Patriot Memory has unveiled the Viper EV330 PCIe M.2 2230 SSD enclosure, specifically designed for users seeking high-speed data transfer, superior heat dissipation, and portability. This enclosure supports the USB 3.2 Gen 2 x2 interface, delivering ultra-fast data transfer speeds of up to 20 Gbps, enabling users to quickly access and back up large files for both professional applications and everyday needs.

The Viper EV330 enclosure accommodates PCIe NVMe SSDs in the M.2 2230 form factor, making it ideal for storage expansion in high-performance laptops, mini PCs, and mobile devices. The aluminium alloy housing not only provides excellent thermal management but also features silicone shock protection to enhance impact resistance, ensuring stable SSD operation even during mobile use.

According to a well-known AMD hardware leaker, Hoang Anh Phu, AMD held a brief press conference where it confirmed that the upcoming Radeon RX 9070 XT and Radeon RX 9070 will carry a 304 Watt and 220 Watt total board power (TBP), respectively. While the post, originally on X is deleted, VideoCardz managed to read the information. AMD's TBP rating is similar to NVIDIA's total graphics power (TGP) metric, which measures both the chip and the memory and other components that the graphics card could contain under full load. So the TBP, and hence TGP, metrics are basically a rough outline of how much power the GPU will draw under full load.

AMD's upcoming Radeon RX 9070 XT and RX 9070 non-XT variants will consume 304 and 220 Watts, respectively, meaning that RDNA 4 IP will be relatively efficient. Interestingly, AMD settled on a 304-watt number instead of rounding it to 305, which usually happens. For reminder, the RX 9070 XT features 4,096 cores at 2.97 GHz boost clock, while the RX 9070 has 3,584 cores at 2.52 GHz. Both cards use a Navi 48 SKU, with 16 GB GDDR6 memory on a 256-bit bus running at 20 Gbps, achieving 640 GB/s bandwidth. Both utilize PCIe 5.0×16 and are rumored to be released on March 6, following their January announcement.

UGREEN, a leading innovator in consumer electronics, is due to showcase its latest innovations at CES 2025 under the theme of "Activate the Possibility of AI." The highlight of the event will be the unveiling of the highly anticipated NASync iDX6011 and NASync iDX6011 Pro devices, which are from the cutting-edge AI NAS lineup of the NASync series. Alongside these groundbreaking products, the Nexode 500 W 6-Port GaN Desktop Fast Charger and the Revodok Max 2131 Thunderbolt 5 Docking Station will also take center stage.

The NASync series AI NAS models are set to redefine expectations with integrated large language models (LLMs) for advanced natural language processing and AI-driven interactive capabilities. Powered by cutting-edge Intel Core Ultra Processors, the iDX6011 and iDX6011 Pro deliver unmatched performance, enabling seamless functionality and exceptional AI applications. These models build on the success of earlier NASync series products, such as the NASync DXP models, which garnered widespread attention and raised over $6.6 million during a Kickstarter campaign in March 2024.

ASMedia Technology Inc., a leading provider of high-speed transmission interface ICs, showcased its latest innovations at CES 2025, focusing on USB4, Thunderbolt 4, and PCIe Gen 4 Switch technologies. These new technologies can be used in many different applications such as consumer electronics, AI, HPC, and industrial computing.

Key highlights include:

• PCIe Gen 4 Switch ASM58000 Series: This low-power, cost-effective solution offers high bandwidth (up to 16 Gbps per lane), ideal for servers, edge computing, NAS, surveillance storage, and industrial PCs. It excels in demanding applications like AI, ML, and HPC.
• USB4 ASM2464PDX with AI Enhancements: The USB4 device controller ASM2464PDX empowers PCs with enhanced AI capabilities. By leveraging USB4 and PCIe, it enables seamless external GPU connectivity for AI tasks, gaming, and content creation.
• Comprehensive USB4 Solutions: ASMedia demonstrated a complete range of USB4 solutions, including host, device, and peripheral controllers, ensuring optimal compatibility and backward compatibility with Thunderbolt 3/4 and legacy USB standards.

ASUS today announced the B850 and B840 series of AM5 motherboards for AMD processors, which includes models from the ROG Strix, TUF Gaming and Prime families. These motherboards give builders higher speeds, more ports and more slots for USB devices and M.2 SSDs. Many ASUS B850 motherboards also break into the next generation of wireless networking with WiFi 7 support, and also provide robust support for fast DDR5 kits. In addition, each model makes the PC building process easier than ever with tool-free M.2 SSD installation, easy troubleshooting and hassle-free installation and removal of graphics cards.

Meanwhile, B840 motherboards are designed to serve the needs of builders who do not intend to overclock their CPUs or install a PCIe 5.0 M.2 SSD or graphics card. For anyone looking to build a very affordable and efficient PC—perhaps one centered around a CPU with powerful integrated graphics capabilities—ASUS B840 motherboards are a cost-effective choice. The first ASUS B840 motherboards all hail from the tried-and-true Prime family.

Cable Matters Inc., a leading connectivity solutions provider, has announced the launch of its Thunderbolt 5 Docking Station, setting an important benchmark for speed, simplicity, and reliability, and offering unmatched performance for modern professionals, gamers, and content creators.

Utilizing Thunderbolt 5 technology, the docking station offers up to 120 Gbps of bandwidth-three times more than Thunderbolt 4-enabling ultra-fast data transfers and superior display experiences. It features an upstream Thunderbolt 5 port compatible with Thunderbolt 3, Thunderbolt 4, Thunderbolt 5, and USB4 based hosts, and three downstream Thunderbolt 5 ports for versatile connections to monitors, SSDs, and other devices. The downstream ports support up to three 8K@60 Hz monitors or two 8K@120 Hz displays for enhanced visual clarity, gameplay, and multitasking.

ASUS today announced the TUF Gaming A2 SSD enclosure for gamers seeking a superfast and reliable storage solution. The enclosure features USB-C 3.2 Gen 2x2 connectivity to deliver data transfer speeds of up to 20 Gbps—twice that of TUF Gaming A1. With IP68-rated water- and dust-resistance plus durability that meets U.S. MIL-STD-810H standards, TUF Gaming A2 keeps the SSD safe in almost any environment. Additionally, its silicone-rubber exterior provides superior on-the-go protection for the SSD. The Q-Latch mechanism allows for easy, tool-free SSD installation, while an efficient thermal management system ensures outstanding cooling to ensure consistent performance. With broad compatibility across PC, Mac, mobile devices, and the latest game consoles, TUF Gaming A2 is perfect for anyone looking to maintain an extensive library of games.

Exceptional SSD performance
TUF Gaming A2 doubles the data transfer speeds of its predecessor, TUF Gaming A1, with USB 3.2 Gen 2x2 connectivity delivering up to 20 Gbps of bandwidth. Even though the latest AAA titles often exceed 100 GB in size, it takes just 90 seconds to transfer an entire game from the SSD to a console such as ROG Ally, or to any other popular gaming platform. When it comes to installing or removing the M.2 SSD, the innovative Q-Latch makes it easy, without requiring specific tools. The latch design employs a simple locking mechanism to secure the drive, so there's no need for screws. Additionally, TUF Gaming A2 features both PCIe NVMe and SATA interfaces, and can accommodate various 2242/2260/2280 SSD form factors while providing up to 4 TB of storage capacity.

QNAP Systems, Inc., a leading computing, networking, and storage solutions innovator, has launched an industrial-grade 16-port full 10GbE L2 managed switch, the QSW-IM3216-8S8T. The rugged industrial case provides eight 10GbE SFP+ fiber and eight 10GbE RJ45 Multi-Gig ports, allowing flexible rackmount, desktop, or wall-mounted installations. It can connect NAS, SNMP, and AV-over-IP devices, making it ideal for deployment in a range of environments to achieve factory automation, smart manufacturing, and meet the diverse needs of smart cities, automotive industry, and traffic monitoring.

"Targeting the Industry 5.0 market which implements AI smart manufacturing, the QSW-IM3216-8S8T assists users in building reliable and secure high-speed 10GbE network infrastructure within an economical budget," said Jerry Deng, Product Manager of QNAP, adding "we also recommend pairing it with the TS-i410X Industrial NAS to create a high-speed data transmission and storage solution for smart factories."

MSI is excited to announce the launch of the DATAMAG 20 Gbps, a magnetic portable SSD designed for fast, reliable, and convenient data storage. Available in 1 TB, 2 TB, and 4 TB capacities, the DATAMAG offers SuperSpeed USB 20 Gbps transfer rates and a unique magnetic attachment system, making it an ideal choice for users who need both speed and portability.

With read/write speeds up to 1,600/1,500 MB/s, the DATAMAG 20 Gbps is built to handle demanding tasks such as ProRes 4K video recording and large file transfers. Whether you're a creative professional working on high-resolution video projects or someone who needs reliable storage on the go, this SSD provides the performance needed to keep your workflow efficient and smooth and supports the highest ProRes formats for video recording on iPhone 16 / 15 Pro (Max).

Infineon Technologies AG today announced the addition of the EZ-USB FX20 programmable USB peripheral controller to its EZ-USB product family. It enables developers to create USB devices that meet the highest performance requirements in AI, image processing and emerging applications. The EZ-USB FX20 peripheral controller offers high-speed connectivity with USB 20 Gbps and LVDS interfaces, increasing the total bandwidth up to six times over its predecessor, the EZ-USB FX3.

"With the growing popularity of USB devices, the demand for compatible and adaptable USB controllers is increasing," said Ganesh Subramaniam, Senior Vice President and General Manager of the Wired Connectivity Solutions Product Line at Infineon. "Therefore, we are continuously improving the features and performance of our EZ-USB peripheral controllers and are pleased to support developers with our new addition to the product family, providing them with a flexible component to create powerful and advanced applications."

Detailed specifications and a somewhat blurry slide showing off all the details of MSI's upcoming high-end MEG Z890 Unify-X motherboard have leaked courtesy of @ChamberTech_ on X/Twitter. The motherboard is likely to appeal to those looking to squeeze every extra bit of performance out of their new Ultra 200K series CPU when they launch next month. You get support for memory speeds of up to 10,000 MT/s and with only two DIMM slots, it might be able to go even higher with the latest DDR5 CUDIMMs. The board also sports a 110 Ampere 20+1+1+1 VRM design on an eight layer PCB, and it also ships with what MSI calls an OC Tuning Controller, suggesting the board is targeting overclockers.

MSI has included a pair of PCIe 5.0 x16 slots for GPUs that presumably operate in dual x8 mode when both slots are in use, and a third PCIe 4.0 x16 slot that operates in x4 mode, as well as a single PCIe 4.0 x1 slot. There's also no less than six M.2 slots, of which two are PCIe 5.0 and one PCIe 4.0, as well as six SATA ports. Where things get a bit interesting is in the network connectivity, as the slide claims that the board features a 5 Gbps Intel Killer Ethernet interface, which is the first we've heard of such a thing, as well as an Intel Killer Wi-Fi 7 and Bluetooth 5.4 module for wireless connectivity.

NVIDIA quietly released a variant of the GeForce RTX 4070 featuring slower 20 Gbps GDDR6 memory, replacing the 21 Gbps GDDR6X that the original RTX 4070 comes with. Wccftech has access to a GALAX branded RTX 4070 GDDR6, and put it through benchmarks focused on comparing it to a regular RTX 4070. Memory type and speed are the only changes in specs, the core-configuration isn't changed, nor is the GPU clock speed. Wccftech's testing shows that the RTX 4070 GDDR6 measures within 0-1% slower than the RTX 4070 (GDDR6X) at 1080p and 1440p resolutions; while the difference between the two is about 2% at 4K Ultra HD.

Wccftech's test-bed is comprehensive, with 27 game tests, each across 3 resolutions; and 7 synthetic tests. The synthetic tests are mainly part of the 3DMark test suite, including Speed Way, Fire Strike, Time Spy, Port Royal, and their presets. Here, the RTX 4070 GDDR6 is nearly identical in performance, with a 0-0.2% delta with the RTX 4070 GDDR6X. In the game tests, performance varies by resolution. 1080p has 0-1% performance delta, with the only noteworthy outliers being "Metro Exodus" (extreme preset), where the RTX 4070 GDDR6 loses 4.2%, and "Alan Wake 2," where it loses 2.3%.

GIGABYTE just released its second AMD 800-series chipset motherboard following last week's debut of the X870E AORUS Master, the more affordable X870 AORUS Elite WIFI7. This board is based on the slightly more affordable AMD X870 (non-E) chipset, which offers connectivity nearly identical to that of the previous-generation AMD B650E, but with the addition of USB4. The board is built in the ATX form-factor, and features a 6-layer PCB. It draws power from a combination of 24-pin ATX and two 8-pin EPS power connectors. The CPU VRM solution consists of a 16+2+2 phase setup, with the vCore side of it being 8-phase with phase doubling. The Socket AM5 is wired to four DDR5 DIMM slots for up to 256 GB of memory.

Expansion slots include a PCI-Express 5.0 x16 wired to the CPU, a PCI-Express 4.0 x16 (electrical Gen 4 x4), and a third PCI-Express 3.0 x16 (electrical Gen 3 x2). The main Gen 5 x16 PEG slot gets a quick-release lever. You press a button, and it ejects the graphics card. Storage connectivity includes three M.2 Gen 5 x4 slots, two of these are wired to the CPU's dedicated Gen 5 x4 NVMe lanes, while the third slot subtracts 8 lanes from the Gen 5 x16 PEG slot. There is a fourth M.2 slot, which is Gen 4 x4, and wired to the X870 FCH. All four slots get heatsinks. Four SATA 6 Gbps ports make for the rest of the storage connectivity.

NVIDIA is possibly launching a more cost-effective variant of its GeForce RTX 4070 graphics card featuring GDDR6 memory, replacing the GDDR6X that it originally launched with. The new SKU would be better differentiated from the RTX 4070 SUPER. When NVIDIA refreshed its RTX 40-series "Ada" product stack in January, it discontinued the RTX 4070 Ti and RTX 4080, which had been replaced in the lineup by the RTX 4070 Ti SUPER and RTX 4080 SUPER, but at the time it didn't tinker with the RTX 4070, which continued to sell at a roughly $50 lower price than the RTX 4070 SUPER. This new SKU could be an attempt by NVIDIA to get further down below the $500-mark in pricing.

The RTX 4070 originally launched with 21 Gbps GDDR6X memory. This new variant sees the memory replaced with 20 Gbps conventional GDDR6. The JEDEC standard GDDR6 chips could be cheaper than GDDR6X, and could very well be the same GDDR6 chip models AMD uses in some of its higher Radeon RX 7000 series SKUs. This, however, comes with a 4.75% drop in memory bandwidth, which NVIDIA probably overcomes with increasing the GPU clocks a touch. The ASIC code for this SKU is AD104-251, compared to the AD104-250 of the original RTX 4070. The core configuration is otherwise unchanged—you get 5,888 CUDA cores across 46 streaming multiprocessors. Galax has a card based on this SKU ready.

Cable Matters, a leader in developing cutting-edge connectivity solutions, introduces a USB4 20 Gbps Switch, meeting the diverse needs of today's dynamic remote workforce. This versatile, powerful device seamlessly blends performance, adaptability, and cost efficiency, offering users an unparalleled connectivity solution.

"The vision of this product grew from the needs of today's remote workforce," says Jeff Jiang, CEO of Cable Matters. "People are doing more than ever with multiple devices and require connectivity solutions that provide performance and flexibility at an affordable price." Unlike traditional switches, the USB4 20 Gbps Switch is engineered to handle two USB-C, USB4 or Thunderbolt enabled devices concurrently, effortlessly scaling to match the demands of gadget-intensive users. Users can instantly share a USB-C, USB4 or Thunderbolt monitor or docking station between two devices, and easily switch between them using either the unit's button or the included remote control.

Gigabyte's new Stealth series of motherboards has to date only had a single product, the Z690 Aorus Elite Stealth, but now, Gigabyte has added its first Stealth board for AMD CPUs. The B650E Aorus Stealth Ice not only has all of its connectors on the back of the board, but it also has a white/silver PCB with a similarly coloured front and back cover. Apart from the DIMM slots, all connectors are also in somewhat matching colours, although judging by the pictures, the colours don't quite match on all the connectors and slots. That aside, the B650E Aorus Stealth Ice is a pretty competent board, as long as you're not interested in adding anything more than a graphics card and some NVMe SSDs, as it has zero additional PCIe expansion slots. It's a rather bold move by Gigabyte, but at the same time, most gamers and consumers don't tend to add PCIe devices to their computers outside of additional storage these days.

Besides the PCIe 5.0 x16 slot, you get two M.2 sockets wired to the CPU, one PCIe 5.0—with its own, larger heatsink—and one PCIe 4.0, as well as a third PCIe 4.0 M.2 socket via the chipset. There's also support for four SATA 6 Gbps drives if more storage is needed. Gigabyte has also added an internal HDMI 1.4 port that's limited to 1080p30, which is intended to be used with Gigabyte's LCD Edge View display. Other connectivity on the back of the PCB includes a USB 3.2 Gen 2x2 (20 Gbps) port, a USB 3.2 (5 Gbps) pin-header as well as the usual USB 2.0, fan headers and LED headers that you'd expect from a modern motherboard. The VRM design is a 12+2+2 design with a 60 Amp DrMOS configuration for the 12+2 phases for the CPU and GPU.

What appears to be the full details of all of Intel's upcoming 800-series chipsets have leaked and although we've already seen what the Z890 chipset will offer, no details of the other SKUs have leaked to date. Maybe the biggest news is that there won't be an H870 chipset this time around, according to the leaker @jaykihn0 over at X/Twitter. The full range appears to be Z890, B860 and H810 for consumers, W880 for workstations and Q870 for corporate desktops. That said, the Z890 and W880 are identical in terms of slots and ports, but whereas the W880 gains full ECC support and Intel vPro and remote management, it loses the CPU and bus overclocking features of the Z890. The Q870 loses four chipset PCIe 4.0 lanes, one 20 Gbps or two 10 Gbps USB 3.2 ports, but is otherwise similar to the Z890, minus all the overclocking features, but with Intel vPro and remote management.

The B860 ends up on the chopping block as always and loses not only 10 chipset PCIe lanes compared to the Z890 chipset, but also four lanes from the CPU and four lanes on the DMI interface to the chipset. On top of that bifurcation goes out the window toigether with PCIe RAID and the chipset is only capable of having two 20 Gbps or four 10 Gbps USB 3.2 ports, but at least memory overclocking is still on the table. Finally, the H810 chipset is as basic as it gets, with no extra CPU PCIe lanes beyond the 16 lanes for a graphics card, but they are at least PCIe 5.0 this time around. The chipset itself is only equipped with a mere eight PCIe 4.0 lanes and it gets zero 20 Gbps and two 10 Gbps USB 3.2 ports. Note that all platforms support one or two Thunderbolt 4 / USB4 ports, but it's unclear if this is native support or via an add-on chip. Also note that the USB ports are shared and the maximum count is that for the 5 Gbps ports in the details below, so the Z890 for example, doesn't have 25 USB 3.2 ports in total, but rather only 10. All in all, we'll see a migration to PCIe 5.0 for the x16 slot across the entire 800-series chipset range and from the B860 and up, all chipsets will have a dedicated PCIe 5.0 x4 interface that will most likely be used for an M.2 slot. Also, the PCIe lane count is the maximum, but due to Intel's HSIO layout, some resources will be shared with SATA and Ethernet, which means not all PCIe lanes will be accessible.

Intel's upcoming Core Ultra "Arrow Lake-S" desktop processor introduces a new socket, the LGA1851, alongside the new Intel 800-series desktop chipset. We now have some idea what the 151 additional pins on the new socket are used for, thanks to a leaked platform map on the ChipHell forums, discovered by HXL. Intel is expanding the number of PCIe lanes from the processor. It now puts out a total of 32 PCIe lanes.

From the 32 PCIe lanes put out by the "Arrow Lake-S" processor's system agent, 16 are meant for the PCI-Express 5.0 x16 PEG slot to be used for discrete graphics. Eight are used as chipset bus, technically DMI 4.0 x8 (these are eight lanes that operate at Gen 4 speed for 128 Gbps per direction of bandwidth). There are now not one, but two CPU-attached M.2 NVMe slots possible, just like on the AMD "Raphael" and "Granite Ridge" processors. What's interesting, though, is that not both are Gen 5. One of these is Gen 5 x4, while the other is Gen 4 x4.

Cable Matters, a leader in providing top-tier connectivity solutions, today announced the launch of its groundbreaking Thunderbolt 5 cable manufactured by Lintes Technology Co., Ltd. The cable is available in 3 different lengths: 1 foot, 1.6 feet and 3.3 feet. This latest innovation delivers unparalleled performance, supporting up to 80 Gbps of bi-directional data transfer and up to 120 Gbps of video bandwidth, a substantial leap from the previous Thunderbolt 4 technology.

The new Thunderbolt 5 cable by Cable Matters represents a significant advancement in connectivity technology, offering increased bandwidth and power delivery capabilities. The Bandwidth Boost technology enables the cable to transmit up to 120 Gbps for the best display experience, providing up to 3 times increase over Thunderbolt 4's maximum bandwidth. This enhancement is particularly beneficial for demanding applications such as high-resolution video streaming, extensive data transfer, and high-speed gaming.

At Computex last week, we got a first look at Gigabyte's new TRX50 AI TOP motherboard—where TOP stands for trillions of operations per second—designed for AMD's Threadripper and Threadripper Pro CPUs. Now the company has released the full specifications of the TRX50 AI TOP and the board is packed to the brim with features. The E-ATX board is home to four PCIe x16 slots, all of which can operate in PCIe 5.0, but one can be limited to PCIe 4.0, depending on your choice of CPU. There's also four NVMe M.2 slots, with CPU dependent functionality. Other features include eight DIMM slots, but these are again CPU dependent, two 10 Gbps Ethernet interfaces via a pair of Marvell AQtion AQC113C controllers, WiFi 7 via a Qualcomm QCNCM865 module and dual audio chips, one for the rear ports and one for the front panel audio.

As this is a 2024 model, USB4 is of course also part of the package, but only one of the two USB4 supports DP Alt-mode, due to there only being one DP input. The board also has an internal USB 3.2 Gen 2x2 (20 Gbps) header for a case mounted USB-C port, a further six rear 5 Gbps USB Type-A ports, plus a pair of internal headers for a further four 5 Gbps USB Type-A ports. The board also has four SATA ports, a 16+8+4 power phase design, a 14-layer PCB, a debug display and the now common easy release features for the M.2 drive heatsinks and the graphics card. The TRX50 AI TOP also sports what Gigabyte calls "UC BIOS" which has a "User-Centred intuitive UX with Quick Access function", but there's currently no details of how this differs from previous UEFI releases from Gigabyte. There was no word on pricing.

Team Group showed off its latest portable SSDs at the 2024 Computex. We begin our tour of the Team Group booth with the T-Force M400, a portable SSD that takes advantage of the 40 Gbps USB4 interface to serve up phenomenal transfer speeds of up to 3700 MB/s reads, with up to 3400 MB/s writes. The M400 comes in capacities of 1 TB, 2 TB, and 4 TB. The T-Create P35 is a compact high-security portable SSD, which comes with native encryption, and the SSD Security Toolbox interface that protects data even against bruteforcing and BadUSB attacks. This drive packs TCG-Opal 2.0 with 256-bit AES hardware encryption. The drive takes advantage of the 10 Gbps USB 3.2 Gen 2 interface, and offers transfer speeds of up to 1000 MB/s reads/writes.

The T-Create CinemaPr P31 is a high performance portable SSD meant for lossless digital cinematography. It's designed to fit with popular cinema camera rigs, as well as video rigs made of DSLRs or even smartphones. The drive features a 20 Gbps USB 3.2 Gen 2x2 interface, and has up to 2000 MB/s of transfer speeds on tap, with enough speeds for 4K or 8K raw captures. There's also the T-Create CinemaPr R41 CFexpress 4.0 Type B, a related product with similar camera-mounting capabilities.

At Computex 2024, we went hands-on with the only custom-design AMD Radeon RX 7900 XTX graphics card to implement the 16-pin 12V-2x6 power connector. We've written quite a bit about this card, but weren't expecting it to be this long. This isn't just the only RX 7900 XTX with a 12V-2x6, but also the only air-cooled RX 7900 XTX that's strictly 2 slots-thick. ASRock made this card for graphics rendering farms, or AI development/inferencing builds, in which a GPU server would have 4-6 of these packed like sardines. The lateral blower helps ensure proper ventilation. The backplate is recessed to ensure better ventilation to the neighboring card.

The card sticks to AMD reference clock speeds of 2270 MHz Game clock, and 20 Gbps (GDDR6-effective) memory speed. ASRock also gave us a technical deep-dive into the card's design. For one, it's confirmed that the power connector is 12V-2x6 (H++) and not 12VHPWR. The cooling solution consists of a large copper vapor-chamber plate that makes contact with the GPU, memory, and VRM; conveying heat to an aluminium channel heatsink, which is ventilated by a high-speed lateral blower. The cooler shroud and backplate are both made of aluminium.

MSI also brought plenty of SSD products to the Computex 2024 show, including the new SPATIUM M560 PCIe 5.0 NVMe M.2 SSD based on the Phison E31T controller and Kioxia's 218-layer BiCS 3D TLC NAND, the new SPATIUM non-metal Vapor Chamber SSD Thermal Solution, and a couple of SSD accessories, including the DATAMAG Magnetic Portable SSD and the DATAVAULT SSD Enclosure.

As said, the new SPATIUM M560 PCIe 5.0 NVMe M.2 SSD will use the Phison 7 nm E31T controller paired up with Kioxia 8th generation 218-layer BiCS 3D TLC NAND and offer transfer rates of up to 10,000 MB/s. The controller, according to MSI, should bring up to a 15 percent reduction in power consumption as well as "maximize bandwidths and link efficiency". It will be available in 1 TB, and 2 TB capacities, and MSI will probably have a version that will include the new SPATIUM non-metal Vapor Chamber SSD Thermal Solution which has been also showcased at the Computex 2024 show. MSI says that this is yet another world's first, which uses "a two-phase flow transition of gas and liquid with minimal signal interference, ensuring rapid and silent heat exchange." According to MSI, the new SPATIUM non-metal Vapor Chamber SSD Thermal Solution should provide up to 11 degrees lower temperature compared to a standard aluminium heatsink.

Silicon Motion Technology Corporation ("Silicon Motion"), a global leader in designing and marketing NAND flash controllers for solid-state storage devices, today launched the SM2322, the industry's fastest single-chip high-performance, low-power, and cost-effective solution for external portable SSDs, supporting up to 8 TB of storage and achieves unparalleled data transfer rates of 20 Gbps for storing and accessing large amounts of content seamlessly from AI smartphones, high-performance multimedia devices, and game consoles.

With the increase of AI-capable devices, high density and high-performance storage solutions are becoming more critical to consumers. Portable SSDs powered by the new SM2322 controller are the only solution that offers low-cost, high-density and high-performance, making it the ideal solution for these applications. Equipped with a USB 3.2 Gen 2x2 interface with 20 Gb/s bandwidth and fully integrated hardware and software solution delivering peak sequential read and write transfer speeds of 2,100 MB/s and 2,000 MB/s, respectively, with up to 8 TB capacity, SM2322 powered high-performance portable SSDs enable an ultra-compact and lightweight form factor. The SM2322 supports the ProRes format and MFi specification for iPhone users while also being compatible with Windows, Android OS, and macOS, making it an ideal high-density, high-performance portable storage solution for AI smartphones, HD content creators, and gaming enthusiasts who require high-density, high-performance portable storage.

Today, we have the latest round of leaks that suggest that AMD's upcoming RDNA 4 graphics cards, codenamed the "RX 8000-series," might continue to rely on GDDR6 memory modules. According to Kepler on X, the next-generation GPUs from AMD are expected to feature 18 Gbps GDDR6 memory, marking the fourth consecutive RDNA architecture to employ this memory standard. While GDDR6 may not offer the same bandwidth capabilities as the newer GDDR7 standard, this decision does not necessarily imply that RDNA 4 GPUs will be slow performers. AMD's choice to stick with GDDR6 is likely driven by factors such as meeting specific memory bandwidth requirements and cost optimization for PCB designs. However, if the rumor of 18 Gbps GDDR6 memory proves accurate, it would represent a slight step back from the 18-20 Gbps GDDR6 memory used in AMD's current RDNA 3 offerings, such as the RX 7900 XT and RX 7900 XTX GPUs.

AMD's first generation RDNA used GDDR6 with 12-14 Gbps speeds, RDNA 2 came with GDDR6 at 14-18 Gbps, and the current RDNA 3 used 18-20 Gbps GDDR6. Without an increment in memory generation, speeds should stay the same at 18 Gbps. However, it is crucial to remember that leaks should be treated with skepticism, as AMD's final memory choices for RDNA 4 could change before the official launch. The decision to use GDDR6 versus GDDR7 could have significant implications in the upcoming battle between AMD, NVIDIA, and Intel's next-generation GPU architectures. If AMD indeed opts for GDDR6 while NVIDIA pivots to GDDR7 for its "Blackwell" GPUs, it could create a disparity in memory bandwidth performance between the competing products. All three major GPU manufacturers—AMD, NVIDIA, and Intel with its "Battlemage" architecture—are expected to unveil their next-generation offerings in the fall of this year. As we approach these highly anticipated releases, more concrete details on specifications and performance capabilities will emerge, providing a clearer picture of the competitive landscape.