X.

Using OpenCL on the NVIDIA GeForce 9400M was 77% faster than on the dual-core Intel P7350 at 2GHz, which isn’t surprising as although this is a relatively GPU, the Open Computing Language is really designed around exploiting the capabilities of a graphics processor.

What is interesting though is that Apple’s OpenCL GPU implementation was 61% faster than the NVIDIA Linux OpenCL implementation for the JuliaGPU test.

With the MandelGPU test it was actually faster running it on the CPU than the GPU, which may be due to the OpenCL kernel not being well optimized or just the lower-end GPU struggling here, but with the Intel Core 2 Duo its speed was better by 29%.

Comparing the NVIDIA OpenCL performance between Apple Mac OS X and Ubuntu Linux the computational speed was much closer this time around, but Apple’s code was faster by 2.5%, which was statistically significant based upon our multi-run values.

When running SmallPT GPU with the caustic rendering scene, the NVIDIA OpenCL driver under Linux finally was faster than that of Apple’s with a lead of 46%.

Running OpenCL on the CPU under Mac OS X had not worked for this OpenCL benchmark.

Like the MandelGPU test where running the OpenCL kernel was faster with the Intel Core 2 Duo than the weak NVIDIA GPU, with MandelbulbGPU this is the same where on the P7350 it was faster by 21%.

When comparing the NVIDIA GPU performance between two of our favorite operating systems, Ubuntu 10.04.1 LTS with NVIDIA’s binary OpenCL implementation there was a 7% favor to the Linux operating system.

While it’s a pity we are unable to test the ATI/AMD OpenCL support under Mac OS X and Linux due to Apple hardware availability (where we would also be able to test the OpenCL CPU Linux performance with the ATI Stream SDK) or with a more powerful NVIDIA graphics processor on Mac OS X, looking at the Open Computing Language performance with these benchmarks on Mac OS X and Ubuntu Linux was interesting.

With the JuliaGPU test profile it heavily was in favor of Mac OS X but with SmallPT GPU it was heavily in favor of NVIDIA’s Linux OpenCL driver and then the MandelGPU and MandelbulbGPU tests did not have a strong preference to one operating system or the other.

We will be back soon with more ATI/NVIDIA/Intel OpenCL tests under Linux.

Source:http://www.phoronix.com/scan.php?page=article&item=apple_opencl_linux&num=1

Share

1 comment »

Posted in Hardware News

Tags: Computing Linux MAC Nvidia

Nvidia geforce gtx 460 sli performance review

August 26th, 2010

Share

It took more time than expected but finally Nvidia was able to counterattack and turn around the tables on ATI last month with the release of the GeForce GTX 460.

This is the first product on Nvidia’s current-gen offerings that will make their competition sweat.

Seriously, the GeForce GTX 460 has it all, the right price, amazing performance, unbelievable overclocking headroom and excellent efficiency.

Not surprisingly this release forced ATI to make a few price cuts, most prominently the Radeon HD 5830 is now available for $200. But that’s not going to be enough to steal the new GeForce’s thunder.

We still prefer the GeForce GTX 460 as there are many factory overclocked 1GB boards that cost around $230 and deliver performance that can match the Radeon HD 5850 and in some cases the GeForce GTX 470.

When wrapping up our initial GeForce GTX 460 coverage on launch day we mentioned the interesting thought of running two 1GB cards in SLI which would cost a whisker less than $500.

This matches the price of a single GeForce GTX 480 that on paper would appear to be the slower alternative.

Many gamers found a similar situation when the Radeon HD 5770 was released.

When placed in Crossfire mode using two cards, these could keep pace with the more expensive Radeon HD 5850 and in some cases beat the GeForce GTX 470.

For roughly $300 this was a killer combination, but no longer, of course.

Those looking for top notch gaming performance without having to spend more than $400 will likely be interested in what the GeForce GTX 460 (768MB) SLI configuration has to offer.

At $400 the GeForce duo will be slightly more expensive than a single Radeon HD 5870, which recently received a slight price drop and can now be found in the $370 range.

Then further up the ATI ladder we have the Radeon HD 5970 at $670, it will be interesting to see how the GeForce GTX 460 SLI cards compare.

Despite our reluctance in recommending dual graphics cards as an upfront solution in the past, recent tests have showed SLI scales very well in a huge variety of games nowadays.

So are GeForce GTX 460 SLI cards the killer combination for PC gamers today? We’ll find out next.

Source:http://www.techspot.com/review/309-geforce-gtx-460-sli-performance/

Share

Comments Off »

Posted in Hardware News

Tags: computer Hardware Nvidia

Nvidia names georgia institute of technology a cuda center of excellence

August 24th, 2010

Share

NVIDIA today recognized Georgia Institute of Technology (Georgia Tech) as a CUDA Center of Excellence.

One of the world’s premier engineering and science universities, Georgia Tech is engaged in a wide number of research, development and educational activities which leverage GPU Computing.

Jeffrey Vetter, joint professor of the Georgia Tech College of Computing and Group Leader at Oak Ridge National Laboratory, will serve as principal investigator of the CUDA Center of Excellence.

“Georgia Tech has a long history of education and research that depends heavily on the parallel processing capabilities that NVIDIA has introduced with its CUDA architecture,” Vetter said. “This award allows us to focus, what is now a large amount of activity across 25 different research groups, under a single center, which will significantly amplify our research capabilities.”

Georgia Tech’s non-profit research arm, Georgia Tech Research Institute, is also leveraging the capabilities of the GPU in its work with industry and government groups such as the U.S. Defense Department.

“By cross-pollinating ideas and skills, sharing software and hardware facilities, and streamlining interactions with priority access to NVIDIA staff and capabilities, this status will add considerable strength to our research and educational programs,” Vetter added.

NVIDIA and Georgia Tech are already collaborating on a number of projects that will help shape the national science infrastructure. The National Science Foundation Track 2D Keeneland Project will initially deploy a significant system of NVIDIA® Tesla™ processors this year, with a larger, petaflop-class system to be in place by 2012. Georgia Tech and Oak Ridge are also collaborating with NVIDIA in the recently announced DARPA Ubiquitous High Performance Computing program, with the goal of designing an energy efficient “petaflop in a cabinet¿ prototype system in 2018.

One example of the work the University is doing in the field of software tools is “Ocelot”, a compiler that allows CUDA code to run seamlessly on multi-core CPUs. The compiler will be available and distributed through the CCOE and will help to catalyze research on top of this open source infrastructure.

Georgia Tech joins a select group of 10 other universities and research organizations in the U.S. and abroad, including Harvard University, Cambridge University and the Chinese Academy of Sciences, that are designated as a CUDA Center of Excellence. More than 350 universities worldwide teach the CUDA programming model within their curriculum.

CUDA™ is NVIDIA’s computing architecture that enables its GPUs to be programmed using industry standard programming languages and APIs, opening up their massive parallel processing power to a broad range of applications beyond graphics.

Source:http://pr-usa.net/index.php?option=com_content&task=view&id=468040&Itemid=30