In the run up for NVIDIA's September 12th launch of mainstream GeForce 400 series products, Chinese technology portal Zol.com.cn shared a picture of the company's GF106 graphics processor, which will be at the heart of several mainstream SKUs, including the talked about GeForce GTS 450. The first three desktop implementations of the GF106 are known to be the GeForce GTS 440, GTS 445, and GTS 450. Of these, the GTS 445 and GTS 440 will be launched first, with the GTS 450 launched just a little later.

The 40 nm GF106 GPU returns to a square package, with a die measuring around 240 mm², which makes it about 45% bigger than that of AMD's Juniper GPU, on which the company's ATI Radeon HD 5700 series is based on. Zol.com.cn suggested that the performance of the GTS 450 in DirectX 9/10 applications can be expected to be comparable to that of the GeForce GTX 260 (216 SP). After launching the GTS 450, NVIDIA will continue on its launch itinerary for the year with a value-segment GF108 GPU.

Following a recent exposé of pictures and performance figures of the Radeon HD 5750, another one covering that of the ASUS Radeon HD 5770 has surfaced. Using a test bed powered by an AMD Phenom II X4 945, 4 GB of DDR3-1333 memory, and Windows 7 64-bit, a member of the Chinese PC enthusiast portal community MyMyPC.com put an ASUS Radeon HD 5770 accelerator through 3DMark Vantage Performance preset (to yield its GPU score), 3DMark06, and FurMark (to check temperatures). It was compared to other popular graphics accelerators in (or around) the sub-$200 league, including Radeon HD 4890, GeForce GTX 260, and GeForce GTX 275. While in the 3DMark06 test the Radeon HD 5770 edges past the GeForce GTX 260, with 3DMark Vantage (GPU score), it lags behind the rest of the league, by at least around 1000 points. This gives an indication that as far as performance goes, the Radeon HD 5770 could be comparable to the Radeon HD 4870, at least in these applications.

Zotac's design team is at it again, releasing a newer graphics card design based on the NVIDIA GeForce GTX 260 GPU. This one, concentrates on making the card shorter, lengthwise. The shorter graphics card also makes do with just one 8-pin PCI-Express power connector. The PCB length is lopped from the typical 26.5 cm to 23 cm, making it easier to fin in smaller cases. With the PCB compact, several heat-producing components are moved closer to each other, making it easier to cool with a smaller common cooler base plate.

Constantly innovating new graphics cards designs, Galaxy pushed the limits of engineering by coming up with the first GeForce GTX 260 with a single-slot air cooler. To achieve this, the company designed its own long cooler based on the Vapor Chamber (VC) technology. A VC plate sits on top of the GPU and the aluminum base-plate of the cooler (to cool the memory) On top of the VC plate is a dense copper channel array through which air directed by the blower passes. There is a separate heatsink over the VRM area, which cools off under the air-flow from the main cooler.

The entire set of heatsinks is neatly packed under an aluminum shroud. The card packs standard features of the GeForce GTX 260, including 216 shader processors, 896 MB of GDDR3 memory across a 448-bit memory interface, 3-way SLI support, and support for DirectX 10. Connectivity options include one each of DVI-D, HDMI, and S-Video/composite. It draws power from two 6-pin PCI-E power connectors. Availability and pricing of this card is not known just as yet.

ASUS is building on its Glaciator Fansink graphics card cooler design, with its newest implementation on an NVIDIA GeForce GTX 260 graphics card. Dubbed Glaciator+ GTX 260 (model: ENGTX260 GL+/HTDI/896MD3), the new card makes use of the Glaciator+ Fansink cooler. The cooler is simpler than its predecessors in having a 5.8 oz copper GPU block from which anodized aluminum fins project radially. A PWM-controlled fan is nested in the center. While the NVIO2 processor gets its own little heatsink, the memory cools off under the fan's air-flow. The 4+1 phase VRM cools passively. The card uses reference NVIDIA clock speeds of 576/1242/999 MHz (core/shader/memory). The 55 nm G200b core provides 216 shader processors, and is wired to 896 MB of GDDR3 memory across a 448-bit interface. ASUS ENGTX260 GL+/HTDI/896MD3 has started selling in the US for $172.99.Images Courtesy: Newegg.com

NVIDIA's SLI multi-GPU technology served as the biggest selling point of nForce series chipsets, as it was exclusive to it. With the advent of LGA-1366 processors and the Intel X58 chipset, NVIDIA allowed the Intel chipset to support the technology, as it soon became clear that it isn't going to be easy for NVIDIA to come up with an LGA-1366 chipset. Users of the older LGA-775 however still have the option of buying nForce 700 series based motherboards to use SLI, and hence no real attempt was made to ensure the technology is available to Intel chipsets. Until now.

A member of Expreview's Chinese enthusiast community has successfully enabled SLI on an Intel chipset based LGA-775 motherboard, the first feat of its kind since much older attempts on i975 platforms using much older GeForce hardware. Firewings_[CCG] successfully ran SLI of GeForce 8600 GT and GeForce GTX 260 cards on his Intel X38+ICH9R chipset based ASUS Maximus Formula motherboard. The feat is headed by software he modified, details of which will surface soon. The mod was validated by Expreview staff, who used the software to run GeForce GTX 260 SLI on a more recent P45+ICH10R based Maximus II Formula motherboard. "By installing the software that Firewings [CCG] provides us, we managed to enable SLI configuration in Directory Services Restore Mode. Due to some 'small problems' according to Firewings [CCG], the SLI configuration can't be realized in normal mode for now, but he says this will be fixed soon," commented Expreview. With SLI enabled, the multi-GPU setup was able offer performance scaling that proves the mod works.

Armed with a creative engineering department, Zotac is in the news for innovating a new graphics card PCB design every now and then. For NVIDIA's GeForce GTX 200 series alone, the firm developed several performance and value oriented PCB designs. It decided to up the ante with a new single-GPU PCB design that is compatible with all GTX 200 series GPUs from NVIDIA, and is out with a GeForce GTX 260-based model out for the Chinese market. The new PCB focuses on strong power circuitry that allow high stability for overclocking. The 12-layered PCB is jam-packed with high-grade power circuitry, which ensures the GeForce GTX 260 GPU runs at 800 MHz out of the box, with even more overclocking headroom to spare.

To begin with, the vGPU portion consists of a lavish 12-phase digital PWM circuit with Renesas LFPAK MOSFETs, BL high-capacity PWM chokes, and Panasonic SP-CAP, flatbed capacitors and a dedicated VRM controller. The vMem portion has a 3-phase power circuit that uses the same kinds of components as the vGPU, and has its own VRM controller. The PCB draws auxiliary power from PCI-E 8 + 6 pin connectors. It holds place for two more memory chips indicating that the same PCB could be used for a GeForce GTX 285 based SKU in the future. On the outputs front, a DVI-D, HDMI, and D-Sub make for the card's connectivity. The new Zotac GeForce GTX 260 model will be cooled by an Arctic Cooling Accelero Xtreme GTX cooler. It is made exclusive for the Chinese market, with a price of RMB 1399 (US $205). Technical data from its electical-testing can be viewed here.

It looks like MSI's beating the drums about military-grade superiority of its N260GTX Lightning over its competitors has finally come to bear some fruit. Renowned enthusiast hipro5 from Greece set a new GPU core clock-speed record for the NVIDIA GeForce GTX 260, with a stress-stable 1100 MHz, with its shader domain clocked at 2205 MHz. Aided with an Intel Core i7 975 XE clocked at 5.57 GHz (HTT disabled, 4 GB DDR3 memory clocked at 2064 MHz DDR), the bench stood Aquamark with scores of 392,063 points (GPU score: 71,955, CPU: 40,070). The score validation can be found here The memory of the graphics card remained at its stock frequency of 999 MHz. The same bench at slightly lower speeds of 1070/2205/999 MHz (core/shader/memory), ran 3DMark03 stable, scoring 73461 points. The stock cooler of the N260GTX however, was excluded from this feat. The HWBot record reveals that the graphics card was cooled using a liquid-nitrogen (LN2) cooler.

Popular memory manufacturer Mushkin, which is also known for power supply units and until last year, for video cards, plans to resume being a video card manufacturer, and an add-in card partner for NVIDIA. The company just announced plans on an entire lineup of products based on the NVIDIA GeForce GTX 200 series, called Mushkin UltimateFX. It includes products starting from the GeForce GTX 260 896 MB base model, going all the way up to GeForce GTX 285 2048 MB, with models carved out based on clock speeds and memory sizes.

For the most part, it looks like Mushkin will use non-reference designs which we may be familiar with. One of their products resembles a similar design by Sparkle, while another bears resemblance with the reference NVIDIA design, albeit a green PCB (something PNY has opted for). The cards will be shipped in an a classic wooden jewel case, similar to to the "WoodBox" it shipped with its older products. Here's what the lineup looks like:

Having tried its hand in several PCB designs for the NVIDIA GeForce GTX 200 series, Zotac created itself enough room to come up with new SKUs at will. The company is using a recently-designed 10-layer PCB it used for a high-end GeForce GTX 285 accelerator featuring Arctic Cooling Accelero Extreme cooler, to design a new GeForce GTX 260 model (GTX260-896D3) with essentially the same design, except for a change: While the GTX 285 card featured 0.77 ns GDDR3 memory chips made by Hynix, this one uses 1.0 ns chips by Samsung.

The card retains the 6+3 phase power circuitry, the Accelero Extreme cooler, and standard features for the GeForce GTX 260: 216 shader processors, 896 MB of GDDR3 memory across a 448-bit interface, support for 3-way SLI, CUDA and PhysX. It will use reference NVIDIA clock speeds of 574/999 MHz (core/memory). Backed by a 3-year warranty, this card will hit stores in China at a price of 1299 RMB (US $190).

With NVIDIA lowering the prices of the GeForce GTX 260 almost at will, its partners are pushed to innovate their own designs that lets them make the most margins out of the tight market segment. At $170~$200, partners have come up with a newer breed of GeForce GTX 260 (55 nm, 216 SP) accelerators with some very interesting designs. MSI on its part has come up with a certain design that looks very familiar for the fact that it looks to be inspired to a large extent by the ATI Radeon HD 4870/4890. The MSI N260GTX-T2D896-OCv4 uses a red PCB, and a cooler design that takes inspiration from that of the reference Radeon HD 4870 to the extent that it looks pleasantly amusing. The heatsink over the card's VRM area coarsely resembles a shorter one on the reference-design Radeon HD 4850. Aesthetics apart, this card is factory overclocked, with speeds of 655/2100 MHz (core/memory). It packs 896 MB of GDDR3 memory. American retailer Newegg.com has it listed for $174.99, which further goes down to $154.99 with the optional rebate.Images Courtesy: Newegg.com

Sparkle on Friday made official its latest graphics card - the Calibre X260/X260 HM. The card uses 55nm GeForce GTX 260 GPU for base model, but adds bumped clock speeds, custom black PCB with enhanced GPU and memory power circuitry, solid capacitors and a custom dual-slot cooler. A 448-bit memory interface and 896 MB of GDDR3 memory are used to finish the whole picture. Sparkle bundles the card with its own SPA Tune, a software that controls the card's clock speeds. It provides three modes - overclocking mode (648 MHz/ 1998 MHz), standard mode (576 MHz/ 1998 MHz), and green mode (400 MHz/ 600 MHz). You can switch between the three modes whenever you want. Click here if you are more interested in the card.

MSI releases two GeForce GTX 260 video cards today - MSI N260GTX Lightning, and N260GTX Lightning Black Edition. First introduced at Cebit 2009, MSI N260GTX Lightning has twice the standard GeForce GTX 260 GDDR3 memory - 1792 MB, utilizes a Twin Frozr cooling with heatpipes, dual slot heatsink and two big fans. The card is reported to utilize 10 Phase PWM design, where 8 phases are reserved for the GPU and 2 phases for the memory.
There are two versions of the card, that are essentially the same, but differ only from the bundle package. The better version N260GTX Lightning Black Edition is paired with the company's AirForce Panel, an external touch panel that can change the card's voltages and clock speeds on the fly. It is not included in the Lightning edition. Detailed information on the clock speeds of both cards is below.

ASUS, the world's leading producer of top-quality graphics solutions, today launched the ASUS Republic of Gamers (ROG) ENGTX260 MATRIX/HTDI/896MD3 graphics card. This innovative piece of hardware comes equipped with the upgraded dual fansink Hybrid Cooler+ for automatic and independent fan speed control. Dubbed as the world's most intelligent cards, the MATRIX Series are also equipped with ROG-exclusive technologies and features that unleash the full power of graphics cards, allowing gamers to enjoy unrivaled in-game visuals. The graphics cards are able to achieve this through several unique features, including:

• Twin fan control application that provides independent adjustments to the Hybrid Cooler+ fans
• Integrated hardware and software for total graphics card control
• Customizable functions that provide a new level of control to gamers
• Accurate adjustments of GPU and memory voltages
• Total monitoring of GPU/memory/Power IC/ambient temperatures
• Total monitoring of GPU/memory/board power consumption
• Automatic fan speed control in relation to the world's only advanced GPU loading detection
• Energy efficiency that surpasses generic boards

On January the 8th, NVIDIA reclaimed the title of the manufacturer of the world's fastest consumer graphics accelerator, with the launch of the GeForce GTX 295. It was received adequately well by the press, for two factors: its performance, and its competitive pricing. NVIDIA, for the first time in generations of graphics technologies, chose to price its high-end offering competitively, and hasn't used its industry-leading position to warrant a high price. Unfortunately, the company seems to be in no mood to hand over the benefit the consumers, not by stepping up its prices, but by not keeping up to its demand.

NVIDIA, which believes in the concept that selling a GeForce GTX 260 for a price well within US $200, at the expense of profits, relies on sales volumes to do the job of making up for its losses, and/or reduced margins. To an extent, the company's moves to redesign the reference PCB for the GTX 260, and subsequently letting partners have a common design kit to come up with their own PCB designs, seems to have helped the cause. That isn't the case with GTX 295. Its power design and the requirement of two sets of high-grade components seems to have hit the company's plans of letting it sell for its competitive price, which it would rather channelize in subsidizing the GTX 260.

As NVIDIA partners attempt to cash in on the good sales of GeForce GTX 260, by lining up numerous custom-designed models based on the said GPU, Galaxy has added yet another to its lineup: a GeForce GTX 260 accelerator with twice the amount of memory as its reference design, all 1792 MB of it. Galaxy worked on two areas of this card: cooling and memory. Towards memory, the company chose to use the high-density Samsung K4J10324QD-HJ1A memory chip, that has a capacity of 1 Gb (128 MB). 14 such chips populate the 448-bit GDDR3 memory bus to dole out a total 1792 MB of memory capacity.

Towards cooling, Galaxy used a custom GPU cooler that spans across the length of the card. Its GPU contact block gives out five heatpipes that direct heat to a dense array of alumium fins. Three fans are employed to cool the card. The airflow passively cools the memory, the 4+1 phase VRM area has a heatsink of its own, so does the NVIO2 processor. Expreview put the card through a series of test to bring out its performance increment over the 896 MB GeForce GTX 260. The games World in Conflict and Company of Heroes saw significant increments in average frame-rates.

Information from an industry source was misinterpreted by VR-Zone yesterday, which wrote in its original report that the ATI Radeon HD 4870 1 GB was going to get a price-cut from $199 down to $179. In reality, it is the GeForce GTX 260 that is expected to get the price cut, sending its price down to $179.

NVIDIA had implemented significant reductions in prices of GeForce GTX 260 accelerators as recently as today, setting its price to $199. AMD may react to this move with its own new pricing scales for the competitive Radeon HD 4870 accelerator. There is no word on when the new pricing would be implemented.

AMD is on the course of releasing a new line of high performance products based on its upcoming RV790 graphics processor. The high-end single GPU SKU, Radeon HD 4890 is expected to be competitive with NVIDIA's GeForce GTX 260 (216 SP). NVIDIA on the other end of the field, is planning a retaliation to the RV790. The GPU giant is carving out a new SKU based on the 55 nm G200b: GeForce GTX 275.

The new SKU will be placed in the US $225~275 range. While the specifications are not known at this point in time, there are two theories: G200b with 216 stream processors and a 512-bit memory interface, and the other theory suggesting 240 stream processors with the existing 448-bit memory interface on GTX 260. The latter looks inexpensive as the former would step up manufacturing costs due to the addition of two memory chips.

Our friends from Guru of 3D alarm that NVIDIA might cut the price of GeForce GTX 260 by up to $30 from its current price tag. With the new price in position the GeForce GTX 260 should drop to around $199 and thus enter in the same league as ATI's Radeon HD 4870 with 1 GB memory. The price cut is direct result of ATI's recent price reductions that soften the price of both Radeon HD 4870 512 MB and Radeon HD 4850 512 MB.

AMD is attempting to revive its competitiveness that took a beating with NVIDIA's introduction of 55 nm G200b-based graphics accelerators. The method AMD seems to be adopting is by giving its existing flagship GPU, the RV770, a series of design improvements that facilitate higher clock-speeds, in turn, better performance on offer.

A lot has been said about RV790 till date, with each commentator coming up with a new version of the story. It has been more or less established that the RV790 will be a improvement over the RV770, though not a revolutionary one. Fresh information gathered by PC Games Hardware places a realistic estimate on up to where RV790 is going to push the performance envelope for AMD.

But wait, that's not all from MSI for today. The company also made public the release of the N260GTX Lightning video card. This NVIDIA GeForce GTX 260 based card features 1792 MB of DDR3 memory, dual fan design and an exclusive "AirForce Panel" with touch buttons for instant overclocking and voltage regulation.

Inno3D are excited to launch the Inno3D GeForce GTX 260 with FreezerX2 cooler that offers double the cooling performance and better value for your buck.

FreezerX2 Features

• Dual 8 cm ball bearing PWM fan, cooling power x 2
• DHT Technology, cooling efficiency x 2
• 2 mm fin distance, airflow x 2
• Smooth sanded base, contact area x 2
• Triple heatpipe cooling performance
• 55 nm Manufacturing Process
• Less power consumption with 175 W

Relatively unknown to the west, yet popular in the Asian OEM and retail markets, Chinese component cooler designer Dongwon has innovated a modular full-coverage design for its water-block which boosts its compatibility. The block serves as a simple answer to NVIDIA's almost ever-changing reference PCB design for its single-GPU GeForce GTX 200 series accelerators. The design allows this water-block to fit the P651 (65 nm GeForce GTX 260 and GTX 280), P654 (first versions of 55 nm GeForce GTX 260 and GTX 285), and the newer P897 PCB design for the 55 nm cards.

The full-coverage block consists of two segments: the main unit which provides coverage for the GPU, the memory and the NVIO2 processor; and the second segment that covers the VRM area. The segment's grooves are designed to cover the LFPAK VRM arrangement on P651 and P654, while it can be detached to allow the rest of the block to be used on the P897. We're not entirely sure if the design concept circled around compatibility with different PCB designs, though it ends up being the block's most useful application. Although the company has seemingly no presence in the western retail channels, one can expect this block to come factory-fitted on graphics cards.

MSI is said to be readying a custom graphics card called the N260GTX Lightning, based on NVIDIA's GeForce GTX 260 GPU. There is currently no word on clock speeds but one would expect this card to come factory overclocked. The graphics card has a custom red PCB, featuring 8+2 phase power, 5000 hour capacitors, 1792 MB of GDDR3, D-Sub, DVI and DisplayPort or HDMI and a dual slot, dual fan cooler, featuring 5 heatpipes. MSI will also be bundling an AirForce panel which appears to allow on the fly adjustments to core and memory voltage, core, shader and memory clock speeds, as well as profile selection and brightness and contrast adjustments. The MSI N260GTX Lightning is expected to make its first official appearance at CeBIT 2009.

Inno3D today announced the launch of their i-Chill GTX 260 with Accelero XXX, a factory overclocked NVIDIA GeForce GTX 260, featuring Arctic Cooling's Accelero XTREME GTX 280. Dubbed by Inno3D as the Accelero XXX for, Xtreme Performance (30°C lower), Xtreme Architecture (3 x 90 mm fans + 5 heatpipes over 110 fins) and Xtreme Silence (PWM support with <30dB noise). The card clocks in at 620 MHz on the core, and 2100 Mhz on the memory. Inno3D have stated that this will be a limited edition card.