One year after its debut, Epic Games' Samaritan Tech Demo is back at the GDC (Game Developers Conference) and it's there to tease the power of NVIDIA's Kepler 28 nm GPU. At last year's GDC Samaritan, which is based on Unreal Engine 3, was running in real-time on three GeForce GTX 580 cards set in SLI. This year Epic is showing off the demo running on a single NVIDIA Kepler graphics card.

Waiting on Kepler before making a new GPU purchase? Well, you have to wait a little longer. Thankfully, this wait can be eased with the latest leaks about NVIDIA's 28 nm chip and the GeForce GTX 680 it powers.

According to VR-Zone, the GTX 680 does indeed feature 1536 CUDA Cores and a 256-bit memory interface, but it also has hotclocks, meaning the GPU is set to 705 MHz but the shaders operate at 1411 MHz. The memory (2 GB most likely) is supposed to be clocked at 6000 MHz giving a total memory bandwidth of 192 GB/s.

NVIDIA's incoming card is 10 inches long and also has 3-way SLI support, and four display outputs - two DVI, one HDMI and one DisplayPort. The GeForce GTX 680 is expected to be revealed on March 12 and should become available on March 23rd.

Thanks to some good understanding of geometry, and great Photoshop skills, the punters have estimated the die-size of NVIDIA's GeForce Kepler 104 (GK104) GPU, using a picture of the NVIDIA reference board that was leaked last week. Compared to the known die-size of an AMD "Tahiti"-based graphics card, the die-size of GK104 was estimated to be around 320 mm², about 45 mm² smaller than Tahiti. The relative transistor density of the chip, compared to AMD's 28 nm chips, were also estimated.

CyberpowerPC Inc. www.CyberpowerPC.com, a global manufacturer of custom gaming machines, today announced it will feature AMD's new Radeon HD 7870 GHz Edition and AMD Radeon HD 7850 GPUs across its entire desktop gaming PC line.

CyberpowerPC thinks it's time to broaden your horizons and tap into the full potential of your GPU. The AMD Radeon HD 7870 and 7850 have been engineered to be the world's most advanced graphics cards. They feature the award-winning GCN Architecture - the industry's first 28 nm GPU design with full support for DirectX 11 to power the next generation of high-def games and multimedia.

VTX3D, a leading brand of graphics card maker today adds the latest HD7800 series to its premium graphics cards selection. Being a new force of HD7000 family, both models of HD7800 series feature revolutionary GPU design, the latest PCI Express 3.0 platform, also advanced DirectX○R 11.1. All these cutting-edge technologies make VTX3D HD7800 series as a perfect choice for PC enthusiasts.

Based on the AMD 28 nm processor, the VTX3D HD7800 series includes two models with 2 GB of GDDR5 memory. VTX3D HD7870 GHz Edition runs at clock speed of 1000 MHz, and HD7850 are clocked at 860 MHz, both together with 4.8 Gbps for the memory delivering class leading performance. What's more, the VTX HD7800 series is ready for CrossFire mode, gamers can get exceptional performance scaling with a second GPU.

Club3D announced its lineup of Radeon HD 7800 series graphics cards, the Radeon HD 7850 (model: CGAX-7856), and Radeon HD 7870 (CGAX-7876). Both models follow AMD reference design to the letter, there's little AIB vendor branding on these cards. The cards stick to AMD reference clock speeds, that's 1000/1200(4800) MHz (core/memory(effective)) for the HD 7870, and 860/1200(4800) MHz for the HD 7850. Based on the 28 nm "Pitcairn" silicon, the Club3D HD 7870 packs 1280 stream processors, and 2 GB of GDDR5 memory, over a 256-bit wide memory interface. The HD 7850 has 1024 stream processors. In all likelihood, Club3D's cards also stick to AMD's target pricing of US $349 for the HD 7870, and $249 for the HD 7850.

AMD just launched the Radeon HD 7800 series graphics cards, consisting of two models, the Radeon HD 7870, and the Radeon HD 7850, targeting crucial price-points that appeal PC gamers. The two models are based on new 28 nm silicon by AMD, codenamed "Pitcairn," which packs 2.8 billion transistors. The new chip takes advantage of AMD's Graphics CoreNext architecture.

The Radeon HD 7870 packs 1280 GCN stream processors, 80 texture memory units (TMUs), 32 raster operations units, and a 256-bit wide GDDR5 memory interface, holding 2 GB of memory. This card is AMD's second "GHz Edition" model (after the Radeon HD 7770), sporting a core clock speed of 1 GHz. The memory is clocked at 1200 MHz (4.80 GHz GDDR5 effective).

LSI Corporation today announced it is sampling the industry's first 28 nm system-on-a-chip (SoC) for the desktop and mobile HDD market segments. The transition to 28 nm SoC technology provides a cost-effective way to increase the amount of data that can be stored on a hard drive by enabling higher areal density and yield through superior signal-to-noise ratio performance.

"The explosion of digital content driven by rich media such as high-definition video and photography is challenging HDD manufacturers to deliver higher-capacity, more energy-efficient drives," said Phil Brace, senior vice president and general manager, Storage Peripherals Division, LSI. "By being first to sample 28 nm SoC technology, we're offering HDD manufacturers a low-risk, time-to-market advantage in meeting next-generation HDD capacity points while staying within the power envelope."

ARM today announced the availability of the ARM Cortex-A9 MPCore Processor Optimization Pack (POP) for GLOBALFOUNDRIES' 28 nm-SLP High-K Metal Gate process technology. Optimized for mobile, networking and enterprise applications, the energy-efficient ARM POP 28 nm-SLP for Cortex-A9 processors delivers a performance range from 1 GHz to 1.6 GHz for worst case conditions, with up to 2 GHz in typical conditions. This provides a wide range of flexibility for System-on-Chip (SoC) designers to optimize performance and energy-efficiency using the ARM Artisan Physical IP Platform and Cortex-A9 POP.

The launch of AMD's Radeon 7800 series is on course for March, as AMD wants to complete the launches of the entire Radeon 7000 series before NVIDIA even has its first GPU out. Radeon HD 7800 will be designed to occupy key price points in the sub-$300 market segment, where it strikes price-performance sweetspots for gamers. Central to this series is a new 28 nm GPU, codenamed "Pitcairn", from which will be derived three SKUs: the Radeon HD 7870, Radeon HD 7850 2 GB, and Radeon HD 7850 1 GB. The specifications look like this:

Radeon HD 7850

• 20 Graphics CoreNext Compute Units, 1280 stream processors
• 80 TMUs, 24 ROPs (de-linked from the memory bus, of course)
• 256-bit wide GDDR5 memory interface, 2 GB and 1 GB variants
• Clock speeds of 900 MHz core, 1250 MHz (5.00 GHz effective) memory

Radeon HD 7870 specifications follow.

AMD's Radeon HD 7770 mid-range graphics card, which is slated for a little later this month, got its first public shot at 3DMark 11. The card was put through the Performance preset of the benchmark, where it scored P3535 points. The bench was driven by an Intel Core i7-3960X processor. The reviewer also took GPU-Z screenshots of this card, revealing low core temperature. Based on the 28 nm Cape Verde GPU, the HD 7700 is said to have 640 Graphics CoreNext stream processors, and 1 GB of memory over a 128-bit wide GDDR5 memory interface. It is designed for sub-$200 price points.

While Intel and AMD are making efforts to come up with low-power x86 processor platforms to compete with the plethora of ARM processor vendors, manufacturers of the target devices of these low-power x86 processors - tablets, netbooks, and smartphones; note that the architecture simply isn't competitive due to its prices. ARM processors are manufactured by a variety of companies, in a variety of different SoC configurations, and as such the tough competition among these companies ensure ARM processor platforms are comparatively cheaper to low-power x86 ones.

Intel recently debuted its 32 nm "Medfield" Atom processors, with power consumption as low as 11W for the platform. In the second half of 2012, it will launch another line of processors with under 10W power consumption, for high-end smartphones. AMD, on the other hand, will unveil "Hondo", which combines its x86 architecture with Radeon graphics IP, and a power consumption target of less than 5W. In 2013, it plans to launch the "Temash" APU, with power consumption under 2W, and built on the 28 nm process.

Ushering the finest and most capable graphics card based on new AMD "Tahiti" 28nm GPUs, the ASUS HD 7970 DirectCU II TOP uses a factory-overclocked core and 3 GB of GDDR5. It is cooled by the exclusive DirectCU II, a dual fan thermal design with six copper heatpipes and a large dissipation area that leads to the lowest temperatures and quietest operation of any HD 7970 card.

ASUS DIGI+ VRM 12-phase digital power delivery comes to graphics cards improved with Super Alloy Power components, while overclockers are bound to appreciate the inclusion of exclusive VGA Hotwire hardware overvolting and GPU Tweak software tuning. The new cards further support adapter-free AMD Eyefinity 6.

Today at its annual Financial Analyst Day, AMD has presented an updated roadmap detailing the hardware it plans to bring to the table during 2012 and 2013. For this year, the Sunnyvale-based company is preparing a processor quarter which includes the Trinity, Brazos 2.0 and Hondo APUs (accelerated processing units) and the Vishera CPU.

Set to be the main weapon in AMD's x86 arsenal, the Trinity APU (aka the 2nd gen A Series) is made on 32 nm process technology, and features DirectX 11 graphics, two/four Piledriver cores (Piledrive is said to deliver 25% better performance than the Stars cores found in Llano APUs), and a TDP that can go as low as 17 W on mobile parts. Trinity is expected to debut in Q2 and already has close to 100 design wins (more than Llano had before its release).

AMD launched the Radeon HD 7950 graphics card, its second high-end product based on its new Graphics CoreNext architecture. Based on the same 28 nm "Tahiti" silicon as the HD 7970, the new SKU packs all its features, while being a tone-down in terms of specifications. To begin with, it features 1792 stream processors brought about by enabling 28 out of 32 of the GCN compute units on Tahiti. It features 112 TMUs, 32 ROPs, and a 384-bit wide GDDR5 memory interface, holding 3 GB of memory.

Its reference clock speeds include 800 MHz core, and 1250 MHz (5.00 GHz GDDR5 effective). Due to reduced power draw, the card requires two 6-pin power inputs, instead of 6+8-pin found on the HD 7970. AIB partners are free to provide custom-design and factory-overclocked models at speeds of up to 900 MHz core with 5.00 GHz memory. Models with AMD reference clock speeds should be typically priced at US $449, with factory-overclocked ones about $15-20 higher. The cards should be available in the markets right away.

Not to be outdone by their US-based brethren, several European stores have jumped the gun and put up for pre-order the (still not officially announced) Radeon HD 7950 aka AMD's second 28 nm-powered graphics card. The listings in Europe reveal three HD 7950s, two that seem to be based on AMD's reference design (one from XFX, one only carrying the AMD brand) and a custom model coming from Gigabyte.

All three cards feature DirectX 11.1 and OpenGL 4.2 support, 1792 Stream Processors, a 384-bit memory interface, 3 GB of GDDR5 VRAM clocked at 5000 MHz, CrossFireX support, and include four display outputs - DVI x 1, HDMI x 1, mini DisplayPort x 2. The 'stock' models have a GPU clock of 800 MHz while Gigabyte's card (GV-R795W3-3GD) is listed with a core clock of 900 MHz (although it may be a typo). The GV-R795W3-3GD also has a blue PCB and a dual-slot, triple-fan (WindForce style) cooler.

Here is the first GPU-Z screenshot of a Radeon HD 7950 graphics card. Although put into a screenshot with ASUS GPU Tweak tool and Republic of Gamers-themed GPU-Z, the card doesn't appear to be an ASUS-made one, and is more likely HIS or PowerColor. The device ID checks out with the one HD 7950 has been associated with. Most other features match expectations. The HD 7950 is carved out of the 28 nm "Tahiti" GPU, with 28 GCN compute units (CUs) active, totaling 1,792 stream processors and 112 TMUs. The ROP count is untouched at 32, so is the memory, that's 3 GB GDDR5 across a 384-bit wide memory interface. With a memory clock of 5.00 GHz effective, it's churning out 240 GB/s of memory bandwidth. It's just the core clock speed where we weren't on target (from the previously-expected 800 MHz), either 880 MHz is the reference core clock speed of HD 7950, or this particular card is a factory-OC variant (PowerColor HD 7950 PCS?).

Making silicon chips is not easy, requiring hugely expensive fabs, with massive clean-room environments and at every process shrink, the complexity and difficulty of making the things goes up significantly. It looks like TSMC and GlobalFoundries are both having serious yield problems with their 28 nm process nodes, according to Mike Bryant, technology analyst at Future Horizons and this is causing a rash of non-working wafers - to the point of having nothing working with some chip designs submitted for production. It seems that the root cause of these problems are to do with the pressures of bringing products to market, rather than an inherent problem with the technology; it just takes time that they haven't got to iron out the kinks and they're getting stuck: "Foundries have come under pressure to release cell libraries too early - which end up with designs that don't work," Bryant said. In an effort to try and be seen to treat every customer equally, TSMC is attempting to launch ten 28 nm designs from seven companies, but it's not working out too well: "At 45-nm, only NVIDIA was affected. At 28-nm any problems for TSMC will be problems for many customers" said Bryant.

Gigabyte Technology has now added to its website a second Radeon HD 7970 graphics card, the (previously-leaked) model codenamed GV-R797OC-3GD which features a dual-slot, triple-fan cooler, a custom PCB, and a GPU clocked at 1000 MHz (the standard frequency is 925 MHz).

Gigabyte's customized HD 7970 also has an Ultra Durable VGA construction (2oz copper PCB, Tier 1 memory, Japanese Solid Capacitor, Ferrite Core Chokes, Low RDS (on) MOSFET), a 384-bit memory interface, 3 GB of GDDR5 VRAM set to 5500 MHz (no overclock here, unfortunately), DVI, HDMI and dual Mini DisplayPort outputs, and CrossFireX support. The GV-R797OC-3GD's price tag has yet to be revealed.

Originally slated for 9th January, and reportedly delayed to February, AMD's second high-end graphics card based on the Tahiti silicon, the Radeon HD 7950 will make it for a late-January launch, sources within add-in board (AIB) partners told DigiTimes. Previous reports mentioned that Radeon HD 7950, when it is launched, will be accompanied with market-availability. In related news, while all Radeon HD 7000M mobile GPUs launched by AMD so far are rebrands from previous generations, it does have designs of 28 nm mobile GPUs based on its new Graphics CoreNext (GCN) architecture in the pipeline, which will take up Radeon HD 7900M/7800M/7700M series, which will be launched in Q2, 2012.

This week the desktop space has officially entered the 28 nm GPU era thanks to the retail release of the Radeon HD 7970. It's a significant milestone for the discrete graphics market and for AMD, but it's only the beginning as the Sunnyvale-based company is currently preparing the arrival of the Radeon HD 7950 and is planning the introduction of the first 28 nm mobile chips.

While the HD 7950 will debut this quarter, the 28 nm GPUs for notebooks will have to wait a little longer, until Q2. The upcoming mobile parts are of course based on the GCN architecture and will bring DirectX 11.1 support, as well as power, image quality and display output enhancements.

The 28 nm mobile cards will (most) likely be added to the recently-introduced Radeon HD 7000M family which also includes multiple 40 nm-based cards (7600M, 7500M, 7400M, 7300M). Look for 28 nm models to be branded Radeon HD 7700M and higher.

That Dodgy Intel Ivy Bridge DX11 'demo' at CES 2012 (UPDATED)

Word has been flying round the internet about Intel's dodgy Ivy Bridge DX11 'demo'. Intel's Mooly Eden, VP, PC Client Group was attempting to demonstrate a racing game on a prototype laptop - 'ultrabook' - fitted with an upcoming 22 nm Ivy Bridge processor with a racing wheel attached and allegedly rendering DX11 graphics. However, as is very apparent at the start, it's actually a video, because the control panel for the free VLC video player pops up for a few seconds. Eden then 'drives' a car and after a few seconds puts up one hand and then the other, because as he says "they are driving it from backstage". However, there was no one driving the game "backstage", as it was just a video and Eden doesn't say anything about this at any point in the presentation.

This gives conspiracy theorists lots of ammunition, as perhaps the game was actually played on a high powered desktop PC with NVIDIA or AMD discrete graphics cards? What game was it? Eden doesn't say. "IB can't really do these graphics!" they cry and so on. Sure, man 'didn't' go to the moon, either... However, we believe that while yes, there was a bit of deception going on, it was nothing more than a white(ish) lie. Why? Because Ivy Bridge comes out in April and people aren't going to forget this demo. They will immediately put IBs DX11 graphics to the test with similar games and if it doesn't deliver, Intel will have a lot of egg on its face. Here's what Intel had to say about this demo in an official statement:

MAINGEAR, an award-winning PC system builder of custom desktops, notebooks, and workstations, gives PC enthusiasts a reason to grin by offering the new AMD Radeon HD 7950 and AMD Radeon HD 7970 graphics solutions on their award-winning desktop systems.

Over the past few days, we were hearing rumors from many quarters that AMD's "Tahiti" high-performance GPU may have been a deviation from an older specification, and that it really has 2304 stream processors spread across 40 GCN compute units (CUs), instead of the 32 the Radeon HD 7970 ended up with. Both AMD and NVIDIA create more redundant components on their chips than their SKUs end up getting, so they could increase yields, it's a process commonly known as "harvesting".

On Tuesday, AMD quashed the rumor in an e-mail to Bright Side of News, in which it said that Tahiti XT (Radeon HD 7970) makes use of all the CUs there are, on the chip. The 40 CU / 2308 SP rumor gained some weight with the fact that since AMD is venturing into unknown territory (TSMC's 28 nm process, built after quite some delays and failures), it could do some heavy harvesting. Examples of harvesting in recent past include Intel Sandy Bridge-E Core i7 processors, which use only up to 6 out of 8 cores on the silicon, and only up to 15 MB out of 20 MB available on it; and GeForce GTX 480, which used only 480 out of 512 CUDA cores available on the GF100 GPU.

Welcome to the first TechPowerUp news post of 2012! Read on for a couple of impressive overclocking feats with the HD 7970 graphics card.

It looks like the new AMD Radeon HD 7970 could be a bit of a dark horse and a lot more potent than its stock specifications would suggest - excellent for creating a competitive graphics card market. The reviews at stock speeds show the flagship HD 7970 to be around 10-15% faster than NVIDIA's flagship GTX 580, which doesn't seem all that impressive since the GTX 580 has been on the market for over a year now. However, what the reviews haven't really shown, is what kind of an overclocking monster the HD 7970 is. It definitely looks like AMD could have easily beaten the GTX 580 by a much bigger margin than they did, had they wanted to and it makes one wonder why they didn't.

VR-Zone have spent the New Year weekend overclocking this beast, having reached a whopping 1.26 GHz core clock speed with their HD 7970 - and decent benchmark improvements to go with it. Also, with the fan at 100%, the card never got above a very comfortable 68 degrees centigrade while running Furmark, which is amazing considering how this test is specifically designed to heat a graphics card to the max - but please see the update at the bottom of the article. The stock cooler may be noisy, but it's certainly very effective: an excellent result which will prolong the working life of the card.