With the recent spurt of non-reference design Radeon HD 5770 graphics cards, ASUS seems to be ready with its EAH5770 CuCore, where "CuCore" stands for copper (Cu) core, used in the GPU cooler. Speaking of which, the GPU cooler consists of a GPU base which has 5.8 oz (164.5 g) of copper, for better transfer of heat, to the rest of the heatsink which has radially-projecting aluminum fins. The PCB designed by ASUS seems to be slightly longer than the reference AMD PCB, yet it only makes room for one CrossFire connector, meaning its can pair with only one more card of its kind (or up to three more cards with two CrossFire connectors).

The 40 nm Juniper GPU has DirectX 11 compliance, and is powered by 800 stream processors, and connects to 1 GB of memory across a 128-bit GDDR5 memory interface. Despite the swanky cooling the card sticks to reference AMD clock speeds of 850/1200 MHz (core/memory), but comes with the ASUS VoltageTweak feature that allows software voltage adjustments. The rear-panel is redesigned too, with one of the slots making for a large vent, with the lower slot holding one each of DVI-D, D-Sub, and HDMI. The ASUS EAH5770 CuCore is expected to be priced under 130 EUR.

TUL Corporation, a leading manufacturer of AMD graphics cards, introduced an upgraded addition into the HD5700 series: the PCS+ HD5770. Part of the award-winning PCS series, the PCS+ HD5770 features outstanding factory overclocked ability with core and memory clocks at 875 MHz and 1225MHz respectively and GDDR5 onboard memory. Memory data transmission rates are enhanced up to 4.9Gbps, easily boosting gaming performance.

The PCS+ HD5770 also features an ultra quite 92mm cooling fan which is made of a low-noise impeller, easily dissipating heat from the copper base that fully covers the GPU at lower fan speed and efficiently cools down the temperature up to 10°C compared to that of reference board design. This allows more headroom for overclocking giving gamers maximum performance of their rig.

Adding to the expansive list of non-reference design Radeon HD 5750, is Club3D with its new silent HD 5750 model. The new accelerator carries a red PCB which seems to resemble the reference PCB, but for the color. The GPU is cooled by a large silent heatsink which consists of a GPU block from which four copper heatpipes arise, conveying heat to an aluminum fin array. The memory seems to be cooled by heatsinks, too. It retains the rear panel of the reference design, with two DVI-D and one each of HDMI and DisplayPort.

The rest of its specs are identical to others in its league: DirectX 11 support, 720 stream processors, 1 GB of memory across a 128-bit GDDR5 interface, ATI CrossFireX and Eyefinity support, it asks 140€ for staying quiet.

AMD's lower-mainstream DirectX 11 compliant graphics card slated for Q1-2010, the ATI Radeon HD 5670 has been pictured and detailed, sourced from a [H]ardOCP HardForum community member. The HD 5600 series is based on a 40 nm GPU codenamed "Redwood". From the specifications the GPU-Z screenshot shows, it has a 50% downscaled SIMD engine, with 400 stream processors, while it retains the 128-bit wide GDDR5 memory interface, with 16 ROPs. Assuming the clock speeds shown in the screenshot to be the reference speeds, they are 775 MHz for the core, and 1000 MHz for the 1 GB of memory (resulting in 64 GB/s of memory bandwidth).

An engineering-sample of the card has also been pictured, revealing a red-colored PCB breaking away from the black PCB scheme of the rest of the HD 5000 series. The card draws all its power from the PCI-Express slot. The GPU cooler consists of a simple heatsink with radially-projecting metal fins, in which is nested a fan. Output connectivity includes DVI, HDMI, and D-Sub, though the leads behind the D-Sub connector shows that offering a DisplayPort in its place might be possible.

Elpida Memory, Inc., Japan's leading global supplier of Dynamic Random Access Memory (DRAM), today announced that it had developed a 1-gigabit GDDR5 (product name: EDW1032BABG) that operates at a world-class high speed of 6Gbps. The new graphic memory (GDDR: Graphics Double Data Rate) device marks Elpida's first successful step in the GDDR market.

Applications for GDDR memory devices used with GPU (Graphics Processing Unit) are found not only in such graphic processing equipment as game consoles and PC graphics cards but also in equipment that require high-performance computing for use in such areas as science and technology, physical simulation, digital image processing and video conversion.

Today HIS unleashed the revolutionary new HIS Radeon HD 5970 2GB GDDR5. Building on the success of HIS 5800 series, HIS 5970 offer the power of two 5800 GPUs and 3200 stream processors on a single board. It delivers groundbreaking graphics performance along with the most advanced feature set in the graphics industry.

Powered by the top-of-the-line 5970 GPU, HIS Radeon HD 5970 2GB GDDR5 accelerates PC gaming with more than 4 teraFLOPS computing power! Running at 725MHz core clock speed and 4Gbps memory clock speed, the, HIS Radeon HD 5970 ensure extreme game play at high resolutions and maximum settings. It delivers full support for Microsoft DirectX 11 graphics and compute functionality. Besides, it features an architecture tailored to excel with DirectCompute 11 (Shader Model 5.0) and OpenCL, taking ATI Stream functionality to next level.

World renowned graphics card and mainboard brand and manufacturer, MSI, captured the hearts of powerful performance enthusiasts after unveiling its all-new R5800, R5700 series of graphics cards. Today, MSI unveils its top model, the R5970-P2D2G graphics card, which not only equips two advanced 40nm GPUs, but also utilizes 3200 stream processors (SPs)*, and 2 GB 512 bit** GDDR5 graphics memory, making it the best performing card on the planet today!

Additionally, MSI's R5970-P2D2G graphics card features ATI Eyefinity triple-display output technology, allowing gamers to get fully immersed in the latest DirectX 11 environments. The card also offers PowerPlay power-saving functionality that shows remarkable power-saving when the machine is in stand-by. Equally noteworthy is MSI's own "Afterburner" overclocking software that allows the user to adjust both the voltage and overclocking settings for better performance. With CrossFire X enabled, the dual GPUs offer outstanding, stable performance that redefines fast. Without a doubt, this card will be the first choice of performance enthusiasts.

Punch up your digital experience with a blast of power. The XFX NVIDIA GeForce GT 240 graphics card gives you an across-the-board speed and performance boost that takes you from "mundane" to "insane". NVIDIA PhysX technology adds speed to your game, so that you can play Batman: Arkham Asylum or Darkest of Days 2X faster than the competition. Transform it into a dedicated PhysX card by pairing it with another NVIDIA based graphics card and enjoy up to 50% pop in performance.

Ninety-six NVIDIA CUDA cores enable you to run the latest GPU-accelerated applications, many at speeds 50% faster than prior generation. Maximize your YouTube- and Hulu-viewing experience in full HD with new GPU-accelerated Adobe Flash 10.1. Enjoy all of the great Microsoft Windows 7 features, including drag and drop video conversion, as well as full support for DirectCompute, OpenGL 3.2 and PC Express 2.0. Available in a 1 GB GDDR5 or 512MB DDR3, put the power of the XFX NVIDIA GeForce GT 240 graphics card at your fingertips.

POINT OF VIEW, premier NVIDIA partner for GeForce Graphics cards, is launching today the first GDDR5 GeForce graphics card, the GeForce GT240. Built from the ground up, the latest GeForce 200 series graphics card is a big step forward. Built on 40nm the GPU will consume less energy and have less heat production. This allows better clock speeds, a higher overclocking potential and lower temperatures. The smaller die-size together with the support for DirectX 10.1 and Shader model 4.1 really adds a future proof coating to the GT240.

Another major improvement is the support of GDDR5 which literally multiplies the possible memory clock by two! The new PCB design also has integrated HDMI audio so no additional SPDIF cable is required for rerouting the sound to the HDMI cable.

AMD's dual-GPU flagship graphics accelerator, the Radeon HD 5970, is closer than you think it is. Slated for 18th Nov, it includes every feature that allows AMD to reclaim the performance leadership it yearned for since the beginning of this year. In a series of company slides sourced from XtremeSystems Forums, it is learned that this could be one of the first accelerators which AMD "openly" markets as having a "Massive Headroom" for overclocking. While the clock speeds on the HD 5970 are lower than those on the single-GPU HD 5870, AMD lifted limits on what the driver-level ATI Overdrive software can offer in terms of clock speeds. While the engine (core) and memory speeds are set at 720/1000 MHz, the unlocked ATI Overdrive lets users take the clock speeds all the way up to 1000/1500 MHz. That's 30% for the core, and a stellar 50% for the memory.

To back such speeds, AMD seems to have splurged heavily on top-notch components on the PCB. To begin with, the PCB holds two high-grade AMD Cypress GPUs, each with all its 1600 stream processors enabled. The GDDR5 memory, while clocked at 1000 MHz or 4 GT/s, is technically rated by its manufacturer to run at 1250 MHz or 5 GT/s. All systems are powered by high-grade digital PWM voltage regulators, with independent Volterra VRM controllers that allow real-time monitoring, and software voltage control. Barring the five-odd cylindrical solid-state capacitors, Japanese pure ceramic surface-mount capacitors are extensively made use of.

NVIDIA Corporation today unveiled the Tesla 20-series of parallel processors for the high performance computing (HPC) market, based on its new generation CUDA processor architecture, codenamed "Fermi".

Designed from the ground-up for parallel computing, the NVIDIA Tesla 20-series GPUs slash the cost of computing by delivering the same performance of a traditional CPU-based cluster at one-tenth the cost and one-twentieth the power.

ASUS has developed its own take on AMD's mid-range Radeon HD 5750 graphics accelerator, with the EAH5750 Formula 1 GB. Being a non-reference design product, it makes use of both cooler and PCB indigenously developed by the company. The black PCB seems slightly shorter than the reference PCB, and has its power connector located on the top, rather than on the end. The Radeon HD 5750 GPU features DirectX 11 compliance, 720 stream processors, and a 128-bit GDDR5 memory interface to connect to the 1 GB of memory on this card. It sticks to AMD's reference clock speeds of 700/1050 MHz (core/memory).

The other, more important selling point of this card is its custom designed GPU cooler. Apparently the design involves a GPU contact block from which metal fins project radially, on which a PWM-controlled fan circulates air. Characteristic to ASUS' Formula series graphics cards, the cooler shroud coarsely resembles a Formula One racing car. The connectors are all rounded off onto one metal bracket, although the card still needs two expansion slots. Connectors include one each of DVI-D, D-Sub, and HDMI. The price is expected to be under € 120.

Elpida Memory, Inc., Japan's leading global supplier of Dynamic Random Access Memory (DRAM), and Winbond Electronics Corp. today announced that they have signed a Memorandum of Understanding (MOU) for DRAM foundry services that will see Winbond manufacture DRAMs under contract to Elpida. The outsourcing agreement is the first step of a business partnership the two companies intend to pursue further.

The agreement pertains to the supply of GDDR3 and GDDR5 graphic DRAMs for Elpida. Prior to this agreement Winbond and Elpida have been working together to commercialize GDDR3 and GDDR5 DRAM products. Winbond is preparing to begin commercial production of these products by the end of 2009 while Elpida plans to begin purchasing the output in the first half of 2010.

HIS today rolled out its own designed graphics accelerator based on the Radeon HD 5750 graphics processor. The HIS HD 5750 iCooler IV is a completely non-reference design, including cooler and PCB. The design involves use of the company's iCooler IV GPU cooler the company has used on mid-range accelerators in the past, the blue PCB, and an overall compact design, including a rear-panel cluster that breaks away from AMD's connector cluster design for the Radeon HD 5000 series. Instead, the card uses a single bracket that includes one each of DVI-D, DisplayPort, and HDMI, although the card itself occupies two slots due to its cooler design.

The Radeon HD 5750 GPU has 720 stream processors available, and is wired to 1 GB of GDDR5 memory across a 128-bit wide interface. It is DirectX 11 compliant, and supports other technologies such as DirectCompute 11, and AMD's Eyefinity, and CrossFireX technologies. The clock speeds of the card stick to AMD's reference speeds, those of 700 MHz (core), and 1150 MHz (effective 4600 MHz, memory). HIS bundles a Collin McRae DIRT-2 coupon, which can be redeemed to get a copy of the game when it's out. Expect standard pricing.

TUL Corporation, a leading manufacturer of AMD graphics cards, is excited to add liquid cooling to the world's fastest single GPU: the PowerColor Liquid Cooling Solution (LCS) HD5870 1GB GDDR5. Equipped with the award-winning EK cooling solution, the LCS HD5870 offers outstanding thermal performance in a single slot design, enhancing overclocking and allowing optimum stability.

The PowerColor LCS HD5870 has a full water block mounted on the card, fully covering the memory and power regulator chip. With a copper base design, it is able to reduce temperatures in the most efficient way possible. PowerColor packages a high-flow 3/8" and 1/2" fittings (barbs) to maximize water flow with captured o-rings to prevent leakage. Gamers can easily customize their own liquid cooling system using these two fittings.

Now, more people than ever can experience real innovation in DirectX 11 gaming with the Club 3D HD5700 graphics card series. Loaded with advanced technology, this graphics card has the power and premium features you need for fully immersive gameplay. Expand your visual real estate across up to three displays and get lost in the action with revolutionary ATI Eyefinity technology. Using ATI Stream technology, accelerate even the most demanding applications and do more then ever with our PC. Enable realistic visual effects and an explosive HD gaming performance to dominate the competition.

Enjoy an incredibly immersive HD gaming/computing experience with ATI Eyefinity Technology. This revolutionary multi-display technology allows you to expand your visual real estate across up to three displays with innovative "wrap around" capabilities to maximize your field of view with incredible sharpness and clarity. Perform tasks on a PC faster with ATI Stream technology, speed up demanding everyday applications and get more done in less time. The first fully Microsoft DirectX 11-compatible GPUs, Club 3D HD5800 Series is designed to deliver ultimate HD gaming performance for the latest games and DirectX 11 applications so you can dominate the competition.

Following the storming success of the recently introduced SAPPHIRE HD 5800 series, SAPPHIRE Technology now introduces the HD 5700 series, setting new standards for performance, value for money and features.

Only the second product line to fully support the forthcoming DirectX 11 from Microsoft, the new HD 5700 series brings many of the exciting architectural and performance features of the HD 5800 series to users on lower budgets. The HD 5700 series is based on a new GPU from the ATI division of AMD built in low power 40nm technology. This new series includes support for the new instruction set of DirectCompute 11, hardware tessellation, accelerated multi-threading capability, support for ATI Stream and the ability to display across up to three monitors simultaneously with ATI Eyefinity.

AMD today released its ATI Radeon HD 5700 series of mid-range graphics processors aimed to increase the company's competitiveness in the sub-$200 market, and present a mainstream graphics card series compliant with the DirectX 11 API. The series makes use of AMD's 40 nm Juniper graphics processor, that packs 1.04 billion transistors, 800 stream processors, 40 texture units, and 16 raster operation units. The 128-bit wide GDDR5 memory interface provides up to 76.8 GB/s of memory bandwidth.

The higher-end product of the two on offer, Radeon HD 5770 is priced around the $150 mark, and has all its 800 stream processors and 40 texture units enabled. It comes with clock speeds of 850/1200 MHz (core/memory), while the Radeon HD 5750 looks to cash in on the $120 point. It has 720 stream processors, 36 texture units enabled from its kitty, with clock speeds of 700/1150 MHz (core/shader). While the Radeon HD 5770 comes with 1 GB of memory, the Radeon HD 5750 comes in variants of 1 GB and 512 MB. Some of the other big selling points of these cards are low rater power consumptions of 16~18 W at idle, and 86~108 W at load. The connectivity options include two DVI-D, and one each of DisplayPort and HDMI. The cards support AMD's Eyefinity technology which allows combining three 2560x1600 pixel display-heads for a single large display head. Each of these can pair with up to three more of its kind for 4-way CrossFireX multi-GPU performance upscaling.

All major AMD board vendors, such as ASUS, Sapphire, HIS, PowerColor, Gigabyte, MSI, VisionTek, XFX, among others, have their offerings based on these GPUs ready to sell from today. Pictured below (in order) are Radeon HD 5770 and Radeon HD 5750.

NVIDIA's upcoming flagship graphics processor is going by a lot of codenames. While some call it the GF100, others GT300 (based on the present nomenclature), what is certain that the NVIDIA has given the architecture an internal name of "Fermi", after the Italian physicist Enrico Fermi, the inventor of the nuclear reactor. It doesn't come as a surprise, that the codename of the board itself is going to be called "reactor", according to some sources.

Based on information gathered so far about GT300/Fermi, here's what's packed into it:

• Transistor count of over 3 billion
• Built on the 40 nm TSMC process
• 512 shader processors (which NVIDIA may refer to as "CUDA cores")
• 32 cores per core cluster
• 384-bit GDDR5 memory interface
• 1 MB L1 cache memory, 768 KB L2 unified cache memory
• Up to 6 GB of total memory, 1.5 GB can be expected for the consumer graphics variant
• Half Speed IEEE 754 Double Precision floating point
• Native support for execution of C (CUDA), C++, Fortran, support for DirectCompute 11, DirectX 11, OpenGL 3.1, and OpenCL

Earlier talked about as being a mere speed bump over RV740 that probably locks horns with Radeon HD 4850 or even HD 4870 512 MB in terms of performance in today's applications, Juniper, codename for AMD's newer mainstream GPU that is looking to handle things under the $200 segment, is looking stronger than thought about before on paper, according to recent publication by Anandtech. Breaking away from the previous conception of a GPU with 800 stream processors, a 128-bit GDDR5 memory, and most other vital-stats resembling its ancestor, the report suggests something that's at least 75% as powerful as Cypress.

It is said to feature 14 SIMD blocks, which going by Cypress' architecture would imply 1120 stream processors (14 x 80 SPs). The report also hints at a stronger memory sub-system to keep up. Instead of a 128-bit wide interface, it hints at 192-bit, GDDR5 at that. This part doesn't seem convincing, as earlier pictures of the card showed it to be holding four memory chips on the reverse side of the PCB, with no fuses to show memory chips that aren't sharing memory channels. Accelerators with 192-bit wide memory interface, should have 6 or 12 memory chips typically. Other numbers include 56 texture memory units (TMUs), and 24 raster operation units (ROPs), if there is a 192-bit memory interface. Two accelerators based on the GPU, Radeon HD 5770 and Radeon HD 5750 should be out around the same time Windows 7 hits retail. Both target sub-$200 price points.

AMD today launched the most powerful processor ever created, found in its next-generation graphics cards, the ATI Radeon HD 5800 series graphics cards, and the world's first and only to fully support Microsoft DirectX 11, the new gaming and compute standard shipping shortly with Microsoft Windows 7 operating system. Boasting up to 2.72 TeraFLOPS of compute power, the ATI Radeon HD 5800 series effectively doubles the value consumers can expect of their graphics purchases, delivering twice the performance-per-dollar of previous generations of graphics products. AMD will initially release two cards: the ATI Radeon HD 5870 and the ATI Radeon HD 5850, each with 1GB GDDR5 memory. With the ATI Radeon HD 5800 series of graphics cards, PC users can expand their computing experience with ATI Eyefinity multi-display technology, accelerate their computing experience with ATI Stream technology, and dominate the competition with superior gaming performance and full support of Microsoft DirectX 11, making it a "must-have" consumer purchase just in time for Microsoft Windows 7 operating system.

"With the ATI Radeon HD 5800 series of graphics cars driven by the most powerful processor on the planet, AMD is changing the game, both in terms of performance and the experience," said Rick Bergman, senior vice president and general manager, Products Group, AMD. "As the first to market with full DirectX 11 support, an unmatched experience made possible with ATI Eyefinity technology, and ATI Stream technology harnessing open standards designed to help make Windows 7 that much better, I can say with confidence that AMD is the undisputed leader in graphics once more."

Unbeknownst to other markets, ATI AIB partners have been working on full-fledged lineups of graphics cards based on the Radeon HD 4860 GPU, a brand that seems to be specific to the Chinese market. Sapphire developed a new premium offering based on the GPU, called Sapphire HD 4860 Platinum, that intends to be a cut above the rest with its cooling and better power stability. The card in many respects looks similar to "Vapor-X" branded offerings based on other GPUs, by Sapphire.

The card sports a dual-slot design with a custom PCB by Sapphire. The PCB includes a 5+2 phase VRM which draws power from two 6-pin PCI-E connectors. The RV790-based Radeon HD 4860 GPU has 640 stream processors, and has core/memory clock-speeds of 700/750 MHz. With its 256-bit wide GDDR5 memory interface, it connects to 512 MB of memory. The GPU is cooled by a robust heatpipe-fin array based GPU cooler. On the connectivity front, the card provides DVI-D, HDMI, and DisplayPort. It is available for RMB 799 (US $117) in China. More pictures can be found at the source.

Force 3D rolled out a refreshing take on the ATI Radeon HD 4890 with its own-design F4890P-G5-350-FS. This model offers all features of the GPU running at AMD reference clock speeds of 850/975 MHz (core/memory), while needing only one 6-pin power connector, while most other accelerators based on the GPU require two, some overclocked models even needing a 6+2 pin power connector to replace one of them. This card has a slim design, with a short PCB, and a decent Arctic Cooling Accelero Twin-Turbo Pro cooling the GPU. It holds 1 GB of GDDR5 memory, and includes DVI-D, HDMI and D-Sub for connectivity. It has started selling in Japan for a fairly standard price.

Gigabyte has a custom-design version of the ATI Radeon HD 4770 accelerator ready, packed with 1 GB of GDDR5 memory instead of the usual 512 MB. Already listed by several German online stores, Gigabyte GV-R477UD-1GI is selling for as low as 114.43 EUR. Although there are no pictures of the actual card, we can infer from the model name that it is going to have a custom design, featuring Gigabyte's Ultra Durable Video technology that makes use of durable components, and perhaps the signature Gigabyte-blue PCB.

Apart from the expanded memory, the rest of its specifications remain the same, including clock speeds of 750/800 MHz (core/memory). Notable specs of the Radeon HD 4770 include 640 stream processors, DirectX 10.1 support, and ATI CrossFireX support. A global launch of the accelerator can be expected soon.

In a release to its AIB partners, AMD listed out the key features of its high performance GPU in the Evergreen family, codenamed "Cypress". While not getting into the GPU specifications, it lists out the key technologies the GPUs support. It comes as no surprise that AMD will name Cypress-XT and Cypress-Pro as Radeon HD 5870 and Radeon HD 5850, a model number scheme that's been running for the past two generations. While product launches, tentatively on September 22, and (p)reviews keep us busy in September, retail availability can be expected only in October, just in time for that of Windows 7.

• 1GB GDDR5 memory
• ATI Eyefinity technology with support for up to three displays
• ATI Stream technology
• Designed for DirectCompute 5.0 and OpenCL
• Accelerated Video Transcoding (AVT)
• Compliant with DirectX 11 and earlier revisions,supports OpenGL 3.1
• ATI CrossFireX multi-GPU support for highly scalable performance
• ATI Avivo HD video and display technology
• Dynamic power management with ATI PowerPlay technology
• 2x DL-DVI, DisplayPort, HDMI
• PCI Express 2.0 interface