Sapphire formally launched its Radeon 6700 series graphics card. The AMD Radeon 6700 is an odd-ball SKU that doesn't yet feature in the company's retail product stack, but is yet being released to retail by Sapphire. So far we've not come across any other board partner with this SKU. The 6700 is unique in its branding—there's neither "RX" nor "XT" in the model name, it's called simply the "Radeon 6700."

Carved out from the same 7 nm "Navi 22" silicon as the RX 6700 XT and RX 6750 XT; the 6700 has 36 out of 40 compute units enabled, working out to 2,304 stream processors, and 144 TMUs. The card is endowed with 10 GB of 16 Gbps GDDR6 memory across a 160-bit wide memory interface. Sapphire has two cards in its lineup, one is an unnamed base model that sticks to the "reference" specs, and a factory-overclocked Pulse 6700 card.

Today at the Intel Vision conference, Habana Labs, an Intel company, announced its second-generation deep learning processors, the Habana Gaudi 2 Training and Habana Greco Inference processors. The processors are purpose-built for AI deep learning applications, implemented in 7nm technology and build upon Habana's high-efficiency architecture to provide customers with higher-performance model training and inferencing for computer vision and natural language applications in the data center. At Intel Vision, Habana Labs revealed Gaudi2's training throughput performance for the ResNet-50 computer vision model and the BERT natural language processing model delivers twice the training throughput over the Nvidia A100-80GB GPU.

"The launch of Habana's new deep learning processors is a prime example of Intel executing on its AI strategy to give customers a wide array of solution choices - from cloud to edge - addressing the growing number and complex nature of AI workloads. Gaudi2 can help Intel customers train increasingly large and complex deep learning workloads with speed and efficiency, and we're anticipating the inference efficiencies that Greco will bring."—Sandra Rivera, Intel executive vice president and general manager of the Datacenter and AI Group

AMD today announced the Ryzen 5000C line of mobile processors for Chromebooks. This is the company's second generation of Chromebook-specific processors after the Ryzen 3000C series based on the original "Zen" microarchitecture. The 5000C series chips are based on "Zen 3," with CPU core counts of up to 8-core, and hence present a big leap in performance over the 3000C series, along with a complete suite of the latest connectivity, display technology, and security and management features specific to Chrome OS.

The Ryzen 5000C series is based on the 7 nm "Cezanne" monolithic silicon. The chip physically features an 8-core/16-thread CPU based on the "Zen 3" microarchitecture, with 16 MB of shared L3 cache; an iGPU based on the Vega graphics architecture, with 8 compute units (512 stream processors), a dual-channel DDR4 or LPDDR4/x memory interface, and unlike the conventional Ryzen 5000-series mobile processors, these chips come with a special microcode to match the security and management features of Chrome OS. AMD also supplies Chromebook vendors with timely driver updates for the various components on these chips.

A leaked specifications sheet of the upcoming PowerColor Radeon RX 6650 XT Hellhound custom-design graphics card, seen by VideoCards, sheds light on AMD's play at carving out the RX 6650 XT. It involves dialing up the engine clocks (GPU clock speed), and memory bandwidth. At this point it is not known if the RX 6650 XT is based on a refined variant of the "Navi 23" silicon, possibly leveraging the TSMC N6 (6 nm) process, or if it's just a case of AMD dialing up clock speeds while pushing up the typical board power, on existing 7 nm (TSMC N7) process.

The RX 6650 XT Hellhound comes with about 4.3% increase in game clocks in its default "OC mode" BIOS, and about 3.7% increase in maximum boost clocks, up from 2593 MHz to 2689 MHz. The "Silent mode" BIOS of the RX 6650 XT Hellhound offers better clock speeds than the "OC mode" BIOS of the RX 6600 XT Hellhound, at 2410 MHz game, 2635 MHz boost, compared to 2382 MHz game, 2593 MHz boost. The other big surprise is memory clocks, with AMD possibly using 17.5 Gbps GDDR6 memory speeds, compared to 16 Gbps on the RX 6600 XT. This results in a 9.4% increase in memory bandwidth. The RX 6600 XT Hellhound uses a single 8-pin PCIe power connector, for an input capacity of 225 W (including the PCIe slot power), which is sufficient for the card's 160 W typical board power. The TBP of the RX 6650 XT Hellhound is not known, but given that its specs sheet still shows single 8-pin, it has to be under 225 W.

Way back in January 2021, we heard a spectacular rumor about "Navi 31," the next-generation big GPU by AMD, being the company's first logic-MCM GPU (a GPU with more than one logic die). The company has a legacy of MCM GPUs, but those have been a single logic die surrounded by memory stacks. The RDNA3 graphics architecture that the "Navi 31" is based on, sees AMD fragment the logic die into smaller chiplets, with the goal of ensuring that only those specific components that benefit from the TSMC N5 node (6 nm), such as the number crunching machinery, are built on the node, while ancillary components, such as memory controllers, display controllers, or even media accelerators, are confined to chiplets built on an older node, such as the TSMC N6 (6 nm). AMD had taken this approach with its EPYC and Ryzen processors, where the chiplets with the CPU cores got the better node, and the other logic components got an older one.

Greymon55 predicts an interesting division of labor on the "Navi 31" MCM. Apparently, the number-crunching machinery is spread across two GCD (Graphics Complex Dies?). These dies pack the Shader Engines with their RDNA3 compute units (CU), Command Processor, Geometry Processor, Asynchronous Compute Engines (ACEs), Rendering Backends, etc. These are things that can benefit from the advanced 5 nm node, enabling AMD to the CUs at higher engine clocks. There's also sound logic behind building a big GPU with two such GCDs instead of a single large GCD, as smaller GPUs can be made with a single such GCD (exactly why we have two 8-core chiplets making up a 16-core Ryzen processors, and the one of these being used to create 8-core and 6-core SKUs). The smaller GCD would result in better yields per wafer, and minimize the need for separate wafer orders for a larger die (such as in the case of the Navi 21).

Here is the first picture of a next-generation AMD EPYC "Genoa" processor with its integrated heatspreader (IHS) removed. This is also possibly the first picture of a "Zen 4" CPU Complex Die (CCD). The picture reveals as many as twelve CCDs, and a large sIOD silicon. The "Zen 4" CCDs, built on the TSMC N5 (5 nm EUV) process, look visibly similar in size to the "Zen 3" CCDs built on the N7 (7 nm) process, which means the CCD's transistor count could be significantly higher, given the transistor-density gained from the 5 nm node. Besides more number-crunching machinery on the CPU core, we're hearing that AMD will increase cache sizes, particularly the dedicated L2 cache size, which is expected to be 1 MB per core, doubling from the previous generations of the "Zen" microarchitecture.

Each "Zen 4" CCD is reported to be about 8 mm² smaller in die-area than the "Zen 3" CCD, or about 10% smaller. What's interesting, though, is that the sIOD (server I/O die) is smaller in size, too, estimated to measure 397 mm², compared to the 416 mm² of the "Rome" and "Milan" sIOD. This is good reason to believe that AMD has switched over to a newer foundry process, such as the TSMC N7 (7 nm), to build the sIOD. The current-gen sIOD is built on Global Foundries 12LPP (12 nm). Supporting this theory is the fact that the "Genoa" sIOD has a 50% wider memory I/O (12-channel DDR5), 50% more IFOP ports (Infinity Fabric over package) to interconnect with the CCDs, and the mere fact that PCI-Express 5.0 and DDR5 switching fabric and SerDes (serializer/deserializers), may have higher TDP; which together compel AMD to use a smaller node such as 7 nm, for the sIOD. AMD is expected to debut the EPYC "Genoa" enterprise processors in the second half of 2022.

AMD's Radeon RX product stack refresh for Spring-Summer, is reportedly set to launch on May 10, 2022. Here's the first picture of what a reference-design RX 6950 XT flagship, RX 6750 XT, and the mid-range RX 6650 XT, could look like. These reference board designs are essentially identical to the original RX 6000 made-by-AMD (MBA) reference designs, but ditch the two-tone silver+black color-scheme for an all-black scheme with some diamond-cut edges around the fan vents, and some piano-black accents.

At this point it is not known if this refresh sees the Navi 20-series ASICs optically-shrunk to the TSMC N6 (6 nm) silicon fabrication node, or if it's the existing 7 nm ASICs with their total graphics power (TGP) values dialed up to make room for increased engine clocks, and faster 18 Gbps-rated GDDR6 memory chips. It's interesting to see the RX 6750 XT now come with a triple-fan cooler that resembles the RX 6800 (non-XT) cooler in design, if not color. We're not sure if the RX 6650 XT reference design will ever make it to the real-world, or if it's just a concept, and the SKU is an AIB-exclusive (custom-designs only).

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.

With the exponential growth of data in the world today, coupled with the shift from centralized clusters of compute and data storage to a more distributed architecture that processes data everywhere—in the cloud, at the edge, and at all points in between—Field-Programmable Gate Arrays (FPGAs) are taking on an increasingly important role in modern applications from the data center to the network to the edge. The flexibility, power efficiency, massively parallel architecture, and huge input/output (I/O) bandwidth make FPGAs attractive for accelerating a wide range of tasks from high-performance computing (HPC) to storage and networking. Many of these applications put enormous demands on memory, including capacity, bandwidth, latency and power efficiency.

To handle these high-demand applications, Intel today introduced product details for the Intel Agilex M-Series FPGAs, built on Intel 7 process technology, the industry's highest memory bandwidth FPGAs with in-package HBM DRAM. The Intel Agilex M-Series incorporates several new functional innovations and features that provide the industry with the high-speed networking, computing and memory acceleration required to meet ever-more ambitious performance and capability goals for networks, cloud and embedded edge applications.

Intel's upcoming "Meteor Lake" and "Arrow Lake" client mobile processors introduce an interesting twist to the chiplet concept. Earlier represented in vague-looking IP blocks, new artistic impressions of the chip put out by Intel shed light on a 3-die approach not unlike the Ryzen "Vermeer" MCM that has up to two CPU core dies (CCDs) talking to a cIOD (client IO die), which handles all the SoC connectivity; Intel's design has one major difference, and that's integrated graphics. Apparently, Intel's MCM uses a GPU die sitting next to the CPU core die, and the I/O (SoC) die. Intel likes to call its chiplets "tiles," and so we'll go with that.

The Graphics tile, CPU tile, and the SoC or I/O tile, are built on three different silicon fabrication process nodes based on the degree of need for the newer process node. The nodes used are Intel 4 (optically 7 nm EUV, but with characteristics of a 5 nm-class node); Intel 20A (characteristics of 2 nm), and external TSMC N3 (3 nm) node. At this point we don't know which tile gets what. From the looks of it, the CPU tile has a hybrid CPU core architecture made up of "Redwood Cove" P-cores, and "Crestmont" E-core clusters.

AMD's final refresh of the RDNA2 graphics architecture, the Radeon RX 6x50 series, could debut in June or July 2022, according to Greymon55, a reliable source with GPU leaks. The final refresh of RDNA2 could see AMD use faster 18 Gbps GDDR6 memory across the board, and eke out higher engine clocks on existing silicon IP. At this point it's not known if these new chips will be built on the same 7 nm process, or are an optical shrink to 6 nm (TSMC N6). Such a shrink to a node that offers 18% higher transistor density, would have significant payoffs with clock-speed headroom. AMD's RDNA3-based 5 nm GPUs could debut only toward the end of the year.

In related news, AMD is preparing to launch another entry-level SKU within the RX 6000 series; the Radeon RX 6500 (non-XT). Based on the same 6 nm Navi 24 silicon as the RX 6500 XT, this SKU could have a core-configuration that's in-between the RX 6500 XT and the RX 6400, in featuring 768 stream processors across 12 compute units; and 4 GB of GDDR6 memory, which is similar to the RX 6400, but with higher engine clocks. The RX 6500 is targeting a $150 (MSRP) price-point.

TSMC has held its quarterly earnings conference today and it's good news all around, at least if you're TSMC or one of its shareholders, as the company reported record profits of US$6.01 billion for the quarter, or an increase of 16.4 percent compared to the same quarter last year. At the same time, the company announced that it's going to increase its CapEx, by no less than US$40-44 billion this year, which should be compared to the US$30 billion in 2021. The company is expecting to continue to rake in money this quarter, with an expected revenue before expenditures and tax of US$16.6 to US$17.2 billion, compared to US$15.74 billion for this quarter.

Looking at the graphs provided by TSMC which shows where its revenue is coming from, its 7 nm and 5 nm nodes are now accounting for 50 percent of TSMC's revenues. The 5 nm node on its own, almost made as much money as its 16 and 28 nm nodes combined in Q4. We can also see that the 5 nm has gone from eight percent of TSMC's revenues in 2020, to 19 percent this year, with the 7 nm node dropping slightly from 33 percent to 31 percent. 2021 saw a massive 51 percent revenue growth in automotive components for TSMC compared to 2020, yet it only accounted for four percent of TSMC's total revenue for 2021. Smartphones and HPC are jointly holding 81 percent of TSMC's business based on revenue, which isn't likely to change any time soon.

XFX started selling the AMD BC-160 cryptocurrency mining card based on the AMD "Navi 12" silicon. The card is available on AliExpress for $2,000. The "Navi 12," if you recall, is an MCM mobile GPU that AMD developed exclusively for the 2019 MacBook Pro. It combined an RDNA-based GPU die with up to 16 GB of HBM2 across a 2048-bit wide interface. Built on the 7 nm node, the GPU die of "Navi 12" on the BC-160 is configured with 36 compute units (2,304 stream processors), and 8 GB of HBM2 across the full 2048-bit memory bus.

The card uses a blower-type cooling solution, and is rated with 150 W of typical board power, with a claimed 69.5 Mh/s (ETH). Drivers are provided for Linux, and mining software supported include Team Red Miner and Phoenix Miner. The card features a PCI-Express 4.0 x16 interface, its driver supports systems with up to 12 of these installed. A marketing slide sheds light on the nomenclature AMD is using for its mining cards. The "BC" in BC-160 represents "blockchain compute," the "1" stands for generation, in this case, first generation; and "60" represents hashrate-class with ETH.

AMD apparently finds itself with quite a bit of undigested 7 nm "Renoir" silicon, which it plans to repackage as Socket AM4 processors, reports VideoCardz, citing sources on ChipHell forums. The most interesting aspect of this leak is that the silicon variant, codenamed "Renoir X," comes with a disabled iGPU. This is hence a case of AMD harvesting enough "Renoir" dies with faulty iGPU components, to sell them off as desktop processors. It is also learned that these chips don't feature all of the 8 "Zen 2" CPU cores present on the silicon, but rather AMD is looking to carve out entry-level SKUs, such as the Ryzen 3 or Athlon. The company lacks Athlon desktop SKUs based on "Zen 2" or later, although traditionally the company sought to include some basic iGPU solution with its Athlon SKUs.

In related news, the source reports that AMD will refresh its Ryzen desktop processor family with the new "Vermeer S" Ryzen processors. Built on the existing Socket AM4 package, these use AMD's "Zen 3" CCDs that feature 3D Vertical Cache (3DV Cache), much like the recently announced EPYC "Milan X" server processors. AMD claimed that the 3DV Cache technology has a significant performance uplift on performance akin to a generational update. These could be the company's first response to Intel Core "Alder Lake," although since they're based on the older AM4 platform, could only feature DDR4 and PCIe Gen 4. Much like the Ryzen 3000XT series, these appear to be a stopgap product lineup, with AMD targeting late-Q2/early-Q3 for next-generation "Raphael" Socket AM5 processors based on the "Zen 4" architecture, with DDR5 and PCIe Gen 5.

With its 12th Gen Core "Alder Lake-P" mobile processors still on the horizon, Intel is already building test batches of the 14th Gen "Meteor Lake" mobile processors, at its Fab 42 facility in Chandler, Arizona. "Meteor Lake" is a multi-chip module that leverages Intel's Foveros packaging technology to combine "tiles" (purpose built dies) based on different silicon fabrication processes depending on their function and transistor-density/power requirements. It combines four distinct tiles across a single package—the compute tile, with the CPU cores; the graphics tile with the iGPU: the SoC I/O tile, which handles the processor's platform I/O; and a fourth tile, which is currently unknown. This could be a memory stack with similar functions as the HBM stacks on "Sapphire Rapids," or something entirely different.

The compute tile contains the processor's various CPU core types. The P cores are "Redwood Cove," which are two generations ahead of the current "Golden Cove." If Intel's 12-20% generational IPC uplift cadence holds, we're looking at cores with up to 30% higher IPC than "Golden Cove" (50-60% higher than "Skylake."). "Meteor Lake" also debuts Intel's next-generation E-core, codenamed "Crestmont." The compute tile is rumored to be fabricated on the Intel 4 node (optically a 7 nm-class node, but with characteristics similar to TSMC N5).

AMD is preparing an update to its compute accelerator lineup with the new MI250X. Based on the CDNA2 architecture, and built on existing 7 nm node, the MI250X will be accompanied by a more affordable variant, the MI250. According to leaks put out by ExecutableFix, the MI250X packs a whopping 110 compute units (7,040 stream processors), running at 1.70 GHz. The package features 128 GB of HBM2E memory, and a package TDP of 500 W. As for speculative performance numbers, it is expected to offer double-precision (FP64) throughput of 47.9 TFLOP/s, ditto full-precision (FP32), and 383 TFLOP/s half-precision (FP16 and BFLOAT16). AMD's MI200 "Aldebaran" family of compute accelerators are expected to square off against Intel's "Ponte Vecchio" Xe-HPC, and NVIDIA Hopper H100 accelerators in 2022.

ASRock, the leading global motherboard, graphics card and mini PC manufacturer, today launched new Challenger series products based on AMD Radeon RX 6600 GPUs. Built on the 7 nm manufacturing process, the new ASRock graphics cards offer support for the DirectX 12 Ultimate API, hardware-accelerated ray tracing, HDMI 2.1, PCI Express 4.0, and the Microsoft Windows 11 operating system. With a wealth of exclusive features, the new graphics cards are designed to provide visually stunning, high-refresh rate 1080p gaming experiences to the midrange market.

The new graphics cards are built on the breakthrough AMD RDNA 2 gaming architecture, designed to deliver the optimal balance of performance and power efficiency. Offering 32 MB of high-performance AMD Infinity Cache, 8 GB of GDDR6 memory, AMD Smart Access Memory and other advanced features, the new graphics cards are designed to bring next-generation desktop gaming experiences to the midrange market. They also support the AMD FidelityFX Super Resolution open-source spatial upscaling solution, which is designed to increase framerates while delivering high-resolution gaming experiences.

Samsung Electronics, a world leader in advanced semiconductor technology, today unveiled plans for continuous process technology migration to 3- and 2-nanometer (nm) based on the company's Gate-All-Around (GAA) transistor structure at its 5th annual Samsung Foundry Forum (SFF) 2021. With a theme of "Adding One More Dimension," the multi-day virtual event is expected to draw over 2,000 global customers and partners. At this year's event, Samsung will share its vision to bolster its leadership in the rapidly evolving foundry market by taking each respective part of foundry business to the next level: process technology, manufacturing operations, and foundry services.

"We will increase our overall production capacity and lead the most advanced technologies while taking silicon scaling a step further and continuing technological innovation by application," said Dr. Siyoung Choi, President and Head of Foundry Business at Samsung Electronics. "Amid further digitalization prompted by the COVID-19 pandemic, our customers and partners will discover the limitless potential of silicon implementation for delivering the right technology at the right time."

XFX formally launched the Radeon RX 6900 XT Speedster ZERO WB, a graphics card it teased last month. The company's new flagship product, the Speedster ZERO WB is a graphics card with a factory-fitted full-coverage water-block, for those with DIY liquid-cooling setups. It appears to be based on an all new PCB with a more tuned-up 14+2 phase VRM setup than that of the air-cooled RX 6900 XT Speedster MERC 319, which pulls power from a trio of 8-pin PCIe power connectors.

Under the hood is a 7 nm "Navi 21" XTXH silicon, which is able to sustain 10% higher engine clocks than the standard "Navi 21," and XFX claims that it has unlocked overdrive slider limit (the de facto maximum overclock), beyond 3 GHz. Out of the box, the card comes with a maximum boost frequency of 2525 MHz, compared to 2250 MHz AMD-reference. The water block came about from a collaboration with EK Water Blocks, and combines a nickel-plated copper primary material with a slightly frosted acrylic top that's studded with addressable RGB LEDs. XFX didn't reveal pricing.

According to Nikkei, TSMC is set to start building a new fab in Kaohsiung, which is Taiwan's third largest city and located in the south of the island. It's also where ASE Technology Holding is located, which is the world's largest chip packaging and testing contractor. So far, TSMC doesn't have any fabs this far south in Taiwan, but it's not without its challenges.

The new fab is said to be designed to build chips on TSMC's 6 and 7 nm nodes, which are currently their most popular nodes, although this is likely to change as their 5 nm node begins to ramp up production. That said, there will still continue to be a huge demand for 6 and 7 nm parts, as these nodes transition to become mainstream production nodes.

BIOSTAR, a leading brand of motherboards, graphics cards, and storage devices, today announced product support for the latest AMD Ryzen 5000G series Cezanne processors. AMD's next-generation Ryzen 5000G series desktop processors codenamed "Cezanne" are ready to invade the global market. The new 5000G series processors are based on Zen 3 architecture, AMD's Ryzen 5000 series of desktop APUs based on the Zen 3 CPU and Vega GPU microarchitectures succeeding the Ryzen 4000 "Renoir" series.

Extreme performance enabled for personal computing with up to 8 cores fueled by the world's most advanced 7 nm processor core technology, the AMD Ryzen 5000 G-series desktop processors with Radeon graphics deliver ultra-fast responsiveness and multi-threaded performance for any use case.

German online retailer Mindfactory may not be as big as Amazon, but it puts out its sales figures of PC hardware components, that often get picked up by the tech-press as a sample size. While using its data as a yardstick for AMD outselling Intel in the DIY market is debatable, sales of individual AMD or Intel products provide valuable insights to what consumers are after these days. Apparently, the recently launched Ryzen 7 5700G and Ryzen 5 5600G APUs are already outselling the Ryzen 7 5800X and Ryzen 5 5600X, respectively, for the week of 2nd August running.

AMD designed the Ryzen 7 5700G to succeed the popular Ryzen 7 3700X, and the 5600G to succeed the best-selling Ryzen 5 3600, which explains the absence of "Ryzen 7 5700X" or "Ryzen 5 5600," at least in the DIY market. It's little surprise then, that just as the 3700X outsold the 3800X, Mindfactory bagged orders for 820 units of 5700G, as opposed to 610 units of the 5800X; and 900 units of the 5600G, compared to 680 units of the 5600X. It's interesting to note that the 5700G even outsold the 5600X. The 5700G and 5600G are based on the 7 nm "Cezanne" silicon, which packs up to 8 "Zen 3" cores, and an iGPU with up to 512 stream processors. Unlike "Vermeer," Cezanne is a monolithic die.

Senior Technology Fellow Yuzo Fukuzaki shed light on the elusive new CPU technology AMD unveiled at its Computex 2021 keynote, 3D Vertical Cache (3DV Cache). The company had then detailed it as an additional 64 MB last-level cache stacked on top of a CCD (CPU core complex die), which significantly improves performance, including a claimed 15% average gain in gaming performance, which accounts for a generational performance gain over "Zen 3." The prototype AMD unveiled in its keynote was based on a Socket AM4 processor with "Zen 3" CCDs that have the 3DV Cache components in place. With two such CCDs, a 16-core processor would end up with 192 MB of L3 cache.

Yuzo Fukuzaki's theory sheds light on the most plausible position of 3DV Cache in the processor's cache hierarchy. Apparently, it expands the CCD's L3 cache, and doesn't serve as an "L4" victim cache to the L3. This way, the cache setup remains transparent to the OS, which sees it as a contiguous 96 MB block of L3 cache (per CCD). The 3DV Cache die is an SRAM chip fabricated on the same 7 nm process as the "Zen 3" CCD. It measures 6 mm x 6 mm (36 mm²), and is located above the region of the CCD that typically has the 32 MB L3 SRAM. Fukuzaki estimates that roughly 23,000 TSVs (through-silicon vias), each about 17 µm in size, connect the 3DV Cache die to the main CCD.

AMD's 7 nm "Navi 21" silicon powers the company's Radeon RX 6800 series and flagship RX 6900 XT graphics cards. It's a big chip, competitive with NVIDIA's fastest GeForce RTX 30-series products, and AMD set 16 GB as the standard memory amount for all products based on this chip, despite its 256-bit wide GDDR6 memory interface. Komachi Ensaka spotted a curious-looking Navi 21 product with 8 GB of memory, on the UserBenchmark database. The card is slower than the desktop RX 6800, but found trading blows with the RX 6700 XT. Speculation is rife as to what it could be.

The most plausible theory is that it could be a prototype, with its user testing out UserBenchmark. The GeForce RTX 3070 Ti has a shaky performance equation with the similarly-priced RX 6800, and any attempt to close the gap between the RX 6700 XT and the RX 6800 would cannibalize the latter, unless that's exactly what AMD wants—a product competitive with the RTX 3070 Ti, but with a leaner bill of materials than the RX 6800 on account of the 8 GB memory.

AMD today announced availability of the new AMD Radeon PRO W6000X series GPUs for Mac Pro. The new GPU product line delivers exceptional performance and incredible visual fidelity to power a wide variety of demanding professional applications and workloads, including 3D rendering, 8K video compositing, color correction and more.

Built on groundbreaking AMD RDNA 2 architecture, AMD Infinity Cache and other advanced technologies, the new workstation graphics line-up includes the AMD Radeon PRO W6900X and AMD Radeon PRO W6800X GPUs. Mac Pro users also have the option of choosing the AMD Radeon PRO W6800X Duo graphics card, a dual-GPU configuration that leverages high-speed AMD Infinity Fabric interconnect technology to deliver outstanding levels of compute performance.