News Posts matching "GM204"

Return to Keyword Browsing

NVIDIA Breathes Life into Kepler with the GK210 Silicon

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.

EK Introduces Short Type GeForce GTX 970 Water Block

EK Water Blocks, Ljubljana based premium computer liquid cooling gear manufacturer, is proud to introduce a new Full-Cover water block for reference design NVIDIA GeForce GTX 970 series graphics cards, powered by the GM204-200 Maxwell generation GPU.

EK-FC970 GTX is a high-performance Full-Cover water block which directly cools the GPU, RAM as well as VRM (voltage regulation module) as water flows directly over these critical areas thus allowing the graphics card and it's VRM to remain stable under high overclocks.

NVIDIA GeForce GTX 960 to Retain Memory Bus from GTX 970

Among other things like CUDA core and TMU counts, NVIDIA was expected to give its next mid-range graphics card, the GeForce GTX 960, a narrower memory bus, with 3 GB of memory, if not less. A sample sniffed out by India's overly transparent customs department, en route testing facilities in the country, reveal that it's not the case.

The GeForce GTX 960, according to description given in the shipping manifest of the sample, features 4 GB of memory, with a full 256-bit wide memory interface. It also reveals clock speeds to be in the neighborhood of 993 MHz core, with 6.00 GHz memory (GDDR5-effective). It doesn't, however, confirm that the GTX 960 is based on a cut-down GM204 silicon. This could still be different chip, the so-called GM206, which succeeds the GK106.

ELSA Launches GeForce GTX 980 SAC Graphics Card

ELSA launched its first custom-design GeForce GTX 980 graphics card, bearing its SAC (silent air-cooling) badge. The card retains its signature Inno3D-sourced look, featuring a cooling solution that lets you easily clean it, to keep its cooling performance up to the mark. The cooling solution consists of an aluminium fin-stack heatsink, to which heat drawn from the GPU, is fed by a number of copper heat pipes, and ventillated by a pair of fans. The shroud on which the two are suspended, can be detached, by pulling out thumb-screws. The PCB under this cooler appears to stick to NVIDIA reference design.

The GTX 980 SAC from ELSA is factory-overclock, although the overclock is nothing to write home about. The core is clocked at 1152 MHz, with 1253 MHz GPU Boost, and 7.00 GHz memory, compared to NVIDIA reference clocks of 1126 MHz core, and 1216 MHz GPU Boost. Based on the 28 nm GM204 silicon, the GeForce GTX 980 features 2,048 CUDA cores, 128 TMUs, 64 ROPs, and a 256-bit wide GDDR5 memory interface, holding 4 GB of memory. The card draws power from a pair of 6-pin PCIe power connectors. Display outputs include three DisplayPorts, and one each of HDMI 2.0 and dual-link DVI.

NVIDIA Could Review GTX TITAN-Z Pricing for the Retail Channel

Under pressure from its own recently launched GeForce GTX 980 cannibalizing its $3,000 flagship GeForce GTX TITAN-Z (you can buy GTX 980 Quad-SLI at $2,200), NVIDIA could finally review its pricing, in the retail channel. The company recently reduced its price for the OEM channel, letting pre-built gaming PC manufacturers source the card at a lower price, whether those price-savings were transferred to the end-users, is a different question. To what measure NVIDIA could lower prices of the GTX TITAN-Z, is not known at this point. Retailers like OCUK were seen offering their GTX TITAN-Z cards at a slightly reduced price, last week. Across the big pond, American retailer Newegg sold-out an ASUS-branded GTX TITAN-Z for as low as US $1,500.

In the wake of GTX 980, AMD cut prices of its dual-GPU flagship product, the Radeon R9 295X2, down to $999. It's hard to imagine that competition from this card, and the GTX 980, are the only factors driving down prices of the GTX TITAN-Z in such a big way. Could NVIDIA be working on its next dual-GPU flagship graphics card already? Perhaps one based on a pair of GM204 chips, with thermal and power requirements as low as those of the GTX 690? Watch this space for more.

Source: Expreview

Gigabyte Outs Second GeForce GTX 970 WindForce OC Graphics Card

Gigabyte rolled out its second factory-overclocked GeForce GTX 970 graphics card after the GTX 970 G1.Gaming, the GTX 970 WindForce OC (model: GV-N970WF3OC-4GD). Its board design is identical to that of the GTX 970 G1.Gaming, but its factory-overclock is milder. The card ships with factory-overclocked speeds of 1114 MHz core, 1253 MHz GPU Boost, and 7.00 GHz memory, compared to the 1178 MHz core and 1253 MHz GPU Boost of the G1.Gaming, and NVIDIA reference speeds of 1050 MHz core, and 1178 MHz GPU Boost. The card features Gigabyte's new generation WindForce 3X cooling solution, which is capable of handling heavy thermal loads, which may look surplus to requirements for a 170W GPU, but serves to keep temperatures off the hair-trigger 80°C thermal limit. Based on the 28 nm GM204 silicon, the GeForce GTX 970 features 1664 CUDA cores, 104 TMUs, 64 ROPs, and a 256-bit wide GDDR5 memory interface, holding 4 GB of memory. Expect this card to be priced around US $350.

MSI Holding Off on GTX 980 Lightning, in Anticipation of Bigger Chip

MSI is reportedly holding off on launching an overclocking-grade GeForce GTX 980 graphics card that you can ride into record-setting competitions, based on its coveted OC Series Lightning brand. The company probably wants to avoid a repeat of last year, when launch of its GeForce GTX 780 Lightning graphics card was closely trailed by NVIDIA's launch of the inherently faster GeForce GTX 780 Ti, and resulting price-cuts for the GTX 780, throwing MSI's inventories in jeopardy. The year before that, MSI launched the Radeon HD 7970 Lightning just weeks before AMD launched the faster HD 7970 GHz Edition.

MSI could stick to selling its Gaming Series GTX 980, equipped with the new Twin Frozr V cooling solution, even offering factory-overclocked variants of the card. This also leads one to wonder if NVIDIA has a third, high-end graphics card based on the GM204 silicon, on the 'cards,' one with higher clock speeds, or a higher temperature-based clock speed throttle, which is currently at a hair-trigger 80°C on the GTX 980.


Source: NordicHardware

EK Unveils Its NVIDIA GeForce GTX 980 Water Blocks

EK Water Blocks, Ljubljana based premium water cooling gear manufacturer, is proud to introduce full-cover water cooling solution for NVIDIA reference (NVA-1G401) design GeForce GTX 980 series graphics cards, based on Maxwell GM204 graphics processor.

EK-FC980 GTX directly cools the GPU, RAM as well as VRM (voltage regulation module) as water flows directly over these critical areas thus allowing the graphics card and it's VRM to remain stable under high overclocks.

NVIDIA Announces the GeForce GTX 980 and GeForce GTX 970

At the Game24 event held at several locations around the world, and online, NVIDIA launched its next-generation GeForce GTX 900 series high-end graphics cards, led by the GeForce GTX 980, and the GeForce GTX 970. The two are based on the company's new 28 nm GM204 silicon, derived from the "Maxwell" GPU architecture. The GeForce GTX 980 leads the pack, featuring 2,048 CUDA cores, 128 TMUs, 64 ROPs, and a 256-bit wide GDDR5 memory interface, holding 4 GB of memory.

The GeForce GTX 970, on the other hand, features 1,664 CUDA cores, 104 TMUs, 64 ROPs, and the same memory configuration. Both cards feature TDP rating of under 170W, and clock speeds above the 1 GHz mark. The GTX 980 features clock speeds of 1126 MHz core, 1216 MHz GPU Boost, and 7.00 GHz memory. The GTX 970, on the other hand, offers 1050 MHz core, 1178 MHz GPU Boost, and 7.00 GHz memory. Both cards offer 224 GB/s of memory bandwidth, but feature technologies that help them make the most of it, such as 3rd generation Delta-color compression, a lossless texture compression algorithm. The GeForce GTX 980 starts at US $549, while the GTX 970 starts at $329.

Gainward Readies GeForce GTX 970 Phantom, GTX 980 Reference

Gainward will launch three graphics cards based on the GM204 silicon later this week, its GeForce GTX 980 board, which sticks to NVIDIA's reference board design, a slightly non-reference GTX 970 which comes with reference clock speeds, and a completely non-reference Phantom Edition board, featuring the company's iconic air-pulling concealed fans, which could ship with a factory-overclock. The GTX 970 Phantom could ship with some of the highest factory-overclocks in the lineup, with core clock speeds of 1152 MHz, GPU Boost speeds of 1304 MHz, and an untouched 7.00 GHz memory.
Source: ComputerBase.de, via VideoCardz

Even More GeForce GTX 980 and GM204 Specs Tumble Out

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.

NVIDIA GeForce GTX 980 and GTX 970 Pricing Revealed

Apparently, NVIDIA is convinced that it has a pair of winners on its hands, with its upcoming GeForce GTX 980 and GTX 970 graphics cards, and is preparing to price them steeply. The GeForce GTX 980 is expected to start at US $599, nearly the same price as the GeForce GTX 780 Ti. The GTX 970, on the other hand, will start at US $399, danger-close to cannibalizing the GTX 780.

Across the brands, the GTX 980 is launching at the same pricing AMD's Radeon R9 290X launched at; and the GTX 970 at that of the R9 290. AMD's cards have since settled down to $449 for the R9 290X, and R9 290 at $350. Both the GTX 980 and GTX 970, will be available in non-reference board designs, although reference-design GTX 980 will dominate day-one reviews. Based on the 28 nm GM204 silicon, the GTX 980 features 2,048 CUDA cores, 128 TMUs, 32 ROPs; while the GTX 970 features 1,664 CUDA cores, and 104 TMUs. Both feature 256-bit wide memory interfaces, holding 4 GB of GDDR5 memory.Source: 3DCenter.org

NVIDIA GeForce GTX 980 Specifications Detailed Some More

NVIDIA's next-generation flagship single-GPU graphics card, the GeForce GTX 980, will feature 2,048 CUDA cores, and not the previously thought of 1,920. These 2,048 will be spread across 16 Streaming Multiprocessor Maxwell (SMM) units. Its sibling, the GeForce GTX 970, in comparison, features 1,664 cores spread across 13 SMMs. The TDP of GTX 980 is rated at just around 175W, and that of the GTX 970 at 145W. In comparison, the GK104 silicon, on the GeForce GTX 770, is rated at 230W. The kicker? Both GM204 and GK104 are based on the same 28 nm silicon fab node. Both cards feature power inputs consisting of two 6-pin PCIe power connectors. NVIDIA is expected to launch the two, on the 19th of September, 2014.


Source: VideoCardz

NVIDIA Readies GeForce GTX 960

It looks like GeForce GTX 980 and GTX 970 won't be the only Maxwell-based graphics cards NVIDIA plans to launch within the next 30 days. The company is readying a third SKU based on the chip, the GeForce GTX 960. The company's next sub-$300 graphics card, the GTX 960 will be launched some time in mid to late-October, 2014. The company's GTX 970 and GTX 980 will come out later this week (19th September), timed with the Game24 event. There's no word on the exact specs of the GTX 960.


Source: Hermitage Akihabara

NVIDIA GeForce GTX 980 Reference Board Pictured

Here are the first pictures of the reference-design GeForce GTX 980 graphics card. This gorgeous looking board is something to get used to, as it could be a while before we see non-reference boards for the GTX 980. Its smaller sibling, the GTX 970, could launch with non-reference design boards right off the bat. Outwardly, the GTX 980 looks exactly like a GeForce GTX 780 reference board, with NVIDIA's space-age cooling shroud.

The only difference here is a metal backplate decking up the card. This backplate isn't cooling anything, and traces reveal that the card is indeed holding just 8 memory chips, confirming its 256-bit wide GDDR5 memory interface. The GeForce GTX 980 is based on NVIDIA's new GM204 silicon, a derivative of the company's new "Maxwell" architecture, which made its debut with the GeForce GTX 750 Ti. The package of GM204 looks roughly as big as a GK104. The card draws power from a pair of 6-pin PCIe power connectors, and uses a 4+1 phase VRM to condition power.

Source: VideoCardz

Radeon R9 290X Sees Price Cuts

AMD is apparently working with its add-in board manufacturers and retailers to bring down prices of its flagship single-GPU graphics card, the Radeon R9 290X. The card can now be had for as low as $449, non-reference design, factory-overclocked cards starting at a $50 premium. Prices could settle down somewhere between $450 and $500. This closely follows AMD's move to bring down price of its dual-GPU flagship Radeon R9 295X2 by a whopping 34 percent, down to $999, offering performance competitive to the $2999 GeForce GTX TITAN-Z. NVIDIA is preparing two new graphics cards competitive in performance to the Radeon R9 290 series, the GeForce GTX 970 and GTX 980. The two are based on the company's new 28 nm "GM204" silicon, implementing the "Maxwell" GPU architecture.

ZOTAC GeForce GTX 970 Pictured

Here's the first picture of a ZOTAC-branded GeForce GTX 970 graphics card, the first one shot with its box, confirming the GTX 900 series nomenclature NVIDIA adopted. ZOTAC's card appears to be based on a non-reference PCB, and features a rather compact-looking aluminium fin-stack heatsink, which is ventilated by a pair of 80 mm spinners. The box doesn't flash any "OC" markings, so it's safe to assume that he card runs at its reference clock speeds of 1051 MHz core, 1178 MHz GPU Boost, and 7012 MHz (GDDR5-effective) memory. The card features 4 GB of memory across a 256-bit wide memory interface. Based on the 28 nm "GM204" silicon, the GeForce GTX 970 reportedly features 1,664 CUDA cores, 138 TMUs, and 32 ROPs.


Source: Overclock.net Forums

MSI Teases GeForce GTX 970 Gaming with TwinFrozr V

MSI showed off the first moneyshot render of its GeForce GTX 970 Gaming graphics card. Trailing an earlier leaked render viewed from the top of the card. The new one reveals MSI giving the TwinFrozr V cooler a bold 2-tone cooler shroud, holding a pair of 100 mm fans that ventilate a chunky copper heat-pipe fed aluminium fin stack. MSI could give the GTX 970 a factory-overclock to take advantage of its meaty cooling solution. Based on the 28 nm GM204 silicon, the GeForce GTX 970 is said to feature 1,664 CUDA cores based on the "Maxwell" architecture, 138 TMUs, 32 ROPs, a 256-bit wide memory interface, and 4 GB of memory. It will launch a little later this month.

Galaxy GeForce GTX 970 Pictured, Specs Confirmed, Early Benchmarks Surface

Here are some of the first pictures of an AIC partner branded NVIDIA GeForce GTX 970 graphics card, the Galaxy GTX 970 GC. Spotted across Chinese PC enthusiast forums and social networks, the latest set of leaks cover not just pictures of what the GTX 970 looks like, but also what's under its hood. To begin with, Galaxy's card appears to be built for the high-end market segment. A meaty twin-fan aluminium fin-stack heatsink, coupled by a spacey backplate cover a signature Galaxy blue PCB, holding NVIDIA's new GTX 970 GPU, and 4 GB of GDDR5 memory. The card appears to feature a high-grade VRM that draws power from a combination of 8-pin and 6-pin PCIe power connectors.

MSI TwinFrozr V Cooling Solution Teased

MSI teased the first CGI sketches of its next-generation TwinFrozr V cooling solution, designed for high-end GPUs, under its Gaming Series. A prototype of the cooler made its first appearance at Computex 2014, in June. The cooler features a large dual-stack aluminium fin heatsink to which heat drawn from the GPU is fed by five 8 mm thick nickel-plated copper heat pipes. The fin-stack is ventilated by a pair of what's now appearing to be two 100 mm fans. MSI is rumored to be innovating a new impeller design that steps up air-flow to noise ratio.

It's interesting to note that in its CGI render of a card equipped with this cooler, MSI showed a PCB with two NVIDIA SLI bridge fingers, and two 6-pin PCIe power inputs. Could this be the NVIDIA GeForce GTX 980, in effect making the card the MSI GTX 980 Gaming OC? Wait until the 19th of September to find out. GeForce GTX 980 and GTX 970 are NVIDIA's next high-end GPUs, based on the 28 nm "GM204" silicon, derived from the company's new "Maxwell" architecture.


Source: MyDrivers

NVIDIA to Launch GeForce GTX 880 in September

NVIDIA is expected to unveil its next generation high-end graphics card, the GeForce GTX 880, in September 2014. The company could tease its upcoming products at Gamescom. The company is reportedly holding a huge media event in California this September, where it's widely expected to discuss high-end graphics cards based on the "Maxwell" architecture. Much like AMD's Hawaii press event that predated actual launch of its R9 290 series by several weeks; NVIDIA's event is expected to be a paper-launch of one or more graphics cards based on its GM204 silicon, with market availability expected in time for Holiday 2014 sales.

The GM204 is expected to be NVIDIA's next workhorse chip, which will be marketed as high-end in the GeForce GTX 800 series, and performance-segment in its following GTX 900 series; much like how the company milked its "Kepler" based GK104 across two series. It's expected to be built on the existing 28 nm process, although one cannot rule out an optical shrink to 20 nm later (like NVIDIA shrunk the G92 from 65 nm to 55 nm). The GTX 880 reportedly features around 3,200 CUDA cores, and 4 GB of GDDR5 memory.Source: VideoCardz

NVIDIA Timing GeForce GTX 880 Launch with Gamescom?

NVIDIA is reportedly preparing new graphics cards based on the "Maxwell" architecture, which judging by the purported 3DMark FireStrike screenshots, could be high-end parts, maybe even the GeForce GTX 880. The cards could launch along the sidelines of Gamescon, a gaming expo held each year in Cologne, Germany, around mid-August. Given how most information to date points to a later-Q3, maybe even mid-Q4 launch of these cards, it wouldn't surprise us if NVIDIA merely teases these cards. NVIDIA's next big consumer GPU is widely expected to be the GM204, a performance-segment chip based on the "Maxwell" architecture.

Source: VideoCardz

NVIDIA Moving Around the Fabled GeForce GTX TITAN II

NVIDIA is moving around engineering samples of what it describes as "GM200 A1 graphics processor," in its shipping manifest. The sample was making its way from Taiwan, to Bangalore, India, from where it's likely pushed to the company's facilities in Bangalore and Hyderabad. A1 steppings of NVIDIA chips are usually pre-production, and bound for just a few more rounds of testing, before being upgraded to "A2" and mass-produced. German tech site 3DCenter.org also pulled out some likely specifications from its sources.

To begin with, the GM200, like the GM204, will be built on existing 28 nm silicon fabrication process, as both NVIDIA and AMD appear to have suffered design setbacks due to their common foundry partner, TSMC, not being able to set its next-gen 20 nm node up to speed in time. The GM200 is expected to feature over 4,000 CUDA cores, although the exact number is unknown. It is expected to widen the memory bus to 512-bit. Given the existing process, the GPU will be huge. Over 600 mm² huge. NVIDIA will probably bank on the energy efficiency of its "Maxwell" architecture to cope with thermal loads put out by a chip that big. The GM200-based "GeForce GTX TITAN II" could launch in the first half of 2015.

Source: 3DCenter.org

GeForce GTX 880 ES Intercepted En Route Testing Lab, Features 8 GB Memory?

An engineering sample (ES) of the GeForce GTX 880 was intercepted on its way from a factory in China, to NVIDIA's development center in India, where it will probably undergo testing and further development. The shipping manifest of the courier ferrying NVIDIA's precious package was sniffed out by the Chinese press. NVIDIA was rather descriptive about the ES, in its shipping declaration. Buzzwords include "GM204" and "8 GB GDDR5," hinting at what could two of the most important items on its specs sheet. GM204 is a successor of GK104, and is rumored to feature 3,200 CUDA cores, among other things, including a 256-bit wide memory bus. If NVIDIA is cramming 8 GB onto the card, it must be using some very high density memory chips. The manifest also declares its market value at around 47,000 Indian Rupees. It may convert to US $780, but adding all taxes and local markups, 47,000 INR is usually where $500-ish graphics cards end up in the Indian market. The R9 290X, for example, is going for that much.
Sources: ChipHell, VideoCardz

No 20 nm GPUs from AMD This Year

It's not just NVIDIA, which will lack 20 nm GPUs in its portfolio this year. AMD senior vice-president Lisa Su, responding to a question by Wells Fargo, in its Q1 investors call, confirmed that her company will stay on 28 nm throughout 2014, and it's only later that it will move on to 20 nm, and FinFET after that. "I think what I said earlier sort of what we're doing in terms of technology strategy, we are 28 this year, we have 20-nanometer in design, and then FinFET thereafter. So that's the overall product portfolio," she said.

AMD and NVIDIA manufacture their GPUs on a common foundry, TSMC, which has faced delays in implementing its 20 nanometer silicon fab node transition, forcing both companies to come up with new GPUs on existing 28 nm nodes. A huge leap in performance could be a tough ask for those new GPUs. NVIDIA is expected to tape out its performance-segment GM204 and mid-range GM206 chips, both of which are 28 nm, later this month, and the first GeForce GTX products based on the two are expected to roll out by late-Q4 2014 and early-Q1 2015, respectively.


Source: Seeking Alpha
Return to Keyword Browsing