NVIDIA's upcoming flagship GPU based on its next-generation "Pascal" architecture, codenamed GP100, is shaping up to be a number-crunching monster. According to a leaked slide by an NVIDIA research fellow, the company is designing the chip to serve up double-precision floating-point (DPFP) performance as high as 4 TFLOP/s, a 3-fold increase from the 1.31 TFLOP/s offered by the Tesla K20, based on the "Kepler" GK110 silicon.

The same slide also reveals single-precision floating-point (SPFP) performance to be as high as 12 TFLOP/s, four times that of the GK110, and nearly double that of the GM200. The slide also appears to settle the speculation on whether GP100 will use stacked HBM2 memory, or GDDR5X. Given the 1 TB/s memory bandwidth mentioned on the slide, we're inclined to hand it to stacked HBM2.

Here are the first PCB shots of NVIDIA's next-gen flagship graphics card, the GeForce GTX TITAN-X. At the heart of this beast is the swanky new 28 nm GM200 silicon, which is rumored to feature 3072 CUDA cores based on the "Maxwell" architecture, a 384-bit wide GDDR5 memory interface, with NVIDIA's latest memory bandwidth management mojo, a staggering 96 ROPs, and 12 GB of memory. The design goal is probably 4K to 5K gaming with a single card, at reasonably high settings. The GM200 silicon appears slightly bigger than the GK110, NVIDIA's previous big chip.

The display I/O of this card looks identical to that of the GTX 980. We're not sure if the DVI connector will make it to the final design. There are no shots of the VRM, although given this architecture's track-record, we don't expect the TITAN-X to have any heavier power requirements than the GTX 780 Ti (6-pin + 8-pin power inputs). NVIDIA is expected to launch the GTX TITAN-X within this quarter. Don't hold off on your GTX 980 purchases just yet, because NVIDIA tends to overprice its "TITAN" branded graphics cards.

NVIDIA's "Maxwell" architecture may have got a rather low-key debut with the GeForce GTX 750 Ti, but nobody saw its performance-segment derivative, the GM204 silicon, driving the GeForce GTX 980 and the GTX 970. The new architecture makes its predecessor, the "Kepler" look inefficient in comparison. It looks like NVIDIA still thinks Kepler is competitive to competition from AMD (GCN) and Intel (Knights Corner), in the high-performance computing era.

The problems here are NVIDIA already launched a GK110 based Tesla HPC card, and its big "Maxwell" chip is nowhere in sight. The GM204 has limited memory bandwidth, and its texture-compression mojo can't bail out bandwidth-hogging HPC applications. The solution? Develop a new big silicon based on "Kepler." Enter, the GK210. That's right, the G-K-210. Launched today with the Tesla K80 dual-chip HPC accelerator, this chip could feature design improvements over the GK110, while offering memory bandwidth and sizes not possible on the GM204.

Ahead of its launch later this week, even more details of NVIDIA's upcoming GeForce GTX 980, and the 28 nm "GM204" silicon it's based on, tumbled out. To begin with, the GM204 silicon is confirmed to be built on the 28 nm silicon fab process. The chip bigger than that of the GK104, with a die area of 398 mm², yet smaller than the GK110, which measures 581 mm². Its transistor count is 5.2 billion, about 2 billion more than the GK104.

The component hierarchy of GM204 is similar to that of the GM107 silicon, on which the GTX 750 Ti is based. The GPU features a 256-bit wide GDDR5 memory interface, and PCI-Express 3.0 x16 bus. The GigaThread Engine dispatches workload between four graphics processing clusters (GPCs), the basic subunit. Each GPC has a common raster engine shared between four streaming multiprocessors Maxwell (SMMs), which each hold 128 CUDA cores. The total CUDA core count is hence 2,048. The L2 cache has been quadrupled over GK104. The chip features 2 MB of it, compared to 512 KB on its predecessor. The GM204 features 64 ROPs, double that of the GK104, and should hence come with a strong geometry processing muscle. The chip features a revolutionary new 3-bit delta color compression technology that makes the most of the limited memory bus width of this chip.

ASUS today took the veil off a new graphics card powered by NVIDIA's GK110 GPU, a custom GeForce GTX 780 that packs 6 GB of memory (double the amount found on 'regular' GTX 780s) and makes use of a DirectCU II cooling solution that can be totally silent, in certain situations. Named GeForce GTX 780 STRIX, the card in question will not be running its two fans when idle or during light workload scenarios (assuming the GPU temperature is below a threshold) but will power them up as soon as the GPU load increases.

The GeForce GTX 780 STRIX has 2304 CUDA Cores, a 384-bit memory interface, SLI support, dual-DVI, HDMI and DisplayPort outputs and will be available in two versions - one with stock clocks and one with an overclocked GPU. ASUS is promising more information 'soon'.

NVIDIA's GeForce GTX TITAN-Z missed the bus on its earlier 29th April, 2014 launch date, which was confirmed to the press by several retailers, forcing some AIC partners to content with paper-launches of cards bearing their brand. It turns out that the delay is going to be by just a little over a week. The GeForce GTX TITAN-Z is now expected to be available on the 8th of May, 2014. That will be when you'll be able to buy the US $3,000 graphics card off the shelf.

A dual-GPU graphics card based on a pair of 28 nm GK110 GPUs, the GTX TITAN-Z features a total of 5,760 CUDA cores (2,880 per GPU), 480 TMUs (240 per GPU), 96 ROPs (48 per GPU), and a total of 12 GB of GDDR5 memory, spread across two 384-bit wide memory interfaces. Although each of the two GPUs is configured identical to a GTX TITAN Black, it features lower clock speeds. The core is clocked at 705 MHz (889 MHz on the GTX TITAN Black), with GPU Boost frequencies of up to 876 MHz (up to 980 MHz on the GTX TITAN Black); while the memory remains at 7.00 GHz. The card draws power from a pair of 8-pin PCIe power connectors, and its maximum power draw is rated at 375W. It will be interesting to see how it stacks up against the Radeon R9 295X2 by AMD, which costs half as much, at $1,500.

Here it is, folks, the fabled monster dual-GPU graphics card from NVIDIA, based on its GK110 silicon, the GeForce GTX TITAN-Z (sounds like "Titans"). The first reference-design graphics card to span across three expansion slots, the GTX TITAN-Z features a cooler design that's an upscale of the GTX 690, with a pair of meaty heat-pipe fed heatsinks being ventilated by a centrally-located fan. The card features a pair of GK110 chips, with all 2,880 CUDA cores enabled, on each. That works out to a total core count of 5,760!

That's not all, the two chips have 480 TMUs, and 96 ROPs between them; and each of the two is wired to 6 GB of memory, totaling a stunning 12 GB on the card. At this point it's not clear if the GPUs feature full-DPFP, but their SPFP totals 8 TFLOP/s. Display outputs on the card include two dual-link DVI, a DisplayPort, and an HDMI. According to its makers, the GTX TITAN-Z is the first graphics card that's truly ready for 5K resolution (5120 x 2700 pixels) on a single display head, for gaming. At US $2,999, the card costs thrice as much as a GTX TITAN Black, for twice its performance.

Equipped with all new voltage management, including 3 rail PWMs with load balancing, the GTX 780Ti iChill DHS can deliver an insane core overclock of over 1300MHz. Backed up by 3GB of VRAM and a 384-bit interface means that even the latest DX11 gaming are delivered to your monitor in their full cinematic glory. Out performing the GTX Titan Black without breaking a sweat, the GTX 780Ti iChill DHS is a true powerhouse of a graphics card.

The DHS-Edition with its bombastic Triple-Fan-Cooler "Herculez X3 Ultra" is a true High-End-Device that not even the GTX Titan Black can beat. The abbreviation DHS (Double Hand Selected) in this case makes calear that already pre-selected Chips for the Graphics Card have been hand picket yet another time. There is no way of getting more quality and performance. Recommended only for Extreme Gamers! Exclusively available at Overclockers UK.

ZOTAC International, a global innovator and manufacturer of graphics cards, mainboards and mini-PCs, today launches new performance and flagship graphics cards with the GeForce GTX 750 family and new GTX Titan Black. The new ZOTAC GeForce GTX 750 family ushers in a new era of performance PC gaming with the next-generation NVIDIA Maxwell architecture while the new ZOTAC GeForce GTX Titan Black injects the venerable flagship series with greater compute power.

"We're embarking on a new generation of NVIDIA graphics architecture with the new ZOTAC GeForce GTX 750 family graphics cards," said Carsten Berger, senior director, ZOTAC International. "The new Maxwell architecture enables us to raise performance by 25-percent while cutting power consumption in half compared to the previous generation."

Here's the first alleged picture of NVIDIA's GeForce GTX TITAN Black, a high-end SKU NVIDIA is working on, to restore the competitiveness of the $999 price-point it commands. Although referred to as "TITAN Black," the card is nowhere close to looking like the CGI renders that surfaced last November. The board looks identical to the original GTX TITAN, except its "TITAN" etching on the cooler shroud is painted in black. The GTX TITAN Black maxes out the 28 nm GK110 silicon, featuring 2,880 CUDA cores, 240 TMUs, 48 ROPs, and a 384-bit wide GDDR5 memory interface, holding 6 GB of memory. It also features full DPFP for the GK110 silicon, which is exclusive to the GTX TITAN, within the GeForce range. NVIDIA is expected to give the GTX TITAN Black a low-key launch some time next week.

NVIDIA's upcoming high-end graphics card, the GeForce GTX TITAN Black, scored its first listing by an online store. Finnish store Multitronic Webshop listed it for €974.90 (including all applicable taxes). The store mentions the pre-tax price at €786.21, which converts to roughly US $1071. These prices confirm the theory that the GTX TITAN Black is being designed to replace the current GTX TITAN from its $999 (pre-tax) price-point. The current GTX TITAN has been cannibalized by the GTX 780 Ti at gaming, and the only two straws it's been clutching onto are 6 GB of memory, and full double-precision floating point GPGPU performance.

The GeForce GTX TITAN Black is a maxing out of the feature-set of the GK110 silicon. It features the full complement of CUDA cores at 2,880 - same as the GTX 780 Ti - but also features full double-precision floating point GPGPU performance, and 6 GB of standard memory amount, which could help in 3DVision Surround, and 4K Ultra HD display setups. Other vital specs of the GPU include 240 TMUs, 48 ROPs, and a 384-bit wide GDDR5 memory interface. There's no word on its clock speeds, but we expect it to be either on par with, or slightly higher than those of the GTX 780 Ti.

In a move by NVIDIA that could present enthusiasts with deep pockets more options in the high-end GPU segment, the company is planning not one, but two high-end graphics cards for this quarter (January to March). The company is planning to follow up its February launch of the GeForce GTX TITAN Black Edition with a dual-GPU card based on the GK110 silicon, presumably named "GeForce GTX 790." The card will be launched some time in March, and could help NVIDIA hold onto its own until it can make high-end GPUs on the teething next-generation 20 nm silicon fab process.

The GeForce GTX TITAN Black Edition replaces the GTX TITAN with a product that's not too different from the GTX 780 Ti in terms of core-configuration, but one that comes with 6 GB of memory, full double-precision floating point performance, and probably higher clock speeds. It offers the full complement of 2,880 CUDA cores, 240 TMUs, 48 ROPs, and a 384-bit wide GDDR5 memory interface. This card could displace the current GTX TITAN from its $999 price-point.

EVGA teased its GeForce GTX 780 Ti Classified K|ngp|n Edition graphics card some more, by posting high-resolution pictures. Designed to be the fastest single-GPU graphics card ever built, it features a non-reference PCB built from the ground up for voltage-assisted overclocking using exotic/subzero cooling solutions, although its included frosty-white cooling solution doesn't look like it will disappoint overclockers, either. There are no pictures of the business side of the PCB, but one can expect an extremely strong VRM powering the GK110 chip and 3 GB of memory. The card draws power from a combination of one 6-pin and two 8-pin PCIe connectors. One can expect EVGA regulars such as EVBot support, redundant BIOS, voltage measurement points, and clearly marked out spots for manual volt-mods, or using EVGA's EPower board. The card will also be designed with fewer but stronger screws, so it takes lesser time to rip off or attach a cooler.

MSI just teased a new VGA cooling solution it's working on, which will likely feature on its upcoming Radeon R9 290 series graphics cards. It's expected to be a long and tall variant of TwinFrozr IV designed for high-end chips like GK110 and "Hawaii," and will feature a frosty see-through acrylic shroud covering the aluminium fin-stack heatsink. TwinFrozr IV is MSI's workhorse VGA cooler for its Gaming, Lightning, and HAWK series of factory-overclocked graphics cards.

Gigabyte, which gave the GeForce GTX 780 the "GHz Edition" treatment, in which the GPU core is factory-overclocked above the 1 GHz mark, meted out the same with the more powerful GTX 780 Ti. Positioned above the GTX 780 Ti OC Edition which debuted early last month, the GHz Edition features a swanky new custom-design PCB by Gigabyte, which features a strong VRM; coupled with the same WindForce 450W cooling solution that's featured on the GTX 780 GHz Edition. The GTX 780 Ti OC Edition, on the other hand, is a combination of the WindForce 450W cooler and NVIDIA reference PCB, with a factory-overclock.

The PCB of Gigabyte GTX 780 Ti GHz Edition is a bustling metropolis, with a 10+2 phase VRM that draws power from a pair of 8-pin PCIe power connectors. The cooler, on the other hand, is a complex aluminium fin-stack heatsink to which heat is fed by five 8 mm-thick copper heat pipes, and which are ventilated by a trio of 100 mm fans.The large heatsink is supported by a base plate that draws heat from the memory and VRM; a support brace that runs along the top of the card, and a back-plate. Gigabyte tuned the card with clock speeds of 1085 MHz core (base), 1150 MHz core (GPU Boost), and a surprisingly untouched 7.00 GHz (GDDR5-effective) memory. Based on the 28 nm GK110 silicon, the GTX 780 Ti features 2,880 CUDA cores, 240 TMUs, 48 ROPs, and a 384-bit wide GDDR5 memory interface, holding 3 GB of memory. There's no word on when Gigabyte will formally launch this card. It shouldn't be too far away.

The GeForce GTX 780 Ti may be the fastest single-GPU graphics card that money can buy, by a long shot, but only to people who game and don't really care about double-precision floating point GPGPU performance. NVIDIA expects you to shell out $999 on the GTX TITAN for full double-precision FP performance at the moment. For all other intents and purposes, the GTX TITAN is cannibalized, but will it stay that way? VideoCardz spotted these pictures of a black-colored GTX TITAN graphics card in the wild, which retains the cooler design of the original, right up to the "TITAN" logo, but swaps out the silver for black. The publication is referring to the card as a sort of "Black Edition." Speculation is rife about what it could be, a GK110-based card with identical core-configuration to the GTX 780 Ti, 6 GB of memory, full double-precision floating-point performance and slightly higher clock speeds, perhaps?

Palit Microsystems Ltd, the leading graphics card manufacturer, releases the Palit GeForce GTX 780 Ti JetStream, the fastest single-GPU on the planet. The Palit JetStream series of graphics cards features innovative cooling and an optimized product design to deliver gamers the ultimate gaming experience. As the latest and most powerful GPU in the series, the Palit GeForce GTX 780 Ti JetStream guarantees gamers cutting-edge performance and an explosive new gaming experience. This Palit GTX 780 Ti JetStream becomes the new king of single GPU graphics cards.

Palit GeForce GTX 780 Ti JetStream has its GK110 core clocked at 980 MHz, making it much faster than the stock 876 MHz of NVIDIA's reference design. Keeping things frosty under the hood are three jet engine-inspired fans (two 80 mm and one 90 mm) that deliver optimized air flow to the heatsink. Further improving the card's cooling performance is the new heatsink design which consists of a copper base and five copper heatpipes for efficient heat dissipation. It also makes use of DrMOS and an 8-phase PWM design to increase the card's stability and improve its overclocking ability. Palit GTX 780 Ti JetStream also equipped with two 8-pin PCI Express power connectors to ensures the power fully supported from the system power supply unit under heavy loading operation.

Launches of non-reference design GeForce GTX 780 Ti graphics cards are only a matter of time, with HardwareLuxx.de scoring pictures of Galaxy GeForce GTX 780 Ti HOF (hall of fame) graphics card. The HOF series from Galaxy is characterized with its milky white PCB. The GTX 780 Ti HOF could feature the same 2+ slot cooling solution, and the same PCB, as the GTX 780 HOF, which the company launched this July. You can tell it's a GTX 780 Ti by looking at its ASIC marking, GK110-425-B1. Its exact clock speeds are still under the wraps. The source expects non-reference design GTX 780 Ti cards to launch around late-November and early-December, strategically timed against non-reference Radeon R9 290X.

Here's the first picture of an AIC partner-branded GeForce GTX 780 Ti graphics card, by Gigabyte in this case. The card appears to stick to NVIDIA's reference board design, which is practically indistinguishable from the standard GTX 780 and GTX TITAN, except for the special markings. Specifications of the card appear to be consistent with older leaks. Based on the 28 nm "GK110" GPU, the GeForce GTX 780 Ti features 2,880 CUDA cores, 240 TMUs, 48 ROPs, and a 384-bit wide GDDR5 memory interface, holding 3 GB of memory. Its core is clocked at 876 MHz, with a maximum GPU Boost frequency of 928 MHz. The memory is clocked on par with the GTX 770, at 7.00 GHz, churning out a stellar 336 GB/s memory bandwidth. NVIDIA is reportedly targeting a $699.99 price-point with the GTX 780 Ti, which would be $150 costlier than AMD's Radeon R9 290X.

NVIDIA's upcoming GeForce GTX 780 Ti is configured to be a notch above the GTX TITAN after all, as leaked specifications sheets reveal it utilize all components available on the GK110 silicon. Specifications sheets of a Galaxy-branded GTX 780 Ti was leaked to the web by @asder00, which reveal it to feature the full complement of 2,880 CUDA cores on the GK110 silicon, which work out to 240 texture memory units (TMUs). Other specifications include 48 ROPs, and a 384-bit wide GDDR5 memory interface, holding 3 GB of memory. Clock speeds include 876 MHz core, 928 MHz GPU Boost, and 1750 MHz (7.00 GHz GDDR5-effective) memory. With these specifications on paper, the GTX 780 Ti shouldn't have too many problems beating the GTX TITAN, and with it, the Radeon R9 290X from AMD. It features everything there is on the GK110, half the memory of the GTX TITAN, but one that's faster.

With AMD's Radeon R9 290X and the upcoming Radeon R9 290, both NVIDIA's GTX TITAN, and GTX 780 are disrupted at their price points. NVIDIA is fixing its GTX TITAN competitive woes with the GeForce GTX 780 Ti, but it's looking like the GTX 780, despite its price cut to $500, could face trouble from the cheaper Radeon R9 290. NVIDIA's more hands-on solution? Launch a new SKU, that's and backed by non-reference designs for the most part, which some of its add-in card (AIC) partners are referring to as "GeForce GTX 780 GHz Edition."

Simply put, the "GeForce GTX 780 GHz Edition" is your ordinary GTX 780 with increased clock speeds of 1006 MHz core, 1046 MHz GPU Boost, and an untouched 6.00 GHz memory. The card is based on a new stepping of the GK110 silicon, labeled "GK110-300-B1," compared to the original's "GK110-300-A1." Expreview discovered its Inno3D GTX 780 iChill HerculeZ 3000 graphics card to be based on this new silicon, and at its given speeds of 1006/1046/6008 MHz, found to to be about 15 percent faster than a standard GTX 780, and about 7 percent faster than a GTX TITAN. It's also about 6.2 percent faster than an R9 290X on the same test-bed. Power consumption isn't up significantly, and the cooler that's an Arctic Cooling solution, does a good job at keeping the temperatures manageable, and keeps throttle limits away. Find the complete review at the source.

NVIDIA's response to AMD Radeon R9 290X, the GeForce GTX 780 Ti, is likely designed to be faster than the company's GeForce GTX TITAN graphics card, at a lower price, although it turns out that it won't end up anywhere close to AMD's pricing. The GeForce GTX 780 Ti will be formally launched on November 7, 2013. It will be priced at $699.99. The card is shaping up to be an overclocked GTX TITAN, based on the same GK110 silicon, and with 6 GB of memory. The company will also launch a game bundle along the lines of AMD Never Settle, which will include Steam/U-Play keys to Batman: Arkham Origins, Assassin's Creed IV: Black Flag, and Splinter Cell: Blacklist.

With benchmarks of the Radeon R9 290X doing rounds, it's getting increasingly clear that NVIDIA's GeForce GTX 780 won't remain competitive with the Radeon R9 290X for too long; and the R9 290X isn't competitive with the GeForce GTX TITAN enough to warrant a price-cut for the $999.99 SKU. NVIDIA's solution to the tangle is a newer SKU that replaces the GTX 780 from its current $649.99 price-point, which trades blows with the R9 290X. Called GeForce GTX 780 Ti, the SKU could be an overclocked GTX 780, or one that ships with a few more CUDA cores. NVIDIA didn't reveal any technical specifications, other than posting a teaser picture CGI render. To quote NVIDIA on this, "Stay tuned for details."

GeForce GTX 295 showed that its possible to place two GPUs with ludicrously high pin-counts next to each other on a single PCB, and if you get a handle over their thermals, even deploy a 2-slot cooling solution. NVIDIA might be motivated to create such a dual-GPU graphics card based on its top-end GK110 chip, to counter AMD's upcoming "Volcanic Islands" GPU family, or so claims a VideoCardz report, citing sources.

The chips on the card needn't be configured, or even clocked like a GTX Titan. The GTX 780 features just 2,304 of the chip's 2,880 CUDA cores, for example. Speaking of 2,880 CUDA cores, the prospect of NVIDIA developing a single-GPU GeForce product with all streaming multiprocessors on the GK110 enabled, the so-called "Titan Ultra," isn't dead. NVIDIA could turn its attention to such a card if it finds AMD's R9 2xxx within its grasp.

MSI posted a reveal trailer (yeah, we have those for graphics cards now) for its flagship GeForce GTX 780 Lightning graphics card. In addition to the highest factory-overclock among MSI's ranks (currently held by the GTX 780 Gaming Series), the card reportedly features redundant BIOS, and a feature that lifts various BIOS-level limits, protections, or fail-safes. If we understand this correctly, the GTX 780 Lightning should be among the first GK110-based graphics cards to feature over-voltage beyond NVIDIA's limits. Very little is known about the card's design, but given that MSI's new TwinFrozr IV cooling solution leaves GTX 780 Gaming Series with overclocking headroom, and runs cool enough, it wouldn't surprise us if MSI sticks to that cooler, albeit paint it yellow. As for the release date, MSI plans to launch the card on the 7th of August, 2013.The reveal trailer follows.