NVIDIA's upcoming GeForce GTX 760 could be the company's last GTX 700 series retail desktop SKU, at least for this year. A leaked slide by the company lays out how this year's three new GTX 700 series SKUs pretty much seal the product stack. It reveals that GeForce GTX TITAN will remain NVIDIA's flagship graphics card throughout the year. The thousand-dollar single-GPU card is based on the GK110 silicon, with 2,688 CUDA cores, and 6 GB of memory. The GeForce GTX 780, introduced this March, replaces the GeForce GTX 680 on the product-stack, even at its much higher launch price of $650, compared to its predecessor's $500. The GTX 780 has no competition from AMD at its price-point.

The slide also reveals that the GeForce GTX 770, which was launched late last month, will replace the GeForce GTX 670 from the stack. Based on the GK104 silicon, it features 1,536 CUDA cores, and 2 to 4 GB of memory. Given that it has a lot in common with the GeForce GTX 680, albeit with higher clock speeds, GPU Boost 2.0, its $400 pricing surprised us. The GeForce GTX 770 outperforms AMD's HD 7970 GHz Edition, and is generally priced on-par. The only definitively faster AMD card is the $1100 HD 7990 "Malta," which makes the GTX 770 the king of its segment.

ASUS today announced two GeForce GTX 780 DirectCU II graphics cards, the new highest-end offerings in its NVIDIA-based graphics card range. They feature an ASUS-designed PCB and a completely new look for the DirectCU II cooler, as well as DIGI+ VRM digital power delivery and robust Super Alloy Power components.

GeForce GTX 780 DirectCU II cards use an NVIDIA GK110-300 GPU with 2304 CUDA cores. They ship with 3GB GDDR5 video memory on a 384-bit interface. For avid PC gamers who demand ultra-high detail and readiness for resolutions beyond full HD 1080p, GeForce GTX 780 DirectCU II cards deliver extreme cooling, stability, and endurance well above reference card requirements.

Galaxy started its GeForce GTX 780 lineup off with a single reference-design card, but plans to make some major additions. One of them is the GTX 780 Twin Fan. Pictured below, the card uses a unique fin-stack heatsink that uses a mix of aluminum and copper fins. Four 8 mm-thick nickel-plated copper heat pipes conduct heat from the GK110 GPU to this colorful fin-stack, which is ventilated by a pair of 80 mm spinners. Alas, this unique dual-slot cooler sits on a bland NVIDIA-reference design PCB, just like on every GTX 780 card ever sold. Galaxy should ship the card with factory-overclocked speeds, but didn't disclose them.

MSI rolled out its GeForce GTX 780 graphics, which we're sure won't be its last, given NVIDIA's laxed policy on custom-designs. MSI's card sticks to NVIDIA reference design, featuring reference-design clock speeds of 863 MHz core, 900 MHz GPU Boost, and 6.00 GHz memory. It tucks in 3 GB of GDDR5 memory. Based on the 28 nm GK110 silicon, GeForce GTX 780 features 2,304 CUDA cores, 192 TMUs, 48 ROPs, and a 384-bit wide memory interface. The card draws power from a combination of 6-pin and 8-pin power connectors. Display outputs include two DVI, and one each of HDMI and DisplayPort. Expect the card to be priced at $649.99.

NVIDIA announced the GeForce GTX 780, the first addition to its desktop GeForce GTX 700 series. Pictured below, the card is as close as it gets to owning a GeForce GTX TITAN, at two-thirds the price. It's based on the same 28 nm GK110 chip, but configured with 2,304 CUDA cores and 192 TMUs, compared to the 2,688 cores and 224 TMUs of the GTX TITAN. The card retains the 384-bit wide GDDR5 memory interface, but holds 3 GB of memory. In that sense, all its key faculties are an exact 50 percent increase over is predecessor, the GeForce GTX 680.

NVIDIA completely recycled the design of GeForce GTX TITAN, down to the PCB and cooler. The only way you can make the two cards out is the lack of memory chips on the GTX 780's back. The card features GPU Boost 2.0, a technology that debuted with the GTX TITAN, boosts clocks and voltage, taking temperatures into account. The GPU core is clocked at 863 MHz, GPU Boost at 900 MHz, and memory at 6.00 GHz (GDDR5-effective). NVIDIA is allowing its partners to launch products with custom design air cooling solutions. Prices should start at $649.99.Read our reviews:

• NVIDIA GeForce GTX 780 (reference)
• GIGABYTE GeForce GTX 780 WindForce OC
• GeForce GTX 780 SLI

It turns out that NVIDIA's upcoming high-end graphics card, the GeForce GTX 780, is indeed based on the GK110 chip, and to the extant of featuring the same reference-design PCB and cooling solution as the GeForce GTX TITAN. Chinese portal IT168 posted press-shots of the card. If you overlook the "GTX 780" embossing on the cooler, half the number of memory chips, and another very subtle difference, you will easily mistake the GTX 780 for a GTX TITAN.

Pictures reveal the card to feature an ASIC bearing the number "GK110-300-A1," which is rumored to feature 2,304 CUDA cores, 192 TMUs, 48 ROPs, and a 384-bit wide memory interface. Given that the card features just twelve 2 Gbit memory chips, the total memory amount should be 3 GB. The card is said to feature clock speeds of 863 MHz core, 902 MHz GPU Boost, and 6.00 GHz memory, which belts out 288 GB/s memory bandwidth. It draws power from a combination of 6-pin and 8-pin PCIe power connectors. The 4-way SLI-capable card gives out display from a pair of dual-link DVI, HDMI, and DisplayPort connectors.

NVIDIA's next-generation GPU family is not far away. We're hearing that new product-launches from the GPU giant could be just weeks away, and within this month. It's only natural that some of these cards could pass through leaky pipes, much to our benefit. One such source in China posted pictures of what he claims to be NVIDIA-reference design GeForce GTX 780 and GeForce GTX 770. Both cards feature a design not unlike the $1000 GeForce GTX TITAN. In fact they look identical. It's not the questionable embossing on the cooler shrouds that caught our attention, it's the subtle differences near the PCI-Express interface - location of PCB number, arrangement of termination resistors, etc., that did, and so we're rating this leak highly plausible.

We know from a previous report that GeForce GTX 780 will be positioned a notch below the GeForce GTX TITAN, in NVIDIA's product stack. It could be based on the same GK110 silicon, and could feature 2,496 CUDA cores, and a 320-bit wide GDDR5 memory interface, holding 5 GB of memory. It won't surprise us if NVIDIA completely recycles the GTX TITAN PCB, as it doesn't particularly have an over-the-top selection of components, apart from the GPU. The GeForce GTX 770 is a different beast altogether. It is based on a GPU not unlike the GK104, with 1,536 CUDA cores, and a 256-bit wide GDDR5 memory interface, holding 4 GB of memory. To sweeten the prospect of upgrading to these new cards, NVIDIA is dropping in the same sexy magnesium alloy-based cooling solution it used on $1000 cards such as the GTX TITAN and GTX 690.

Late last month, we learned that NVIDIA plans to unveil its GeForce GTX 7-series desktop GPU family just a little later this month. According to a new report by SweClockers, the company plans to take full advantage of AMD's lethargy or console-fixation, in launching its next GPU generation much later this year. The premium GeForce GTX 780, which is reportedly based on the GK110 silicon, could command a price much higher than the $499.99 GeForce GTX 680 started out on, when it launched last March.

Pricing of the GeForce GTX 780 could be closer to that of the GeForce GTX TITAN, than today's GTX 680, according to the report. It asks us not to be surprised if the card is priced on-par with the TITAN, making us wonder if TITAN remains NVIDIA's fastest single-GPU graphics card for long, or if NVIDIA is re-branding TITAN to GTX 780, or even if it ends up being the fabled "TITAN Ultra."

AMD may have declared that its next-generation GPU family won't arrive before October, but that isn't stopping NVIDIA from launching its GeForce GTX 700 series much earlier. While AMD's lineup is banking on sales during the X'mas shopping season, NVIDIA is going after the pre-Summer system upgrade crowd. According to a Bright Side of News (BSN) report, NVIDIA's new lineup will make its debut no later than this May.

According to the BSN report, GeForce GTX 700 series will be heavily based on existing GeForce Kepler silicon, with a handful feature-set updates, and some clever product stack adjustments. The part that succeeds today's GeForce GTX 680, the GeForce GTX 780, could be based on the 28 nm GK110 silicon, and could very well be the fabled "GTX TITAN LE" part that's been in the news for some time now, as being a scaled down GeForce GTX TITAN, with 2496 CUDA cores, 208 TMUs, 40 ROPs, and a 320-bit wide GDDR5 memory interface, holding 5 GB of memory.

NVIDIA may decisively hold on to the single-GPU performance lead, with its GeForce GTX Titan graphics card, but at roughly $1000, it could attract a very small market. According to a SweClockers report, NVIDIA is looking to woo gamers just ahead of Summer with the second GK110-based GeForce GTX graphics card. Similar in specifications to the fabled Quadro K6000, the new SKU could feature 13 out of 15 streaming multiprocessors on the GK110 silicon, working out to 2,496 CUDA cores, 208 texture memory units, a 320-bit wide GDDR5 memory interface holding 5 GB of memory, and 40 ROPs. Given that there's a deep ravine between the ~$450 GeForce GTX 680 and ~$1000 GTX Titan, NVIDIA could pick a price-point in the middle. The report claims the new SKU could launch some time between July and August, 2013.

Inno3D rolled out its newest flagship graphics card, the GeForce GTX Titan iChill Black Series. The card ships with Arctic's Accelero Hybrid cooling solution pre-fitted to an NVIDIA reference design GTX Titan PCB. The cooler uses a combination of a closed-loop liquid block to cool the GPU, and fan-heatsink to cool other hot components on the card, such as memory chips and VRM.

The GTX Titan iChill Black series ships with factory-overclocked speeds of 937 MHz core, and 980 MHz GPU Boost, against NVIDIA reference speeds of 837 MHz core and 876 MHz GPU Boost. Memory clock speed stays at 6.00 GHz. Based on the 28 nm GK110 silicon, NVIDIA's GeForce GTX Titan features 2,688 CUDA cores, 224 TMUs, 48 ROPs, and a 384-bit wide GDDR5 memory interface, holding 6 GB of memory. The card draws power from a combination of 6-pin and 8-pin PCIe power connectors, display outputs include a pair of dual-link DVI, HDMI, and DisplayPort. The company did not release pricing or availability information.

Following the launch of its consumer-oriented GeForce GTX Titan, NVIDIA is planning its second graphics card based on the 28 nm GK110 silicon, this one for professionals, and featuring in the Quadro family. Likely bearing the model name Quadro K6000, the card is expected to feature 13 of the GK110's 15 streaming multiprocessors (SMXs), which work out to 2,496 CUDA cores, 208 TMUs, 40 ROPs, and a 320-bit wide GDDR5 memory interface, likely holding 5 GB of memory. The configuration could be just about enough for 1 TFLOP/s double-precision floating-point performance, and could hence offer an upgrade over Fermi-based Quadro 6000.

MSI announced its GeForce GTX Titan graphics card, adding to the list of partners selling the same exact wine in a different bottle. MSI's package includes Live Update and Afterburner overclocking software. The GTX Titan is based on NVIDIA's 28 nm GK110 silicon, featuring 2,688 CUDA cores, 224 TMUs, 48 ROPs, and 6 GB of GDDR5 memory across a 384-bit wide interface. It features NVIDIA-reference clock speeds of 837/876/6008 MHz (core/GPU Boost/memory), draws power from a combination of 6-pin and 8-pin PCIe power connectors, and its display outputs include two dual-link DVI, HDMI, and DisplayPort. MSI's GTX Titan is priced at $1000 before taxes.

Palit's twin brand Gainward announced its GeForce GTX Titan, too. The card, like every other GTX Titan launched today (and will be), sticks to NVIDIA's reference board design, and reference clock speeds of 837/876/6008 MHz (core/GPU Boost/memory). Based on the 28 nm GK110 silicon, the GeForce GTX Titan features 2,688 CUDA cores, 224 TMUs, 48 ROPs, and a 384-bit wide GDDR5 memory interface, holding 6 GB of memory; and like every other GTX Titan, it is priced at $1000 before taxes.

Gigabyte announced its GeForce GTX Titan graphics card. Much like every other GTX Titan announced till now, it sticks to NVIDIA reference design. To spice things up, Gigabyte is tossing in a few extras from its Aivia line of gaming peripherals, by including a gaming mousepad, and a pack of Aivia-themed playing cards that you can also play "memory" with. Based on the 28 nm GK110 silicon, the GeForce GTX Titan features 2,688 CUDA cores, 224 TMUs, 48 ROPs, and a 384-bit wide GDDR5 memory interface, holding 6 GB of memory. The card features NVIDIA reference clock speeds of 837/876/6008 MHz (core/GPU Boost/memory). It should be priced at $1000 before taxes, $1100~$1200 after it.

Palit joined other NVIDIA partners in making a late-evening announcement of its GeForce GTX Titan graphics card, even though we're informed that partners have zero GTX Titan inventories, and could begin having them only by early-March. Palit's GTX Titan is no different from any of the other SKUs launched today. NVIDIA restricts partners from modifying the board design, much like it did with the GeForce GTX 690. Based on the 28 nm GK110 silicon, the GeForce GTX Titan features 2,688 CUDA cores, 224 TMUs, 48 ROPs, and a 384-bit wide GDDR5 memory interface, holding 6 GB of memory. It is priced at US $1000, which could go anywhere up to $1100 and $1200 including taxes.

You may have read our report from earlier today, covering the main points that AMD was trying to make in its recent teleconference with the European press (which includes us). While in the call, we were a little jolted by the choices of words some of AMD's executives used to describe their company's consumer graphics outlook for 2013, how they believe they can hold out for almost the entire year with little or no major updates to their product stack, and more interestingly, a few above-the-belt jabs at NVIDIA and its upcoming GeForce GTX Titan product.

The crux of AMD's emergency meeting with the press was to bust some misconceptions spread in the press over the last couple of weeks, to tell them a Graham's Number of times that they still hold the fastest single GPU on the planet, which powers the fastest graphics card there is (ASUS ARES II). The most ironic part of AMD's emergency meeting with the press was the one in which they called GeForce Titan NVIDIA's emergency/knee-jerk reaction to AMD's getting cozy with game developers, and netting some of the biggest PC game launches of the season for its Never Settle Reloaded bundle.

Here it is, folks; the first pictures of NVIDIA's newest pixel crunching dreadnought, the GeForce GTX Titan. Pictures leaked by various sources east of the Greenwich Median reveal a reference board design that's similar in many ways to that of the GeForce GTX 690, thanks to the magnesium alloy cooler shroud, a clear acrylic window letting you peep into the aluminum fin stack, and a large lateral blower. The card features a glowy "GeForce GTX" logo much like the GTX 690, draws power from a combination of 6-pin and 8-pin PCIe power connectors, and features two SLI bridge fingers letting you pair four of them to run 3DMark Fire Strike as if it were a console port from last decade.

When we first heard of NVIDIA launching its GK110-based consumer graphics card by as early as February, it took us by surprise. Intimidating naming (GeForce Titan 780?) aside, the graphics card is hoping to better NVIDIA's current-generation flagship, the dual-GPU GeForce GTX 690, in a single-GPU package, but does the graphics card market really need NVIDIA to launch its card at the moment? Perhaps not, but the answer lies not with AMD and competition in the graphics card market, but Sony, and competition between PC and console platforms.

Over the weekend, it surfaced that Sony would introduce its next-generation PlayStation console (codenamed "Orbis") later this month, and it would mark the beginning of the next-generation of game consoles. PlayStation 4 features an updated hardware feature-set, and promises to raise the bar with graphics detail that the console industry held with an iron fist for the past half decade. This presents a challenge for not only NVIDIA, but PC gaming in general. Here's how.

2013 started off on a rather dull note for the PC graphics industry. NVIDIA launched its game console platform "Project: Shield," while AMD rebranded its eons-old GPUs to Radeon HD 8000M series. Apparently it could all change in late-February, with the arrival of a new high-end single-GPU graphics card based on NVIDIA's GK110 silicon, the same big chip that goes into making the company's Tesla K20 compute accelerator.

NVIDIA may have drawn some flack for extending its "GTX" brand extension too far into the mainstream and entry-level segment, and wants its GK110-based card to stand out. It is reported that NVIDIA will carve out a new brand extension, the GeForce Titan. Incidentally, the current fastest supercomputer in the world bears that name (Cray Titan, located at Oak Ridge National Laboratory). The GK110 silicon physically packs 15 SMX units, totaling 2,880 CUDA cores. The chip features a 384-bit wide GDDR5 memory interface.

EK Water Blocks is working on a new Full Cover water block for the latest NVIDIA Quadro 5000K graphics- and Tesla K20 computing card thus providing world's first water cooling solution for NVIDIA Maximus technology.

The upcoming EK-FCQ5000K will be designed specifically for the latest GK100 based Nvidia Quadro 5000K professional graphics card while the upcoming EK-FCTK20 (GeForce GK110 based) will be designed specifically for Nvidia Tesla K20 computing card.

A 3DCenter.org report shed light on what NVIDIA's GPU lineup for 2013 could look like. According to the report, NVIDIA's next-generation GPUs could follow a similar path to previous-generation "Fermi Refresh" (GF11x), which turned the performance-per-Watt equation around back in favor of NVIDIA, even though the company's current GeForce Kepler has an established energy-efficiency lead. The "Kepler Refresh" family of GPUs (GK11x), according to the report, could see significant increases in cost-performance, with a bit of clever re-shuffling of the GPU lineup.

NVIDIA's GK104 GPU exceeded performance expectations, which allowed it to drive this generation's flagship single-GPU graphics card for NVIDIA, the GTX 680, giving the company time to perfect the most upscaled chip of this generation, and for its foundry partners to refine its 28 nm manufacturing process. When it's time for Kepler Refresh to go to office, TSMC will have refined its process enough for mass-production of GK110, a 7.1 billion transistor chip on which NVIDIA's low-volume Tesla K20 GPU compute accelerator is currently based.

Specifications of NVIDIA's Tesla K20 GPU compute processor, which was launched way back in May, are finally disclosed. We've known since then that the K20 is based on NVIDIA's large GK110 GPU, a chip never used to power a GeForce graphics card, yet. Apparently, NVIDIA is leaving some room on the silicon that allows it to harvest it better. According to a specifications sheet compiled by Heise.de, Tesla K20 will feature 13 SMX units, compared to the 15 available on the GK110 silicon.

With 13 streaming multiprocessor (SMX) units, the K20 will be configured with 2,496 CUDA cores (as opposed to 2,880 physically present on the chip). The core will be clocked at 705 MHz, yielding single-precision floating point performance of 3.52 TFLOP/s, and double-precision floating point performance of 1.17 TFLOP/s. The card packs 5 GB of GDDR5 memory, with memory bandwidth of 200 GB/s. Dynamic parallelism, Hyper-Q, GPUDirect with RDMA are part of the new feature-set. The TDP of the GPU is rated at 225W, and understandably, it uses a combination of 6-pin and 8-pin PCI-Express power connectors. Built in the 28 nm process, the GK110 packs a whopping 7.1 billion transistors.

With its competition checked thanks to good performance by its GK104 silicon, NVIDIA was bold enough to release die-shots of its GK110 silicon, which made its market entry as the Tesla K20 GPU-compute accelerator. This opened flood-gates of speculation surrounding minute details of the new chip, from various sources. We found one of these most plausible, by Beyond3D community member "fellix". The source of the image appears to have charted out component layout of the chip by some pattern recognition and educated guesswork.

It identifies the the 7.1 billion transistor GK110 silicon to have 15 streaming multiprocessors (SMX). A little earlier this week, sources close to NVIDIA confirmed the SMX count to TechPowerUp. NVIDIA revealed that the chip will retain the SMX design of GK104, in which each of these holds 192 CUDA cores. Going by that, GK110 has a total of 2880 cores. Blocks of SMX units surround a centrally-located command processor, along with six setup pipelines, and a portion holding the ROPs and memory controllers. There are a total of six GDDR5 PHYs, which could amount to a 384-bit wide memory interface. The chip talks to the rest of the system over PCI-Express 3.0.

NVIDIA today unveiled a new family of Tesla GPUs based on the revolutionary NVIDIA Kepler GPU computing architecture, which makes GPU-accelerated computing easier and more accessible for a broader range of high performance computing (HPC) scientific and technical applications.

The new NVIDIA Tesla K10 and K20 GPUs are computing accelerators built to handle the most complex HPC problems in the world. Designed with an intense focus on high performance and extreme power efficiency, Kepler is three times as efficient as its predecessor, the NVIDIA Fermi architecture, which itself established a new standard for parallel computing when introduced two years ago.