TrendForce reports that continued production cuts by DRAM suppliers have led to a gradual quarterly decrease in overall DRAM supply. Seasonal demand, on the other hand, is helping to mitigate inventory pressure on suppliers. TrendForce projects that the third quarter will see the ASP for DRAM converging towards a 0~5% decline. Despite suppliers' concerted efforts, inventory levels persistently remain high, keeping prices low. While production cutbacks may help to curtail quarterly price declines, a tangible recovery in prices may not be seen until 2024.

PC DRAM: The benefits of consolidated production cuts on DDR4 by the top three suppliers are expected to become evident in the third quarter. Furthermore, inventory pressure on suppliers has been partially alleviated due to aggressive purchasing by several OEMs at low prices during 2Q23. Evaluating average price trends for PC DRAM products in 3Q23 reveals that DDR4 will continue to remain in a state of persistent oversupply, leading to an expected quarterly price drop of 3~8%. DDR5 prices—influenced by suppliers' efforts to maintain prices and unmet buyer demand—are projected to see a 0-5% quarterly decline. The overall ASP of PC DRAM is projected to experience a QoQ decline of 0~5% in the third quarter.

AMD Radeon RX 7800 XT will be a much-needed performance-segment addition to the company's Radeon RX 7000-series, which has a massive performance gap between the enthusiast-class RX 7900 series, and the mainstream RX 7600. A report by "Moore's Law is Dead" makes a sensational claim that it is based on a whole new ASIC that's neither the "Navi 31" powering the RX 7900 series, nor the "Navi 32" designed for lower performance tiers, but something in between. This GPU will be AMD's answer to the "AD103." Apparently, the GPU features the same exact 350 mm² graphics compute die (GCD) as the "Navi 31," but on a smaller package resembling that of the "Navi 32." This large GCD is surrounded by four MCDs (memory cache dies), which amount to a 256-bit wide GDDR6 memory interface, and 64 MB of 2nd Gen Infinity Cache memory.

The GCD physically features 96 RDNA3 compute units, but AMD's product managers now have the ability to give the RX 7800 XT a much higher CU count than that of the "Navi 32," while being lower than that of the RX 7900 XT (which is configured with 84). It's rumored that the smaller "Navi 32" GCD tops out at 60 CU (3,840 stream processors), so the new ASIC will enable the RX 7800 XT to have a CU count anywhere between 60 to 84. The resulting RX 7800 XT could have an ASIC with a lower manufacturing cost than that of a theoretical Navi 31 with two disabled MCDs (>60 mm² of wasted 6 nm dies), and even if it ends up performing within 10% of the RX 7900 XT (and matching the GeForce RTX 4070 Ti in the process), it would do so with better pricing headroom. The same ASIC could even power mobile RX 7900 series, where the smaller package and narrower memory bus will conserve precious PCB footprint.

Intel issued a product change notification (PCN), announcing the discontinuation of the reference-design Arc A770 16 GB Limited Edition graphics card. The product is now marked end-of-life (EOL) from Intel's end. What this means is that retailers cannot order more of it from Intel. The Arc A770 has 8 GB as its standard video memory size, and in its reference spec, uses 8 GB of 16 Gbps-rated GDDR6 memory across a 256-bit wide memory bus. The first-party A770 Limited Edition card comes with 16 GB of faster 17.5 Gbps memory. Memory aside, both the A770 Limited Edition, and custom-design A770 graphics cards have the same GPU core-configuration.

The Moore Threads MTT S80 gaming-oriented graphics card has been tested mostly by Chinese hardware publications, but Japan's PC Watch has managed to get hold of a sample unit configured with 16 GB GDDR6 (14 Gbps) for evaluation purposes and soon published their findings in a "HotHot REVIEW!" The MTT S80 GPU appears to be based on PowerVR architecture (developed by Imagination Technologies), but official Moore Threads literature boasts that their own Chunxaio design is behind all proceedings with 4096 "MUSA" cores. The GPU's clock speed is set at 1.8 GHz, and maximum compute performance has been measured at 14.2 TFLOPS. A 256-bit memory bus grants a bandwidth transfer rate of 448 GB/s. PC Watch notes that the card's support for PCIe Gen 5 x 16 (offering up to 128 GB/s bandwidth) is quite surprising, given the early nature of this connection standard.

Moore Threads has claimed in the past that their cards support Direct X, but PC Watch has discovered that the S80 does not work with DX12, and their tests also demonstrated significant compatibility issues under DX11—with plenty of system freezes and error messages logged. The reviewer(s) had to downshift in some cases to DX9 game environments, in order to gather reliable/stable data. TPU's GPU-Z utility is shown to have no registration information for the S80, and it cannot read the GPU's clock. PC Watch compared their sample unit to an NVIDIA GeForce GTX 1050 Ti graphics card—the entry level 2016-era GPU managed to best the newer competition in terms of in-game performance and power efficiency.

NVIDIA could be advancing the launch of its GeForce RTX 4060 (non-Ti) graphics card from its July 2023 launch the company originally announced. Leaked documents shared by MEGAsizeGPU say that NVIDIA could make the RTX 4060 available on June 29, which means reviews of the card could go live on June 28 for the MSRP cards, and June 29 for the premium ones priced above MSRP. It was earlier expected to launch alongside the 16 GB variant of the RTX 4060 Ti, in July.

The RTX 4060 is a significantly different product from the RTX 4060 Ti the company launched in May, it is based on the smaller AD107 silicon. The card is expected to feature 3,072 CUDA cores, 24 RT cores, 96 Tensor cores, 96 TMUs, and 32 ROPs, compared to the 4,352 CUDA cores, 34 RT cores, 136 Tensor cores, 136 TMUs, and 48 ROPs, of the RTX 4060 Ti. The memory configuration is similar, with 8 GB of GDDR6 memory across a 128-bit wide memory bus, however, the memory speed is slightly lower, at 17 Gbps vs. 18 Gbps of the Ti. The RTX 4060 has a TGP of just 115 W. The company hasn't finalized its price, yet.Update Jun 14th: NVIDIA confirmed the launch date on Twitter:

NVIDIAThe GeForce RTX 4060 will now be available to order starting June 29, at 6AM Pacific.

Zephyr has produced an NVIDIA GeForce custom graphics card that sports a very unique pink printed circuit board—bright and pastel colors have featured on cooling solutions in the past, but this new-ish product presents the first example of a PCB with a tinge of blush. Renowned hardware tipster harukaze5719 broke from his normal delivery of very cold and macho tech on social media, and shared his discovery of Zephyr's GeForce RTX 3060 Ti compact ITX card.

International buyers will be disappointed to learn that the pink Ampere card is a China market exclusive, with the company only offering a limited number of products on JD.com. VideoCardz notes that the card's specifications are not at all special, despite its interesting compact form factor and brightly toned cooling solution design. It is a non-overclocked model based on the older RTX 3060 Ti GDDR6 GPU variant with a standard 1750 MHz boost clock, 8 GB VRAM configuration, and a single 8-pin power connector.

The price of GDDR6 memory has continued to fall sharply - over recent financial quarters - due to an apparent decrease in demand for graphics cards. Supply shortages are also a thing of the past—industry experts think that manufacturers have been having an easier time acquiring components since late 2021, but that also means that the likes of NVIDIA and AMD have been paying less for VRAM packages. Graphics card enthusiasts will be questioning why these savings have not been passed on swiftly to the customer, as technology news outlets (this week) have been picking up on interesting data—it demonstrates that spot prices of GDDR6 have decreased to less than a quarter of their value from a year and a half ago. 3DCenter.org has presented a case example of 8 GB GDDR6 now costing $27 via the spot market (through DRAMeXchange's tracking system), although manufacturers will be paying less than that due to direct contract agreements with their favored memory chip maker/supplier.

A 3DCenter.org staffer had difficulty sourcing the price of 16 Gb GDDR6 VRAM ICs on the spot market, so it is tricky to paint a comparative picture of how much more expensive it is to equip a "budget friendly" graphics card with a larger allocation of video memory, when the bill-of-materials (BoM) and limits presented by narrow bus widths are taken into account. NVIDIA is releasing a GeForce RTX 4060 Ti 16 GB variant in July, but the latest batch of low to mid-range models (GeForce RTX 4060-series and Radeon RX 7600) are still 8 GB affairs. Tom's Hardware points to GPU makers sticking with traditional specification hierarchy for the most part going forward: "(models) with double the VRAM (two 16 Gb chips per channel on both sides of the PCB) are usually reserved for the more lucrative professional GPU market."

Chips and Cheese published their very in-depth review of AMD's Radeon RX 7600 GPU last weekend - a team member (Jiray) took it upon themselves to actually buy the card, since a sample unit was not supplied for evaluation. The site's exploration of this graphics processing unit on an architectural level revealed a couple of positive aspects - which comes as a minor surprise since the Radeon RX 7600 received a generally lukewarm reception upon launch at the end of last month. Thanks to the Radeon RX 7600's Navi 33 XL GPU being a monolithic chip it seems to outpace—in terms of cache and memory latency performance—chiplet-based designs as featured in the vastly more powerful (and expensive) Radeon RX 7900-series cards.

Factoring in the smaller space that the RDNA 3 Navi 33 die occupies - it seems that it gains an advantage over the flagship card. Chips and Cheese reports that AMD's RX 7900 XTX takes up to 58% longer to access and pull data from its pool of Infinity Cache, when contrasted with the recently released sibling. The RX 7600 GPU exhibits 15% lower VRAM latencies compared to the RX 7900 XTX when retrieving data from the onboard GDDR6 VRAM chiplets. The review points to a greater disparity between current high-end and mid-range cards when looking back at equivalent models from the preceding generation: "The difference is especially large with RDNA 3. With RDNA 2, the RX 6900 XT had 151.57 ns of Infinity Cache latency compared to 130 ns on the RX 6600 XT, or a 16.5% latency penalty for the larger GPU." Chips and Cheese reckons that AMD's Navi 31's "chiplet configuration may be causing higher latency."

ASRock, the leading global motherboard, graphics card and mini PC manufacturer, today launched the new Phantom Gaming, Steel Legend and Challenger series graphics cards based on the new AMD Radeon RX 7600 GPUs. In addition to the popular Phantom Gaming and Challenger models, the new Steel Legend series graphics cards are designed in white with eye-catching snow camouflage elements and fancy ARGB lighting effects that support Polychrome SYNC technology, providing more premium choices for consumers who want to assemble white-themed PC builds.

The new ASRock AMD Radeon RX 7600 Series graphics cards are built on the groundbreaking AMD RDNA 3 architecture, featuring up to 32 redesigned compute units and second-generation AMD Infinity Cache technology. They also offer the latest features and capabilities including AMD Radiance Display Engine, full AV1 encoding and more to power high-performance 1080p gaming, streaming and content creation applications. The new ASRock AMD Radeon RX 7600 Series graphics cards are equipped with high-speed 8 GB GDDR6 memory at 18 Gbps and are pre-overclocked to deliver higher levels of performance. The AMD Radiance Display Engine provides 12 bit-per-channel color for up to 68 billion colors for incredible color accuracy.

The SAPPHIRE PULSE AMD Radeon RX 7600 8 GB Graphics Card headlines with 2048 stream processors running with a Boost Clock of up to 2755 MHz and a Game Clock of up to 2355 MHz. The latest 8 GB of GDDR6 high-speed memory clocked at 18 Gbps Effective with 32 MB of AMD Infinity Cache, which dramatically reduces latency and power consumption, ensuring higher overall gaming performance than traditional architectural designs. To support the latest display monitors in the market, it is equipped with 4 output ports, including 1x HDMI and 3x DisplayPort 1.4 with DSC outputs. The SAPPHIRE PULSE AMD Radeon RX 7600 8 GB Graphics Card series features 32 powerful enhanced Compute Units and 32 Ray Accelerators.

Noise vs. Temperature
The SAPPHIRE PULSE AMD Radeon RX 7600 Graphics Card is designed around the renowned Dual-X Cooling Technology for a whisper quiet and low temperature gaming experience. Two large fans equipped with sturdy Dual Ball Bearings for longevity are designed with the effective Hybrid Fan Blades to work in conjunction with the Intelligent Fan Control feature and Precision Fan Control feature. Airflow is streamlined for optimal heatsink coverage, and GPU component temperatures will be low to balance performance and fan noise. Rest assured with the Fuse Protection feature, built into the circuit of the external PCIe power connector to preserve the components and protect the graphics card.

Achronix Semiconductor Corporation, a leader in high-performance FPGAs and embedded FPGA (eFPGA) IP, today announced that its VectorPath accelerator card featuring a Speedster 7t FPGA has been certified by PCI-SIG for PCIe Gen 5 and is the first and only FPGA accelerated CEM add-in card certified for PCIe Gen 5 x16 at 32 gigatransfers per second (GT/s) on the PCI-SIG Integrator's list. VectorPath S7t-VG6 accelerator cards are designed to reduce time to market when developing high-performance compute and acceleration functions for AI, ML, networking and data center applications and are shipping today.

"Achronix continues to drive acceleration in the high-performance FPGA add-in card market," said Craig Petrie, Vice President at BittWare. "Achieving PCI-SIG Gen 5 certification is an important milestone. Our customers can be assured of the highest PCIe bandwidths and have confidence that VectorPath cards will interoperate with Gen 5 servers."

The "AD107" is expected to be the smallest client GPU based on the NVIDIA "Ada Lovelace" graphics architecture. The upcoming GeForce RTX 4060 (non-Ti) is rumored to be maxing this silicon out (enabling all available shaders). MEGAsizeGPU scored one of the first pictures of the "AD107" in the flesh, revealing a small fiberglass substrate, and a visibly smaller die than the "AD106" powering the RTX 4060 Ti.

The "AD107" silicon is expected to feature 3,072 CUDA cores across 24 SM (streaming multiprocessors). The GPU features a 128-bit wide memory interface much like the "AD106," and NVIDIA is expected to use conventional 17 Gbps GDDR6 memory, which works out to 272 GB/s memory bandwidth. 8 GB is the standard memory size for the RTX 4060. With "Ada," NVIDIA has rebalanced the memory sub-system with greater reliance on on-die caches, and the "AD107" features a 24 MB L2 cache. Much like the "AD106," the smaller "AD107" features a PCI-Express 4.0 x8 host interface. The ace up its sleeve has to be power, with even the maxed out RTX 4060 only being rated for 119 W of TGP.

Listings for two custom AMD Radeon RX 7600 GPUs have appeared on PC Canada's webstore, according to VideoCardz - who were notified by keen-eyed locals from the Great White North. The online electronics seller has likely created these page entries by mistake - the information on hand is not exactly refined, indicating that the temporary text has come straight from hardware distributors - Althon Micro is Sapphire's chosen distribution and support partner for North American territories.

The prematurely published pages show that the Sapphire Pulse and MSI Mech 2X Classic models are both overclocked and specced with 8 GB of GDDR6 memory. The custom graphics cards are not ready to purchase, as expected, with an indication of zero units being in-stock. At the time of writing, PC Canada has the Sapphire RX 7600 PULSE OC listed at $335.80 (451.99 CAD), and the MSI Mech 2X RX 7600 at $329.85 (443.99 CAD). These prices are subject to change, given that the information presented seems to have originated from placeholder sources.

NVIDIA's upcoming GeForce RTX 4060 Ti and RTX 4060, expected to make for a twin-launch in May 2023, will come with 8 GB as the standard memory size. This was confirmed in screenshots of a leaked listing of an MSI pre-built gaming desktop. The upcoming MSI MAG Infinite S3 comes with GeForce RTX 4060 Ti and RTX 4060 GPU options, both of which have been mentioned in the listing as featuring 8 GB of graphics memory. The CPU options span between the Intel Core i5-13400F and the i7-13700F. 16 GB (2x 8 GB) main memory, and 1 TB SSD are the other standard equipment.

An 8 GB memory size confirms the 128-bit memory interface of the "AD106" silicon the RTX 4060 series is expected to be based on. 8 GB would actually be a downgrade compared to the current RTX 3060, which offers 12 GB of GDDR6 memory across a 192-bit memory bus. To be fair, the RTX 3060 Ti gives you just 8 GB of memory (albeit over a 256-bit wide memory bus). With the "Ada" graphics architecture powering the RTX 40-series, NVIDIA has significantly redesigned the memory sub-system of the GPU, with greater design emphasis on large on-die caches, so the GPU relies less on discrete memory bandwidth.

Sparkle, a Taiwanese computer electronics maker, is again entering the GPU market after almost ten years of inactivity in the space. A while back, Sparkle was one of NVIDIA's original Add-In Board (AIB) partners and helped them launch the GeForce 7900 PCI GPU. The company continued to make NVIDIA-based GPUs until GTX 700 series in 2013. After a decade, Sparkle is back again with ambitions to be Intel's AIB partner and announced not one but three graphics cards to start. Called A750 Titan, A750 Orc, and A380 Elf, these cards feature triple-fan, dual-fan, and single-fan coolers, respectively.

The first in line is the Sparkle A750 Titan, a triple-fan, 2.5-slot design based on Intel Arc A750 GPU. Featuring 8 GB of GDDR6 VRAM, this top-end A750 SKU is clocked at 2300 MHz, up from the factory's 2050 MHz frequency. Titan's smaller brother is Sparkle A750 Orc, a dual-fan, "2.2"-slot (we assume smaller than two and a half and bigger than two slots) GPU with the same A750 GPU; however, it clocked slightly lower at 2200 MHz. Both models feature single HDMI 2.0 and 3x DisplayPort 2.0 output ports and require two 8-pin PCI power connectors. Lastly, we have the Sparkle A380 Elf, a half-length, ITX-sized GPU that fits in two slots and has a single-fan cooler. It is based on Intel Arc A380 and has identical specifications without factory overclocks applied. Pricing and availability are currently unknown.

NVIDIA has begun selling its compact RTX 4000 Ada Lovelace graphics card, offering GeForce RTX 3070-like performance at a mere 70 W power consumption, allowing it to fit in almost all desktop PCs. The low-profile, dual-slot board is priced higher than the RTX 4080 as it targets professional users, but it can still be used in a regular gaming computer. PNY's RTX 4000 Ada generation graphics card is the first to reach consumer shelves, currently available for $1,444 at ShopBLT, a retailer known for obtaining hardware before its competitors. The card comes with four Mini-DisplayPort connectors, so an additional mDP-DP or mDP-HDMI adapter must be factored into the cost.

The NVIDIA RTX 4000 SFF Ada generation board features an AD104 GPU with 6,144 CUDA cores, 20 GB of GDDR6 ECC memory, and a 160-bit interface. With a fixed boost frequency floating around 1560 MHz to reduce overall board power consumption, the GPU is rated for just 70 Watts of power. To emphasize the efficiency, this card requires no external PCIe power connector, as all the juice is fed through the PCIe slot. The GA104 graphics processor in this configuration delivers a peak FP32 performance of 19.2 TFLOPS, comparable to the GeForce RTX 3070. The 20 GB of memory makes the card more valuable for professionals and AI researchers needing compact solutions. Although the card's performance is overshadowed by the recently launched GeForce RTX 4070, the RTX 4000 SFF Ada's professional drivers, support for professional software ISVs, and additional features make it a strong contender in the semi-professional market. Availability and pricing are expected to improve in the coming weeks as the card becomes more widely accessible.More images, along with specification table, follow.

Rambus Inc., a premier chip and silicon IP provider making data faster and safer, today announced a new product milestone for GDDR6 memory interface performance. The Rambus GDDR6 PHY delivers a market-leading data rate of up to 24 Gigabits per second (Gb/s), providing 96 Gigabytes per second (GB/s) of bandwidth per GDDR6 memory device. As part of a system-level solution, the Rambus GDDR6 offering enables cost-efficient, high-bandwidth memory performance for AI/ML, graphics and networking applications.

"With the new level of performance achieved by our GDDR6 PHY, designers can deliver the bandwidth needed by the most demanding workloads," said Sean Fan, chief operating officer at Rambus. "As with our industry-leading HBM3 memory interface, this latest achievement demonstrates our continued commitment to advancing state-of-the-art memory performance to meet the needs of advanced computing applications such as generative AI."

An unreleased GeForce RTX 3060 "Super" graphics card surfaced on the web. The original and popular RTX 3060 falls short of maxing out the 8 nm "GA106" siicon it is based on, with 28 out of 30 streaming multiprocessors being enabled (that's 3,584 out of 3,840 CUDA cores). This odd-ball graphics card maxes the silicon out, enabling all 30 SM and 3,840 CUDA cores, 120 Tensor cores, 30 RT cores, 120 TMUs, and 48 ROPs. This card reportedly has the ASIC code "GA106-400-A1" and device ID of 10DE-2501. The memory interface is still 192-bit wide, as is the memory speed of 15 Gbps (GDDR6-effective), and it has the same 12 GB of memory. Besides more shaders, the card has been given higher clock speeds than a production RTX 3060, with up to 1875 MHz boost, compared to 1777 MHz. Alas, this is one of many unofficial rare graphics cards that never went into production, and which NVIDIA doesn't officially support with driver updates.

NVIDIA is preparing its fifth GeForce RTX 40-series "Ada" graphics card launch in May 2023, with the GeForce RTX 4060 Ti. Red Gaming Tech reports that the company could target the USD $450 price-point with this SKU, putting it $150 below the recently launched RTX 4070, and $350 below the RTX 4070 Ti. The RTX 4060 Ti is expect to nearly max-out the 5 nm "AD106" silicon, the same one that powers the RTX 4070 Laptop GPU. While the notebook chip maxes it out, featuring all 4,608 CUDA cores physically present across its 36 SM, the desktop RTX 4060 Ti will be slightly cut down, featuring 34 SM, which work out to 4,352 CUDA cores. The "AD106" silicon features a 128-bit wide memory interface, and NVIDIA is expected to use conventional 18 Gbps-rated GDDR6 memory chips. The design goal behind the RTX 4060 Ti could be to beat the previous-generation RTX 3070, and to sneak up on the RTX 3070 Ti, while offering greater energy efficiency, and new features such as DLSS 3.

It looks like Spring-Summer will see NVIDIA's GeForce RTX 40-series get a rapid ramp up to high-volume market segments. In April, the company is planning to release the GeForce RTX 4070, which should hold the performance-segment end of things, albeit at an eye-watering $750 rumored price. May 2023 could see NVIDIA come out with two product launches before the 2023 Computex; the GeForce RTX 4060 Ti, and the RTX 4060.

Both the RTX 4060 Ti and RTX 4060 are expected to be based on the "AD106" silicon, which is expected to have around 36 streaming multiprocessors (4,608 CUDA cores, 144 Tensor cores, 36 RT cores, 144 TMUs, 48 ROPs); and a 128-bit wide memory interface. NVIDIA is expected to use 18 Gbps-rated conventional GDDR6 memory chips with these SKUs, which should lower costs (for the company) in comparison to the more exotic 19 Gbps GDDR6X.

TrendForce reports that several suppliers, such as Micron and SK hynix, have started scaling back DRAM production. The ASP of DRAM plunged 20% in 1Q23, and this price decline is predicted to slow down to 10~15% next quarter. It's uncertain whether or not demand will recover in 2H23. Therefore, the ASP of DRAM has continued to fall as inventory levels are high from the suppliers' side, and prices will only rebound if there is a significant decrease in production.

PC DRAM: Purchase quantity from buyers has fallen drastically over the past three quarters; buyers have around 9~13 weeks of PC DRAM stock remaining. Despite suppliers having already cut production in the PC DRAM segment, DDR4 8 GB module is still likely to fall by more than 10% in 2Q23. There is a possibility that PC OEMs may purchase more DRAM because prices have been down to a relatively low level, but it is still under observation whether or not this can mitigate the inventory overstock situation from the suppliers' side. TrendForce predicts the ASP of PC DRAM will fall between 10~15%.

A curious OEM-only AMD Radeon RX 6300 desktop graphics card surfaced on Chinese peer-to-peer marketplaces. The card is a true single-slot, half-height (low-profile) graphics card, with an active fan-heatsink based cooling; and two display outputs—a pair of HDMI 2.1 ports. The card draws all its power from the PCIe slot, and features a PCI-Express 4.0 x4 host interface. The RX 6300 is based on the same 6 nm "Navi 24" silicon as the RX 6400 and RX 6500 XT, although it is rumored to be heavily cut down. The mobile RX 6300M has 768 out of 1,024 stream processors enabled, so one can expect a similar cut-down for the RX 6300. Perhaps the most interesting piece of specifications is the memory—2 GB GDDR6. It's possible that the card has a puny 32-bit memory interface, half that of the 64-bit interface the "Navi 24" is capable of. The person selling it has it listed at just ¥399 RMB (around $60).

After GUNNIR showed the same product back in January, ADLINK is now offering both Intel Arc A-series GPUs in MXM form factor. The MXM (Mobile PCI Express Module) is a standardized form factor that is used mostly in laptops and some small form factor PCs. Product pages confirm that ADLINK offers both the Intel Arc A370M and the Intel Arc A350M in MXM form factor.

According to specifications The ADLINK MXM-AXe, as the product is called, is MXM 3.1 Type A based on Intel Arc GPU, packing 8 Xe-cores, 128 Execution Units, 4 GB of GDDR6 memory, and TDP of 35-50 W, which is pretty much standard for the Arc A370M GPU. The company also offers the same product with A350M GPU with TDP of 25-35 W. With decent power efficiency, full AV1 hardware encoding, and support for up to four 4K displays, such a GPU would be perfect for small form factor machines, and could be even a decent upgrade for some laptops.

NVIDIA's mid-range, high-volume graphics card based on the GeForce "Ada" graphics architecture, the GeForce RTX 4060, a 60-class product, could feature specs that make it possible for NVIDIA to carve it out either with a maxed out 5 nm AD107 silicon, or a heavily cut-down AD106. Kopite7kimi, a reliable source with NVIDIA leaks, says that the RTX 4060 has specs which align with the full-spec AD107, with 3,072 CUDA cores across 24 streaming multiprocessors (SM), 96 Tensor cores, 96 TMUs, an unknown ROP count, and 8 GB of 18 Gbps GDDR6 memory. The memory bus width is hard to predict with this generation. The GPU's on-die L2 cache is 24 MB in size. The card has a typical graphics power (TGP) of 115 W, making it possible to build cards with just one 6-pin PCIe power connector.

The new HEATKILLER V PRO graphics card coolers for Nvidia's RTX 4090 Founders Edition have been specially developed to cope with the high power dissipation of 450 watts and more. In order to achieve the necessary performance increase, the German manufacturer decided to introduce a modular high-performance bottom plate as well as implementing existing proven technologies.

To increase the cooling performance in the GPU core area, Watercool relies on a modular high-performance base plate. Due to a fine-slit fin structure and the resulting increase in the cooling surface, this ensures a significantly higher cooling performance compared to a monolithic design. At the same time, the German manufacturer has optimized the fin structure in order to achieve a high flow rate as well. The distances to heat-relevant components were kept tight, which means that the heat of the thermal problem zones like GDDR6X Ram and voltage converters can be dissipated ideally.