After finally reaching mature yields (comparable to those of planar NAND processes), Micron's 32-layer first generation 3D NAND has grown increasingly prominent in the company's NAND output. Now, the company is looking to ramp-up production of their (currently sampling) 64-layer 3D NAND, promising "meaningful output" by the end of December 2017, looking for an 80% increase in total GB per wafer and a 30% decrease in production costs.

When it comes to the graphics subsystem memory, Micron is looking to transition their 20nm production to a "1x nm" (most likely 16nm) node, in a bid to improve cost per GB by around $20, with introduction of 16nm GDDR5 memory to be introduced later this year. However, GDDR5X volume is expected to grow significantly, in a bid to satisfy bandwidth-hungry uses through GPUs (like NVIDIA's GTX 1080 and potentially the upcoming 1080 Ti) and networking, with GDDR6 memory being introduced by the end of 2017 or early 2018. The company is still mum on actual consumer products based on their interpretation of the 3D XPoint products through their QuantX brand, though work is already under way on the second and third generation specifications of this memory, with Micron planning an hitherto unknown (in significance and product type) presence in the consumer market by the end of this year.

Joining in on other manufacturer's offerings for low-profile graphics solutions for HTPCs and other small fomr factor PCs, GALAX has recently announced that they too will be offering low-profile variants of the GTX 1050 and GTX 1050 Ti graphics cards. With 2 GB and 4 GB of GDDR5 at 7 Gbps respectively, both over a 128 bit bus, both cards make do without any additional power connectors, making do with exclusive power delivery from the PCIe socket.

The core clocks are set at 1303 MHz and 1366 MHz base clocks for the 1050 and the 1060 Ti respectively, with boost clocks seeing a healthy (but somewhat traditional)n uplift to 1417 Mhz and 1468 Mhz. As to connectors, both cards carry 1x DisplayPort 1.4, 1x DVI-D, and 1x HDMI 2.0b. Considering the cards' characteristics, they probably won't be sold much higher than NVIDIA's reference MSRP for both cards: $139 for the GTX 1050 Ti and $109 for the GTX 1050.

At its WX call, AMD focused on shifts in creativity from traditional design flows such as Solidworks, Adobe and Autodesk towards game engines as solutions for design visualization (Unreal Engine, Unity, CryEngine, or Autodesk's own Stingray platform), which signal changes in the creator ecosystem. Thanks to globalization, the Internet, and the available wealth of knowledge one can access through it, the line between amateurs and professionals is becoming more and more blurred. Now, those who would once be called amateurs are also using professional tools, and AMD plans to be at the forefront of technologies empowering creators to deliver their vision.

Radeon PRO serves to give creators more flexible and powerful solutions, leveraging open-source resources and centering the ecosystem back on creators and the tools they choose to use, with focused support on VR. As such, AMD is giving them the tools they need, by introducing three new products featuring the Polaris architecture, including 3 year standard + 7 year free extended warranty (including components such as the PCB itself, the PCI-Express slot, and the heatsinks), with AMD taking that extra 7 years as company commitment towards the quality of their products. Those three products are the WX4100, the WX5100, and the WX7100, and have planned, staggered availability throughout November.

NVIDIA is giving finishing touches to its next enthusiast-segment graphics card based on the "Pascal" architecture, the GeForce GTX 1080 Ti. Its specifications were allegedly screengrabbed by a keen-eyed enthusiast snooping around NVIDIA website, before being redacted. The specs-sheet reveals that the GTX 1080 Ti is based on the same GP102 silicon as the TITAN X Pascal, but is further cut-down from it. Given that the GTX 1080 is unflinching from its $599-$699 price-point, with some custom-design cards even being sold at over $800, the GTX 1080 Ti could either be positioned around the $850-mark, or be priced lower, disrupting currently overpriced custom GTX 1080 offerings. By pricing the TITAN X Pascal at $1200, NVIDIA appears to have given itself headroom to price the GTX 1080 Ti in a way that doesn't cannibalize premium GTX 1080 offerings.

The GTX 1080 Ti is carved out of the GP102 silicon by disabling 4 out of 30 streaming multiprocessors, resulting in 3,328 CUDA cores. The resulting TMU count is 208. The card could retain its ROP count of 96. The card will be endowed with 12 GB of GDDR5 memory across the chip's 384-bit wide memory interface, instead of GDDR5X on the TITAN X Pascal. This should yield 384 GB/s of memory bandwidth, significantly lesser than the 480 GB/s bandwidth the TITAN X Pascal enjoys, with its 10 Gbps memory chips. The GPU is clocked at 1503 MHz, with 1623 MHz GPU Boost. The card's TDP is rated at 250W, same as the TITAN X Pascal.

Sony today announced an addition to its PlayStation family of entertainment systems, with the new PlayStation 4 Pro (PS4 Pro). The PS4 Pro isn't meant to succeed the PS4, and is still a "current generation" console, in that all of its games are compatible with the PS4, and there won't be PS4 Pro-exclusive games. What sets it apart is higher-resolution game rendering that looks best on 4K Ultra HD TVs. This doesn't necessarily mean that the games are being rendered at 3840 x 2160 pixels. They could be rendered at resolutions higher than 1080p, and using just the right upscaling algorithms, made to look significantly better than 1080p. Another key visual component is hardware HDR. The PS4 Pro falls in line with Sony's big push for 4K HDR TVs this holiday shopping season.

Under the hood, the PlayStation 4 Pro features an all new semi-custom SoC designed by Sony and AMD. The CPU component sticks to the "Jaguar/Puma" architecture, with 8 cores, but features higher CPU clock speeds. The 8 GB GDDR5 memory is said to run at higher clocks, too, and is bolstered by new lossless memory compression tech by AMD. The GPU component is where the action is. The GPU features over double the shading power of the PS4, and is based on the "Polaris" GPU architecture. The PS4 Pro also ships with a bigger 1 TB hard drive. There is seamless interplayability between PS4 and PS4 Pro. PS4 games you already own are being added with higher-resolution content meant for PS4 Pro. The console also has the muscle for PS-VR. For those still holding on to 1080p HDTVs, the PS4 Pro rewards with higher visual detail, more geometric complexity, and better effects, at that resolution. The PS4 Pro starts at $399.

AMD today unleashed the new Radeon RX 460 graphics card, engineered from the ground up for eSports gamers who demand pristine HD gaming, extraordinarily smooth beyond-HD streaming capabilities, and a distinguished assortment of future-proof gaming technologies. With an ultra-quiet cooling solution and sub-75W power footprint, the Radeon RX 460 brings this wide array of enthusiast class features with a stunning SEP starting at $109.

"The Radeon RX 460 delivers the perfect balance of price, power, performance and package size, the four key pillars of modern GPUs," said Raja Koduri, senior vice president and chief architect, Radeon Technologies Group, AMD. "The Radeon RX Series is built on architecture designed for extraordinary power efficiency and is especially well suited to desktop gaming PCs targeted at mainstream price points. Radeon RX 460 users will enjoy an optimized software and hardware graphics card solution ideal for both eSports and iCafé systems."

TUL Corporation, a leading and innovative manufacturer of AMD graphic cards since 1997, has released a brand new video card in PowerColor Red Dragon RX 460 2GB/4GB GDDR5. It is based on AMD's latest GCN 4 architecture designed for GloFo 14nm FinFET that delivers premium VR capability, increased level of performance, smooth VR, seamless support for next-gen gaming monitors and CPU-free game streaming or recording. Furthermore, the model also supports AMD's newest technologies such as Direct 12 and Vulkan, FreeSync, and Liquid VR.

PowerColor Red Dragon RX 460 4 GB GDDR5 utilizes 4 GB of GDDR5 memory with 896 stream processors, ships with 1212 MHz core clock speed, and has 1750 MHz memory clock speed which is connected via a new high speed 128-bit memory interface.

The official specifications of two the two upcoming mainstream graphics cards by AMD, the Radeon RX 470, and the Radeon RX 460, were leaked to the web as slides from the company's official press presentation ahead of their early-August product launches. The RX 470 is based on the same "Ellesmere" Polaris10 silicon as the RX 480. It features 2,048 stream processors across 32 GCN compute units; 128 TMUs, 32 ROPs, and 4 GB of memory across a 256-bit wide GDDR5 memory interface.

The RX 470 features clock speeds of 926 MHz core, 1206 MHz boost, and 6.6 Gbps memory, working out to 211 GB/s memory bandwidth. The RX 460, on the other hand, is based on the "Baffin" Polaris11 silicon, featuring 896 stream processors, 48 TMUs, 16 ROPs, and 4 GB of memory across a 128-bit wide GDDR5 memory interface. Its core is clocked at 1090 MHz, with 1200 MHz boost, and 7 Gbps memory, working out to 112 GB/s memory bandwidth. The RX 470 draws power from a single 6-pin PCIe power connector as its TDP is rated at 120W; the RX 460 relies entirely on the PCIe slot for its power, as its TDP is rated at <75W. The RX 470 will be available from 4th August, 2016; with the RX 460 following on 8th August.

AMD confirmed specifications of its second and third "Polaris" architecture graphics cards in a leaked presentation, the Radeon RX 470, and the Radeon RX 460. The RX 470 will be AMD's attempt at a graphics card that plays everything at 1080p resolution, under $150. The Radeon RX 460, on the other hand, is based on the new 14 nm Polaris11 "Baffin" silicon, and could be ideal for MOBA games with light GPU requirements.

The Radeon RX 470 is carved out from the Polaris10 "Ellesmere" silicon that the RX 480 is based on, it features 2,048 stream processors across 32 GCN compute units, 128 TMUs, 32 ROPs, and a 256-bit wide GDDR5 memory interface, holding 4 GB of memory. The card draws power from a single 6-pin PCIe power connector. The Radeon RX 460, on the other hand, features 896 stream processors across 14 compute units, 2 GB of GDDR5 memory across a 128-bit wide memory interface, and relies on the PCI-Express slot entirely for power. The reference RX 460 board looks quite similar to the Radeon R9 Nano, but features a simpler spiral heatsink under the fan. Despite rumors to the contrary, it looks like Vega is on-course for a 2017 launch after all.

Here's one of the first pictures of the NVIDIA GeForce GTX 1060 Founders Edition (reference) PCB. The PCB is about 2/3rds the length of the actual card, and despite that, it's pretty barren. Power is drawn from a 6-pin PCIe power connector, however, this connector isn't on the PCB, but is on a receptacle towards the end of the cooler. NVIDIA designed this in response to complaints that on cards with PCB shorter than the cooler, the power connector would be in the middle of the card. It would also block the illuminated GeForce GTX logo along the top.

The 6-pin PCIe power receptacle connects to the card at big solder points. This approach has one downside. If you want to change the cooler (to, say, an aftermarket air cooler), you will have to deal with that ugly cabling. The card uses a simple 3+1 phase VRM to power the GPU, with its TDP rated at just 120W. The GP106 GPU is neighbored by six 8 Gbps GDDR5 memory chips populating its 192-bit memory bus. There's no SLI support. Display outputs include three DisplayPort 1.4, and one each of HDMI 2.0b and DVI.

MSI unveiled a pair of APAC-exclusive DUKE Edition graphics cards, the GeForce GTX 1080 DUKE Edition, and the GeForce GTX 1070 DUKE Edition. The two are characterized by a large triple-fan cooling solution that's optimized for the tropical climate of its target markets; and features a custom-design PCB that appears to be similar in design to the Gaming X series cards.

The cooler is longer than the PCB itself. A back-plate covers the PCB. The cooler features a twin-stack aluminium-fin cooler that's ventilated by three 90 mm fans, an RGB LED-lit "DUKE" logo tops the cooler. The GTX 1080 DUKE Edition, in its "OC Mode" offers clock speeds of 1708 MHz core, 1847 MHz GPU Boost, and 10108 MHz (GDDR5X-effective) memory; while the GTX 1070 DUKE Edition, in its "OC Mode" offers 1607 MHz core, 1797 MHz GPU Boost, and 8108 MHz (GDDR5-effective) memory.

Here's a sight for sore eyes, a stream of Radeon RX 480 graphics cards by XFX are making their way down the production line to packaging. The picture reveals that XFX could top up the reference-design RX 480 board with a back-plate, to make it even more marketable. To give you an idea of just how cool the 14 nm "Ellesmere" chip runs, with the cooler shroud taken apart, you'll find that the GPU is cooled by just a chunky, monolithic aluminium heatsink with a copper core, and not an elaborate aluminium fin-stack/channel setup with heat-pipes crisscrossing it. A base-plate cools the memory and VRM. In related news, ChipHell scored a GPU-Z screenshot of the XFX Radeon RX 480, confirming its stream processor count of 2,304 and GPU clocks of 1328 MHz, with the memory ticking at 8 GHz (GDDR5-effective).

In its post-E3 press-deck, AMD confirmed the specifications of its Polaris10 "Ellesmere" and Polaris11 "Baffin" GPUs. The two chips will drive at least three desktop discrete SKUs between them, the Radeon RX 480, the RX 470, and the RX 460. Of these, the RX 480 and RX 470 appear to be based on the "Ellesmere" silicon. This chip features 2,304 stream processors spread across 36 compute units (CUs), and a 256-bit wide GDDR5 memory interface. The "Baffin" silicon, on the other hand, features 1,024 stream processors spread across 16 CUs, and a 128-bit GDDR5 memory interface, likely ticking at 128 GB/s.

AMD made a bold move in launching its first "Polaris" architecture based performance-segment GPU, the Radeon RX 480 at a starting price of US $199. The company claims that it will perform on-par with $500 graphics cards from the previous generation, directly hinting at performance being on par with the Radeon R9 Fury and R9 Nano. Although it's not in the league of the GTX 1070 or the GTX 1080, this level of performance at $199 could certainly disrupt things for NVIDIA, as it presents an attractive option for people still gaming on 1440p and 1080p resolutions (the overwhelming majority). The R9 Fury can handle any game at 1440p.

The Radeon RX 480 is based on the 14 nm "Ellesmere" silicon, fabbed by GlobalFoundries. It's publicly known that GloFlo has a 14 nm fab in Malta (upstate New York), USA. The RX 480 is based on AMD's 4th generation Graphics CoreNext architecture, codenamed "Polaris." It features 2,304 stream processors, spread across 36 compute units (CUs). Its single-precision floating point performance is rated by AMD to be "greater than 5 TFLOP/s." The chip features a 256-bit wide GDDR5 memory interface, with memory clocked at 8 Gbps, yielding memory bandwidth of 256 GB/s. There will be two variants of this card, 4 GB and 8 GB. It's the 4 GB variant that starts at $199, the 8 GB variant is expected to be priced at $229. AMD confirmed that the GPU will support DisplayPort 1.4 although it's certified up to DisplayPort 1.3. The typical board power is rated at 150W. The card could be available from 29th June.

Here's the first picture of an NVIDIA reference-design PCB for the GeForce GTX 1070. The PCB (PG411) is similar to that of the GTX 1080 (PG413), except for two major differences, VRM and memory. The two PCBs are pictured below in that order. The GTX 1070 PCB features one lesser VRM phase compared to the GTX 1080. The other major difference is that it features larger GDDR5 memory chips, compared to the smaller GDDR5X memory chips found on the GTX 1080. These are 8 Gbps chips, and according to an older article, its clock speed is maxed out to specifications, at which the memory bandwidth works out to be 256 GB/s. The GeForce GTX 1070 will be available by 10th June.

NVIDIA posted the product page of its upcoming GeForce GTX 1070 graphics card, confirming its clock-speeds, and related specifications. The card features a nominal GPU clock speed of 1506 MHz, with a maximum GPU Boost frequency of 1683 MHz. The memory is clocked at 2000 MHz (actual), or 8 GHz (GDDR5-effective), working out to a memory bandwidth of 256 GB/s. The company also rates the card's single-precision floating point performance at 6.45 TFLOP/s. Other key specs include 1,920 CUDA cores, 120 TMUs, and 64 ROPs. The GeForce GTX 1070 goes on sale, on the 10th of June, 2016.

Industry sources revealed to TechPowerUp some pretty interesting specifications of AMD's two upcoming GPUs based on the 4th generation Graphics CoreNext "Polaris" architecture. The company is preparing a performance-segment GPU and a mainstream one. It turns out, that the performance-segment chip, which the press has been referring to as "Ellesmere," could feature 32 compute units (CUs), and not the previously thought 40.

Assuming that each CU continues to consist of 64 stream processors (SP), you're looking at an SP count of 2,048. What's more, this chip is said to offer a single-precision floating point performance of 5.5 TFLOP/s, as claimed by AMD. To put this into perspective, the company had claimed 5.2 TFLOP/s for the "Hawaii"/"Grenada" based FirePro W9100, which launched earlier this February, and that SKU featured all 2,816 SP present on the chip. So this chip is definitely faster than most "Hawaii" based SKUs.

AMD posted a new webpage for its upcoming "Polaris" GPU architecture, outlining its various innovations - 4th gen. Graphics CoreNext, 4K H.265 60 Hz game-streaming, next-generation display engine with support for DisplayPort 1.3 and HDMI 2.0, XConnect Technology, and the foundation of GPUOpen. In this page, the company inadvertently leaked pictures of its upcoming Polaris 10 "Ellesmere" and Polaris 11 "Baffin" ASICs.

The mast image of the page has a faded 3-quarter shot of a "small" GPU with a die that's almost 30% of the package area. This hints at Polaris 11 "Baffin." This chip is rumored to feature a 128-bit GDDR5/GDDR5X memory interface, and so its pin-count, and conversely, package-size is less. Then in its "4th gen GCN" heading image, AMD showed a picture of a bigger GPU. At first glance, you could assume that it's either "Tonga XT" or "Tahiti" looking at its support brace, but VideoCardz observed that the on-package electrical components in this image are arranged nothing like on the "Tonga" or "Tahiti." This could very well be Polaris 10 "Ellesmere."

A new picture of NVIDIA GP104 "Pascal" ASIC surrounded with GDDR5X memory chips hints at the possibility of NVIDIA reserving the new fast memory standard for the GTX 1080, and older GDDR5 for the more affordable GTX 1070. The picture reveals a GP104 chip with the ASIC code "GP104-400-A1," surrounded by eight Micron-made GDDR5X memory chips. We know from an older article that this ASIC code denotes the top-tier GTX 1080. A second picture (recently posted) reveals a "GP104-200-A1" ASIC surrounded by conventional GDDR5 memory chips. This ASIC corresponds to the second-fastest GTX 1070.

GDDR5 and GDDR5X are nearly identical electrically, and it's quite conceivable that the GP104 chip features a memory controller that supports both standards. GDDR5 can be had at speeds of up to 8 Gbps, while GDDR5X chips can range between 10 Gbps thru 12 Gbps initially, with 14 Gbps chips planned for a little later. Besides memory, CUDA core count could be another factor that sets the two SKUs apart. NVIDIA is planning to launch a total of three SKUs based on the GP104 silicon, in June 2016, beginning with the GTX 1080 and GTX 1070 in early-June (probably along the sidelines of Computex 2016), and a third SKU in mid-June.

AMD finalized the GPU architecture roadmap running between 2016 and 2018. The company first detailed this at its Capsaicin Event in mid-March 2016. It sees the company's upcoming "Polaris" architecture, while making major architectural leaps over the current-generation, such as a 2.5-times performance/Watt uplift and driving the company's first 14 nanometer GPUs; being limited in its high-end graphics space presence. Polaris is rumored to drive graphics for Sony's upcoming 4K Ultra HD PlayStation, and as discrete GPUs, it will feature in only two chips - Polaris 10 "Ellesmere" and Polaris 11 "Baffin."

"Polaris" introduces several new features, such as HVEC (h.265) decode and encode hardware-acceleration, new display output standards such as DisplayPort 1.3 and HDMI 2.0; however, since neither Polaris 10 nor Polaris 11 are really "big" enthusiast chips that succeed the current "Fiji" silicon, will likely make do with current GDDR5/GDDR5X memory standards. That's not to say that Polaris 10 won't disrupt current performance-thru-enthusiast lineups, or even have the chops to take on NVIDIA's GP104. First-generation HBM limits the total memory amount to 4 GB over a 4096-bit path. Enthusiasts will have to wait until early-2017 for the introduction of the big-chip that succeeds "Fiji," which will not only leverage HBM2 to serve up vast amounts of super-fast memory; but also feature a slight architectural uplift. 2018 will see the introduction of its successor, codenamed "Navi," which features an even faster memory interface.

Sony's upcoming 4K Ultra HD game console, which its fans are referring to as the "PlayStation 4K," while being internally referred to by Sony as "NEO," could feature a very powerful GPU. AMD could custom-design the SoC that drives the console, to feature an 8-core 64-bit x86 CPU based on the "Jaguar" micro-architecture, running at 2.10 GHz; and a GPU component featuring 36 compute units based on "next-generation Graphics CoreNext" architecture.

36 next-gen GCN compute units sounds an awful lot like the specs of the Polaris10 "Ellesmere" chip in its Radeon R9 480 configuration, working out to a stream processor count of 2,304 - double that of the 1,152 on the current-gen PlayStation 4. The SoC is also rumored to feature a 256-bit GDDR5 memory interface holding 8 GB of memory. This memory will be used as both system and video memory, just like on the current-gen PlayStation 4. The memory bandwidth will be increased to 218 GB/s from the current 176 GB/s. Besides 4K Ultra HD gaming, this chip could also prepare Sony for VR headsets, leveraging AMD's LiquidVR tech.

Computex 2016 could see some major consumer graphics action, with AMD reportedly planning to launch two mid-thru-performance segment products on the sidelines of the event - the Radeon R7 470, based on the 14 nm "Baffin" (Polaris 11) silicon, and the Radeon R9 480, based on the 14 nm "Ellesmere" (Polaris 10) silicon. The R7 470 could succeed the R7 370 series in not just performance, but also offer a leap in energy efficiency, with a TDP of less than 50W. The R9 480, on the other hand, could feature a TDP of just 110-135W (R9 380 is rated at 190W).

The R9 480, based on the "Ellesmere" (Polaris 10) is shaping up to be a particularly interesting silicon. It's rumored to feature 2,304 stream processors based on the 4th generation Graphics CoreNext architecture, with 2,560 stream processors being physically present on the chip; and a 256-bit wide GDDR5 (GDDR5X-ready) memory controller. 8 GB could be the standard memory amount. AMD could keep the clock speeds relatively low, with 800-1050 MHz GPU clocks. Imagine R9 390-like performance at half its power-draw.

A picture of NVIDIA's next performance-segment GPU based on the upcoming "Pascal" architecture, the GP104, was leaked to the web, revealing a heap of raw material to speculate from. To begin with, GP104 retains the traditional component layout of a simple GPU die sitting on a conventional fiberglass substrate package, with memory chips surrounding it. NVIDIA is reserving exotic specs such as stacked HBM2 memory for the high-end GP100 silicon.

Some fairly straightforward trignometry reveals that the rectangular die of the GP104 measures 15.35 mm x 19.18 mm, with one source speculating a transistor-count of 7.4-7.9 billion. The card is expected to feature 8 gigabit GDDR5 memory chips, which tick at 8 GHz (GDDR5-effective). If the memory bus width is 256-bit, then you're looking at a memory bandwidth of 256 GB/s. The CUDA core count of the GP104 could be closer to 2,560, than the 4,096 from an older report.

At the NVIDIA Drive PX compute module unveiling, company CEO Jen-Hsun Huang gave us the first glimpse of a mid-range GPU based on the "Pascal" architecture. This chip looks a lot more conventional than the fancy HBM2-infused multi-chip module that's at the heart of the Tesla P100. Its package is a traditional green fiberglass substrate with a rectangular die at the center; and is surrounded by conventional-looking GDDR5-class memory chips (which could very well be GDDR5X). The Drive PX is a GPU-accelerated deep-learning box that NVIDIA is basing much of its self-driving car tech around; and uses a pair of these mid-range "Pascal" MXM boards.

NVIDIA announced the Quadro M6000, its new high-end workstation single-GPU graphics card. Based on the GM200 silicon, and leveraging the "Maxwell" GPU architecture, the M6000 maxes out all the hardware features of the chip, featuring 3,072 CUDA cores, 192 TMUs, 96 ROPs, and a 384-bit wide GDDR5 memory interface, holding 24 GB of memory, double that of the GeForce GTX TITAN X. Its peak single-precision floating point performance is rated at 7 TFLOP/s.

Where the M6000 differs from its the GTX TITAN X is its workstation-grade features. It drops the HDMI 2.0 connector for a total of four DisplayPort 1.2 connectors, supporting a total of four 4K Ultra HD displays. The dual-link DVI connector stays on. There's also an optional stereoscopic 3D connector. The nView MultiDisplay tech provides more flexible display-head configurations than the ones you find on NVIDIA's consumer graphics GPUs; you also get NVIDIA GPUDirect support, which gives better memory sharing access for multi-GPU systems. The M6000 supports most modern 3D APIs, such as DirectX 12, OpenGL 4.5, and Vulkan; with compute capabilities over CUDA, OpenCL, and DirectCompute. NVIDIA didn't reveal pricing.