EK, has partnered up with XFX to bring you a factory water-cooled Radeon RX 6900 XT GPU. The XFX Speedster ZERO Radeon RX 6900XT RGB EKWB is one of the fastest AMD Radeon-based graphics cards on the market. This new Speedster series GPU is equipped with a 14-phase VRM power delivery system, consisting of DrMOS and high polymer capacitors. To make sure these lightning-fast graphics processors manage to hit their maximum clocks, a unique EK water block is pre-installed which also brings a prolonged lifespan due to the superior thermals that the liquid cooling provides. This also makes sure that no precious gaming time is spent on the water block mounting, and there are no questions regarding the warranty.

A powerful 14+2 Phase Power Design allows more stable performance by better distributing power across more power phases in the VRM which results in more overclocking and boosting headroom. Couple that with the incredibly cool components due to the full cover EK water block and you get a recipe for high performance, stability, and a long lifespan.

Earlier this week, we learned about AMD making several additions to its Ryzen 5000 Socket AM4 desktop processor lineup, to better compete against the bulk of the 12th Gen Intel Core "Alder Lake" processors. It turns out that there are three more additions to the lineup that we missed, because they're slated for a slightly later availability from the other chips (later by weeks).

The first of these three is the Ryzen 7 5700 (non-X). This chip is uniquely different from the Ryzen 7 5700X and the Ryzen 7 5600G. It is an 8-core/16-thread processor that's based on the 7 nm "Cezanne" silicon, with its iGPU disabled. This means you still get eight "Zen 3" CPU cores, but no iGPU, just 16 MB of L3 cache, and the PCI-Express interface of the chip is limited Gen 3. The Ryzen 3 5100 is the spiritual successor to the very interesting Ryzen 3 3100. It is a 4-core/8-thread processor based on the same "Cezanne" silicon with "Zen 3" cores, but with only 8 MB of L3 cache, and the iGPU remaining disabled. The third chip on the anvil is the Ryzen 7 4700, an interesting 8-core/16-thread offering based on the older "Renoir" silicon with "Zen 2" CPU cores.

Less than 24 hours after Apple's launch event, the first Geekbench numbers for the new Apple M1 Ultra CPU are out and the numbers are interesting to say the least. For starters, the system the Geekbench numbers are from, is the top of the range 20 Core SKU with 128 GB of RAM. This helps us get some additional insight into Apple's new CPUs. As Apple didn't provide much in technical terms yesterday, nor on its website, we now know that the clock speed of the M1 Ultra is the same 3.2 GHz as the regular M1. It also appears that the CPU cache remains the same, even though Geekbench is only listing the cache of the efficiency cores for some reason.

Although Geekbench isn't a reliable cross-platform benchmark, we do at least get an idea of how the new SoC from Apple performs. The single core performance is more or less on par with the Apple M1 Max, but loses out quite easily to Intel's Alder Lake processors. However, once we move to the multi-threaded test, the M1 Ultra really shows what it's capable of. Surprisingly the performance scaling is almost linear with the double of performance CPU cores compared to the M1 Max, which suggests that Apple's multi-chip module design is extremely capable. The interesting thing will be to see how well this design scales for GPU intensive applications. Stepping outside of the Apple ecosystem, the M1 Ultra ends up somewhere around an AMD Ryzen Threadripper 3970X in terms of multi-core performance. Scaling over some of the detailed tests aren't somewhere between 80-90 percent depending on the particular test compared to the M1 Max, if we compare to the faster results on Geekbench, which is still quite impressive considering we're looking at two M1 Max CPUs that are technically glued together.

MSI Is ready with the monstrous WS WRX80 motherboard for AMD Ryzen Threadripper PRO workstation processors. Slotted in the company's MSI PRO lineup, the board is based on the AMD WRX80 chipset, features the sWRX8 CPU socket, and comes with out of the box support for Ryzen Threadripper PRO 5000WX processors based on the "Zen 3" architecture, as well as previous-generation Threadripper 3000WX "Zen 2" processors. Built in the E-ATX form-factor, the board draws power from a 24-pin ATX, two 8-pin EPS, as well as two 6-pin PCIe power. It uses an expensive 14-layer PCB, and the CPU VRM solution is made up of 11 phases, using 105 A power-stages.

The sWRX8 socket is flanked by eight DDR4, each with a dedicated 1DPC path to the processor's 8-channel memory interface. There are seven PCI-Express 4.0 x16 slot, all wired to the processor and running at full x16 bandwidth. There could be as many as four M.2 NVMe slots with PCI-Express 4.0 x4 wiring, with an included M.2 to PCIe card with four additional M.2 slots. Other storage options include two U.2 ports, and eight SATA 6 Gbps. Networking options include 10 GbE, 2.5 GbE, and Wi-Fi 6E with Bluetooth 5.2. There's also an ASpeed AST2600 BMC IPMI remote-management chip with its dedicated 1 GbE management port. USB connectivity includes 20 Gbps USB 3.2 Gen 2x2 ports. We'll know more about this board as Threadripper PRO 5000WX nears retail availability. Right now it's exclusive to the Lenovo ThinkCenter P620 pre-built workstation.

TechPowerUp has verified a rumour posted over on VideoCardz that is quite puzzling, as AMD has asked motherboard makers to remove support for overclocking in the UEFI/BIOS for the Ryzen 7 5800X3D. When we asked for a reason as to why this was the case, we were told that AMD was keeping that information to themselves for the time being. The details provided by AMD are short and to the point "5800X3D 8C16T 100-xxxxxxxxx 105 W AGESA: PI 1206b 1/28 Please hide Vermeer-X CPU OC BIOS SETUP options".

The information suggests that this happened back at the end of January, although it's no surprise that this information took some time to leak, as it's not the kind of information that would normally make its way outside of the motherboard manufacturers. AGESA 1.2.0.6 B is also the most current release for a wide range of motherboards, even though it doesn't seem to be offered as a final release from all of the board makers just yet. It's unclear why AMD has done this, but it suggests that there might be some issues related to the 3D V-Cache and overclocking.

Intel is reportedly targeting early-Summer (May-June) for the launch of its ambitious attempt at AAA gaming graphics cards for desktops, the Arc "Alchemist" series, based on the Xe-HPG graphics architecture, according to a report by Igor's Lab. Product launches are expected anywhere between May 2 and June 1, so one could expect some market availability within Summer. The Arc "Alchemist" series is designed to be sold through a handful board partners Intel already has strong industry relations with. The Arc "Alchemist" lineup will initially target four market segments, including the performance segment, meant for maxed out AAA gaming, with XeSS possibly even enabling 4K Ultra HD gameplay. Intel's entry to the gaming graphics space is expected to introduce an element of competitive pressure against both NVIDIA and AMD, as the company has the financial muscle to keep investing in this market if it tastes success with "Alchemist."

AMD today announced availability of the new AMD Radeon PRO W6600X GPU for Mac Pro, developed to help professional users push the limits of what is possible. Built on the award-winning AMD RDNA 2 architecture, AMD Infinity Cache and other advanced technologies, the new GPU delivers stunning visuals and exceptional performance to power a variety of today's popular professional applications and workloads.

AMD Radeon PRO W6000X Series GPUs provide several graphics options for Mac Pro, which is engineered for extreme performance, expandability and configurability. The new AMD Radeon PRO W6600X GPU delivers an outstanding combination of image quality and exceptional performance, helping enable Mac Pro users to achieve amazing levels of productivity and creativity. Users can also select from several other powerful AMD GPUs to power an extensive range of professional workloads, including the previously announced AMD Radeon PRO W6900X, AMD Radeon PRO W6800X and AMD Radeon PRO W6800X Duo GPUs.

AMD Ryzen 7 5800X3D, the company's ambitious new 8-core/16-thread Socket AM4 processor that claims to match the Core i9-12900K "Alder Lake" in gaming performance, will launch at an MSRP of USD $449, according to prices of several upcoming AMD Ryzen processors leaked to the web. The 5800X3D is clocked lower than the 5800X, with 3.40 GHz base and 4.50 GHz boost frequencies, but the large 96 MB L3 cache from the 3D Vertical Cache memory, overcomes this.

The Ryzen 7 5700X is an interesting new 8-core/16-thread part. It's based on the "Vermeer" MCM just like the 5800X, and unlike the 5700G that's based on the "Cezanne" monolithic silicon. The 5700X is clocked at 3.40 GHz base, with 4.60 GHz boost, compared to the 3.80 GHz base and 4.70 GHz boost frequency of the 5800X. The Ryzen 7 5700X is launching at $299 MSRP, which implies that the company is cutting the MSRP of the Ryzen 5 5600X that originally occupied this price-point.Update Mar 9th: Correction: the Ryzen 5 5500 is a 6-core/12-thread part.

CORSAIR, a world leader in enthusiast components for gamers, creators, and PC builders, today announced a powerful new tool to help gamers design their next PC or upgrade, tailor-made for their needs: CORSAIR PC Builder. Combining a vast PC part compatibility database with expert hardware insights and building tips, CORSAIR PC Builder is an invaluable resource when researching and configuring a new PC build.

CORSAIR PC Builder curates your components checklist to your PC gaming needs, drawing on a database covering the full specs from thousands of currently sold PC components. Users need only to specify their chosen Intel or AMD processor, graphics card, and motherboard, and CORSAIR PC Builder will offer a comprehensive parts list of award-winning CORSAIR components, all guaranteed for compatibility with your desired system. PC Builder also ensures that all the parts fit inside your chosen case, whether it be a CORSAIR case or third-party. The result is a carefully curated shopping cart, filled with exactly what you need to build the optimal PC to run your favorite games.

AMD today launched its first Ryzen Threadripper processors based on the "Zen 3" microarchitecture, with the Ryzen Threadripper PRO 5000WX series. Designed to be drop-in compatible with workstations and motherboards based on the AMD WRX80 chipset, these processors come in core-counts of up to 64-core/128-thread, with an enormous I/O offering that includes 8-channel DDR4 memory with ECC support, and a 128-lane PCI-Express 4.0 root complex. The biggest change over the previous generation Threadripper PRO 3000WX series has to be the use of "Zen 3" CCDs, each with 8 CPU cores, sharing a common 32 MB of L3 cache. AMD isn't using the "Zen 3" chiplets with 3DV Cache.

The full AMD PRO management feature-set from Ryzen PRO is available on these processors, including PRO Security, PRO Management, and a special support channel that includes planned parts and software availability. What's more, AMD has been working with ISVs of most professional content-creation software since the past generation of Ryzen Threadripper PRO, to optimize their software for the processors (high core-counts, NUMA topology, etc.). The benefits of these are shared with all generations of Threadrippers. Although all parts in the Threadripper PRO 5000WX series are rated for a TDP of 280 W, AMD claims to have worked on power-management, offering up to 67 percent lower power per core, compared to the competition (2P Xeon Scalable Platinum 8280).

According to yet another leak, it would appear that AMD is planning to release no less than four new CPUs in its Ryzen 5000-series this month, with the obvious headline product being the already announced Ryzen 7 5800X3D. However, details of a further three CPUs have turned up on Twitter and it looks like AMD is planning to go head to head with Intel, if the rumoured price brackets are indeed correct. The expected three new CPUs are the Ryzen 7 5700X, Ryzen 5 5600 and Ryzen 5 5500.

The Ryzen 7 5700X is as expected an eight core, 16 thread CPU that is said to be cheaper than an Intel Core i5-12600KF, which means an MSRP around the US$250-270 mark. The six core, twelve thread Ryzen 5 5600 on the other hand, is said to be cheaper than the Core i5-12400, so it should get a sub US$200 MSRP. Finally the six core, six thread Ryzen 5 5500, is said to land at the same price point as the Core i3-12100, pointing at a US$130 MSRP. Unfortunately, no indication of pricing for the Ryzen 7 5800X3D was given, but based on the fact that AMD seems to be dropping the pricing of its current Ryzen 5000-series of processors in the US market, it'll hopefully get a competitive price point.

The rumor mill is attempting to nail down the specific launch date of AMD's purported RDNA 2 series refresh. According to renowned leaker Enthusiastic Citizen over at Chiphell forums, AMD is now planning to launch updated versions of its RDNA2 graphics cards around April 20th or April 21st. It seems that AMD is updating three different SKUs based on the RDNA2 silicon, perhaps in order to increase their market attractiveness (and competitiveness) against both NVIDIA's lineup and Intel's upcoming Arc Alchemist series, which is expected to launch in the next several weeks as well.

The new cards, which are expected to carry updated model names, are currently expected to be the RX 6950XT (a response to NVIDIA's oft-delayed RTX 3090 Ti graphics card), the RX 6750XT (likely meant to compete against Intel's upcoming Arc Alchemist A700 series, and the RX 6650XT. The only available details purport to the RX 6950XT, which is expected to not only carry upgraded 18 Gbps GDDR6 VRAM, but also an increased power limit of 350 W (above the 300 W from the reference RX 6900XT. The other two GPU updates should follow suit along the memory frequency and power consumption increases.

Does it take ages for the taskbar calendar and notification center to load on your Windows 11 PC powered by an AMD Ryzen processor? Notice random stutters in performance? Chances are, the lag is caused not due to user-interface bugs by Microsoft, but hardware. AMD discovered that certain Ryzen-powered Windows 11 and Windows 10 PCs experience intermittent performance stutters when running with fTPM (firmware TPM) enabled.

The performance stutter is caused due to background memory transactions between Windows and the fTPM, to authenticate an action, as the fTPM serves the function of a hardware root of trust. Since the fTPM is part of the UEFI firmware that resides on the SPI flash EEPROM chip, the performance stutter is caused due to fTPM-related memory transactions with this chip.

Intel and AMD will be releasing new CPUs that support DDR5 DRAM solutions for PCs and servers this year. In response, the DRAM industry led by South Korean suppliers is developing solutions to complement the arrival of the new CPUs. In the midst of the gradual shift to DDR5, DRAM suppliers will also scale back the supply of DDR3 solutions, according to TrendForce's latest investigations. With Korean suppliers accelerating their withdrawal from DDR3 production, Taiwanese suppliers yet to kick off mass production using newly installed capacities, and Chinese suppliers falling short of their expected yield rate, the global supply of DDR3 solutions will undergo an impending decline. With respect to the demand side, however, not only has the supply of networking chips been ramping up, but material shortage issues are also gradually easing. As such, buyers are now procuring DDR3 solutions ahead of time, resulting in a tight supply and demand situation in the DDR3 market. TrendForce therefore expects DDR3 DRAM prices to recover from a bearish first quarter and undergo a 0-5% QoQ increase in 2Q22.

There is a significant cooling down of graphics card prices across the board, according to an exhaustive set of retail pricing data compiled by 3DCenter.org, for the European market. We see entry-level cards like the Radeon RX 6500 XT get closest to their MSRP, with prices as low as 235€, and the GeForce RTX 3050 as low as 349€. The RX 6600 XT sees its price as low as 519€, and the popular RTX 3060 around the same price, starting at 529€. The performance segment sees the 768€ RX 6700 XT square off against the 869€ RTX 3070. The high-end sees the RTX 3080 10 GB as low as 1,179€, compared to the RX 6800 XT, as low as 1099€. 3DCenter observes a downward trend for these graphics cards across the board.

MSI released a new gaming desktop, the MAG Trident S 5M, a machine dedicated to cloud gaming and mobile gaming. It is different from traditional gaming desktops and focuses more on helping gamers enjoy games in the living room. Gamers only need to activate the Android system emulator (MSI APP Player) and the 30-day free XBOX GAME PASS ULTIMATE through the Game Stadium software exclusively developed by MSI to control mobile games on the sofa with only one joystick. You can easily access cloud games and other entertainment software.

The exclusive MSI APP Player allows players to emulate the Android system in the Windows system. Gamers can experience mobile games on a larger screen with better performance. They can also use the Multi-Instance Manager function to open different games in different windows. Certain games can also enable joystick mode, allowing players to use the same joystick to play. In STEAM, you can play games through the built-in Big Screen Mode. In Xbox Game Pass Ultimate, there is the same operation mode to let you use the joystick directly so that it can be perfectly applied to various game software.

AMD's Threadripper Pro WX or workstation processors are set to finally join the 5000-series of AMD CPU's if the latest leaks are to be believed. It would seem that AMD is planning no less than five new SKUs, or one more than the current 3000-series of workstation chips. The new entry is a 24 core, 48 thread chip, which was lacking from the current range. As such, the entire series will consist of 12, 16, 24, 32 and 64 core options, all with twice as many threads. All chips have a common 4.5 GHz peak turbo clock, but the base clocks vary by quite a bit, depending on the core count.

The 12 core 5945WX has a base clock of 4.1 GHz, with the 16 core 5955WX coming in at 4 GHz, the 24 core 5965WX then drops to 3.8 GHz, followed by the 32 core 5975WX at 3.6 GHz. Finally the 64 core 5995WX is said to only muster a base clock of 2.7 GHz. All five CPUs have a TDP rating of 280 Watts. The new Threadripper chips are expected to work in current boards that sport a WRX80 socket. As such, PCIe lanes and memory support is expected to remain the same as for the 3000-series.

Unless you have been living under a rock, chances are you are following the news about the Russia-Ukraine war disputes. Not to get too political, we are here to report about your favorite rivals—AMD and Intel—officially stopping the delivery of processors to Russia and Belarus. Firstly, an AMD representative told PCWorld that "Based on sanctions placed on Russia by the United States and other nations, at this time AMD is suspending its sales and distribution of our products into Russia and Belarus. It is all AMD products and products we power (PCs, etc) in Russia and Belarus." Additionally, Intel posted an official quote, which you can find below.

IntelIntel condemns the invasion of Ukraine by Russia and we have suspended all shipments to customers in both Russia and Belarus. Our thoughts are with everyone who has been impacted by this war, including the people of Ukraine and the surrounding countries and all those around the world with family, friends and loved ones in the region.

"We are working to support all of our employees through this difficult situation, especially those with close ties to this region. We have launched an employee donation and matching campaign through the Intel Foundation that has already raised over $1.2 million for relief efforts, and we are proud of the work our teams in surrounding areas including Poland, Germany and Romania are doing to aid refugees. We will continue to stand with the people of Ukraine and the global community in calling for an immediate end to this war and a swift return to peace.

The Radeon 680M integrated graphics powering the AMD Ryzen 6000-series mobile processors is proving to be an entry-level discrete-GPU killer. TechEpiphany posted a video presentation showing the iGPU's real-world gameplay performance with the AAA title "Cyberpunk 2077" at Full HD (1080p), with a little help from FidelityFX Super Resolution (FSR). TechEpiphany used an ASUS TUF GAMING F17 notebook powered by an AMD Ryzen 7 6800H "Rembrandt" processor that has the full Radeon 680M iGPU unlocked, with all its 12 RDNA2 compute units (768 stream processors), 12 Ray Accelerators, 16 ROPs, and 48 TMUs, enabled. The notebook also features a GeForce RTX 3050 Laptop GPU, but for this testing, it was disabled.

The first part of the video shows the game running at 1080p and Medium-High settings, with FSR set at Ultra Quality. Here, the iGPU is managing 30-40 FPS. Real-time ray tracing is disabled. In the second part, they enabled ray tracing and FidelityFX Super Resolution, but this is where the iGPU runs out of steam. Frame-rates drop to unplayable levels, but there still aren't any noticeable visual artifacts or rendering errors typically associated with iGPUs made to render games above their pay-grade. It's still impressive to see that AMD following through on its promise of bringing 1080p gaming across a broader range of titles. The TechEpiphany video presentation can be watched in the source link below.

Intel, along with Advanced Semiconductor Engineering Inc. (ASE), AMD, Arm, Google Cloud, Meta, Microsoft Corp., Qualcomm Inc., Samsung and Taiwan Semiconductor Manufacturing Co., have announced the establishment of an industry consortium to promote an open die-to-die interconnect standard called Universal Chiplet Interconnect Express (UCIe). Building on its work on the open Advanced Interface Bus (AIB), Intel developed the UCIe standard and donated it to the group of founding members as an open specification that defines the interconnect between chiplets within a package, enabling an open chiplet ecosystem and ubiquitous interconnect at the package level.

"Integrating multiple chiplets in a package to deliver product innovation across market segments is the future of the semiconductor industry and a pillar of Intel's IDM 2.0 strategy," said Sandra Rivera, executive vice president and general manager of the Datacenter and Artificial Intelligence Group at Intel. "Critical to this future is an open chiplet ecosystem with key industry partners working together under the UCIe Consortium toward a common goal of transforming the way the industry delivers new products and continues to deliver on the promise of Moore's Law."

It will be an unexpectedly busy March for AMD, with the company launching three distinct products across its processor lines. The first one, which we reported earlier this morning, speaks of a late-March availability of the Ryzen 7 5800X3D 8-core/16-thread Socket AM4 processor, which AMD claims offers gaming performance on par with the Core i9-12900K "Alder Lake." It turns out, there are two more surprises.

Apparently the company is ready with Ryzen Threadripper PRO 5000 series workstation processors. Designed for Socket sWRX8 motherboards based on the only chipset option available—the AMD WRX80, these are the first Threadripper products based on the "Zen 3" microarchitecture, and feature 8-channel DDR4 memory, and up to 128 PCI-Express Gen4 lanes for workstation connectivity. Unfortunately, you can't buy one of these in the retail channel, as AMD is making them OEM-only. The first pre-built workstations will arrive as early as next week (March 8). At this point we still don't know if these chips use the newer "Zen 3" CCD with 3D Vertical Cache, or the conventional "Zen 3" CCD with 32 MB planar L3 cache.

Based on details posted by the Commercial Times in Taiwan, it would appear that Samsung Foundry is working on ways to win over customers from TSMC. The backstory is that fabless chip makers have had to change foundries over time, due to the fact that from a foundry side, it's expensive to be on the cutting edge and many foundries have stepped out of the race. This has left the fabless chip makers with fewer options and with TSMC as the de facto industry leader today. That said, as we've seen, NVIDIA and Qualcomm picked Samsung Foundry for some of their latest chips, but based on industry rumours, the yields aren't great.

Samsung Foundry has been trying to win over customers by offering attractive pricing compared to TSMC, but far from everyone has been interested. Some companies, like Apple, have tried to work with Samsung, but abandoned them in the end in favour of TSMC. Samsung is said to be courting both AMD and NVIDIA, as well as others, although only time will tell if we'll see any future product from either company being manufactured by Samsung Foundry. The fact that there are investigations into the yields at Samsung Foundry should be a concern for any future customer and isn't a good sign of where the company is heading.

With motherboard vendors posting UEFI firmware updates to support it, we were wondering where the actually processor is. It turns out, that the new AMD Ryzen 7 5800X3D started shipping from factories. Greymon55, a reliable source with hardware leaks, states that 5800X3D is shipped, and the first retail batch of these processors could reach markets by the end of the month (the end of March, 2022).

The 5800X3D is the first client processor to feature a 3D stacked-die design. AMD is introducing the new 3D Vertical Cache, a 64 MB addition to the processor's 32 MB on-die cache, which works contiguously as a 96 MB L3 cache that each of the processor's eight "Zen 3" cores can access. Apparently "Zen 3" with 3D Vertical Cache has major performance gains, with AMD claiming gaming performance parity with the Core i9-12900K.

Pricing of AMD's Ryzen 9 5950X 16-core/32-thread "Zen 3" processor dropped to a surprising all-time-low of $600 on Newegg, down from its original $800, a 25% drop in price. This puts the 5950X more or less on par with the pricing of the Intel Core i9-12900K (±$20 variance). The 12-core/24-thread 5900X is going for $450, which is still significantly higher than the $380 that the Core i7-12700K commands.

The 5950X continues to lead the i9-12900K in various heavily multi-threaded productivity tasks, although it has lost the gaming performance edge to the new Intel chip. AMD is attempting to remedy this with the Ryzen 7 5800X3D processor that the company claims offers gaming performance parity, but will fall behind in multi-threaded productivity on account of fewer cores. If you're planning to pick a 5950X from Newegg for a build-from-scratch, don't forget to check out combo deals with motherboards, where you get to save a further $30-odd.

The recent NVIDIA data-leak springs up information on various upcoming graphics parts. Besides "Ada Lovelace," "Hopper," we come across a new codename, "Blackwell." It turns out that NVIDIA is splitting the the graphics and compute architecture naming with the next generation, not unlike what AMD did, with its RDNA and CDNA series. The current "Ampere" architecture is being used both for compute and graphics, with the streaming multiprocessor for the two being slightly different—the compute "Ampere" has more FP64 and Tensor components, while the graphics "Ampere" does away with these in favor of RT cores and graphics-relevant components.

The graphics architecture to succeed GeForce "Ampere" will be GeForce "Ada Lovelace." GPUs in this series are identified in the leaked code as "AD102," "AD103," "AD104," "AD106," "AD107," and "AD10B," succeeding a similar numbering for parts with the "A" (GeForce Ampere) series. The compute architecture succeeding "Ampere" will be codenamed "Hopper." with parts in the series being codenamed "GH100" and "GH202." Another compute or datacenter architecture is "Blackwell," with parts being codenamed "GB100" and "GB102." From all accounts, NVIDIA is planning to launch the GeForce 40-series "Ada" graphics card lineup in the second half of 2022. The company is in need of a similar refresh for its compute product lineup, and could debut "Hopper" either toward the end of 2022 or next year. "Blackwell" could follow "Hopper."