Samsung has reportedly produced the world's first "standalone" DRAM module that uses a process tech below 10nm. Samsung's New Sub-10nm DRAM Technology To Boost Densities & Uses New Materials For long, the DRAM industry has relied on a 10nm process technology for producing integrated circuits. 10nm DRAM technologies range from 1x, 1y, 1x, 1a, 1b, 1c, and 1d. Now, Samsung is working on a brand new 10a process technology for DRAM that goes below the official "10nm" process limit. Samsung Electronics has produced the world's first single-digit nanometer DRAM working die. It is reported that the company plans to rapidly […]

Read full article at https://wccftech.com/samsung-breaks-10nm-dram-barrier-new-4f-cell-structure-boosts-density/

Think of DRAM, but in a NAND-like structure, that's the basic concept of 3D X-DRAM, a revolution for the memory markets, bringing higher densities for AI. 3D X-DRAM Is Now Closer To Reality, An HBM-Replacement That Offers Higher Densities For AI In 2023, US-based NEO Semiconductor announced its brand new project called 3D X-DRAM, which was going to address the DRAM capacity bottleneck by leveraging a 3D NAND-like architecture. The company also unveiled two 3D X-DRAM cells, which will be integrated into memory solutions based on 3D X-DRAM. These include 1T1C and 3T0C DRAM cells, offering up to 512Gb, a […]

Read full article at https://wccftech.com/3d-x-dram-hits-proof-of-concept-hbm-replacement-10x-density-high-yield-memory/

As AI computing capacity continues to grow, an expert from computing infrastructure provider Nebius sat with AlphaSense to describe the state of the industry. While NVIDIA's leading-edge AI GPUs remain the top in the industry when it comes to performance, the expert believes alternatives are growing in popularity, particularly as the industry shifts its cost metrics. The demand for AI computing capacity also remains high, as providers can easily run at 100% utilization rates to drive down costs and earn the most from their investment. Alternatives To NVIDIA Chips Grow In Popularity As Industry Shifts Towards Cost Per Million Tokens […]

Read full article at https://wccftech.com/nvidias-ai-chips-see-alternatives-emerge-amidst-pricing-model-shift-to-cost-per-million-tokens/

OpenAI is on a mission, an incredible mission, you might say, to scale up its AI compute capacity to 30GW by the end of this decade. 30GW By 2030: OpenAI Sets Ambitious Goals, but Will The Semiconductor Industry Be Able To Supply This Much Compute? Yesterday, Amazon and Anthropic announced that they were going to bring up 6GW of AI compute capacity by the end of this year. But their primary competitor, OpenAI, has more ambitious & mighty plans. Today, OpenAI announced that since the launch of ChatGPT, its revenue and compute resources have scaled at a rapid pace. This […]

Read full article at https://wccftech.com/openai-sets-goal-to-scale-up-ai-compute-capacity-to-a-whopping-30gw-by-2030/

Global wafer foundry revenue in 2026 is estimated to grow 23.5% year-over-year, expected to reach US$250 billion.

One of the first recipients of TSMC’s 2nm ‘N2P’ process will be the Dimensity 9600 Pro, enabling MediaTek to move over to a new CPU cluster while also rumored to target higher 5.00GHz clock speeds for significantly improved single-core and multi-core performance. Now, one tipster has shared the first Geekbench 6 scores belonging to the chipset, with the results being up to 25 percent faster than the A19 Pro, Snapdragon 8 Elite Gen 5, and the Dimensity 9600 Pro’s immediate predecessor, the Dimensity 9500. At a 5.00GHz clock, the Dimensity 9600 Pro has already been pointed out to overheat, suggesting that commercial […]

Read full article at https://wccftech.com/dimensity-9600-pro-prototype-25-percent-faster-than-a19-pro-and-snapdragon-8-elite-gen-5/

The 2nm node is expected to be the most cutting-edge lithography the industry has ever laid eyes on, with none other than TSMC leading from the front with orders secured for AI chips and mobile SoCs. Unfortunately, the complexities of mass-producing these advanced wafers at a monumental scale pose a major challenge for multiple smartphone companies planning to use these chipsets in their devices. In fact, one tipster suggests that TSMC will partly be responsible for firms introducing silicon downgrades to their handsets later this year, with the other culprit being the DRAM crisis. The advanced 2nm process has ‘insufficient […]

Read full article at https://wccftech.com/tsmc-to-force-smartphone-makers-to-make-chipset-downgrades-this-year/

The race to achieve chipset autonomy for Samsung began with the Exynos 2600, which is found in various Galaxy S26 and Galaxy S26+ models, and is the company’s first 2nm GAA SoC. Needless to say, the Korean giant has made up its mind to put all of its faith in Qualcomm and has decided to jumpstart its silicon division to ensure that it doesn’t need to rely on any other firm for this specific part. However, the Exynos 2600 would likely not have materialized if Samsung’s incessant mistakes had not led to a $3 billion loss last year in the […]

Read full article at https://wccftech.com/exynos-2600-was-only-made-because-samsung-lost-3-billion-in-2025/

The timing of the Snapdragon X2 Elite could not have been more perfect, as the release of a significantly powerful chipset to take on Apple’s M5 series and x86 CPUs was only matched by how the Windows on ARM platform improved with more native apps available for the operating system. In short, developers are finally recognizing that Qualcomm’s latest and greatest SoC offers a decent footing for them to make the required effort towards optimizing their programs. Unfortunately, in the latest discussion on Reddit, OEMs using Snapdragon X2 Elite in their machines are targeting short-term gains rather than joining hands to […]

Read full article at https://wccftech.com/snapdragon-x2-elite-solid-ecosystem-foundation-but-oems-risk-future/

The Snapdragon 8 Elite Gen 6 and Snapdragon 8 Elite Gen 6 Pro are expected to be exclusively mass produced on TSMC’s 2nm ‘N2P’ process, making it yet another year where Qualcomm and Samsung cannot come to terms on becoming partners. To be fair, it’s not the San Diego firm’s fault, as a new report states that even the Korean giant’s 2nm GAA node is running into stability issues, to the point that the technology cannot reach the desired benchmark. These limitations have apparently created a rift between the two companies, preventing them from forming a long-term foundry business. With the […]

Read full article at https://wccftech.com/samsung-2nm-process-not-satisfied-qualcomm-quality-benchmark/

Companies like Apple, Qualcomm, and MediaTek have benefited greatly from extending their partnership with TSMC, as not only have they developed the most cutting-edge smartphone chipset technology, but the level of performance is uncanny and would likely not be possible without the Taiwanese semiconductor behemoth providing its advanced nodes. Later this year, we shall bear witness to SoCs reaching clock speeds of 5.00GHz, making it the first time in history, with Huawei sadly unable to keep up with its rivals as it doesn’t have access to the foundry giant’s technology. Huawei didn’t jump into EUV lithography fast enough, as its […]

Read full article at https://wccftech.com/tsmc-advanced-lithography-allows-smartphone-socs-to-reach-5ghz-huawei-missing-out/

The Fusion Architecture that Apple introduced with the M5 Pro and M5 Max allowed the company to scale the limitations of chipset development, packing in more CPU cores operating at higher clock speeds while maintaining impressive efficiency. This leaves only the M5 Ultra to be announced, but unlike the other SoCs, there are few details surrounding its architecture. Fortunately, a series of logical explanations indicate why Apple will bring its UltraFusion process to this silicon, just like it has with the M3 Ultra. However, this approach also means that we’ll witness the pairing of both Fusion Architecture and UltraFusion to form […]

Read full article at https://wccftech.com/m5-ultra-to-use-ultrafusion-process-to-combine-two-m5-max-dies/

The Exynos 2600 may have ushered in the first step towards Samsung obtaining self-reliance in smartphone chipset development, but there is still significant room for improvement, as issues lingering with the company’s first 2nm GAA SoC could be addressed by its immediate successor, the Exynos 2700. This may explain why the latter has been spotted in the latest Geekbench listing, indicating that Samsung wishes to put the silicon through some early paces to introduce better optimization ahead of its launch in the Galaxy S27 series. A 10-core CPU cluster is expected to remain unchanged for the Exynos 2700, but Samsung may […]

Read full article at https://wccftech.com/exynos-2700-spotted-on-geekbench-again/

MediaTek may have conceded its plans to bring only a single chipset to its Dimensity 9600 lineup because a new rumor claims a ‘Pro’ version will arrive later this year, likely as a direct counter to Qualcomm’s Snapdragon 8 Elite Gen 6 Pro. Like its closest rival, a tipster states that the SoC is bringing desktop-level performance with a target frequency of 5.00GHz and a new CPU cluster. The only catch is that MediaTek has yet to find a way to cool it properly. New target frequency for the Dimensity 9600 Pro may only be in short bursts, as overheating problems plague the chipset, […]

Read full article at https://wccftech.com/dimensity-9600-pro-bringing-desktop-level-performance-with-5ghz-clock-speed-target/

One of the ways that enabled Apple to get to that $599 price point for the MacBook Neo was using binned, or leftover A18 Pro chipsets, which feature a 5-core GPU instead of the 6-core configuration used on the iPhone 16 Pro and iPhone 16 Pro Max. With its attractive entry-level positioning, the new portable Mac would undoubtedly be picking up steam, just as long as Apple has the available parts, which, according to the latest report, it doesn’t. Now, the technology giant is stuck tackling a conundrum: ask its suppliers to resume production or stick with the current shipment tally and […]

Read full article at https://wccftech.com/apple-faces-macbook-neo-production-dilemma-as-a18-pro-supply-runs-dry/

The sub-3nm expansion plans for Samsung are progressing somewhat smoothly, with its 2nm GAA yields estimated to be at 60 percent, giving the Korean foundry sufficient confidence to proceed with trial production of its next-generation lithography at its Taylor plant. Initially constructed for 4nm wafer manufacturing, Samsung has previously been reported to transform the overseas facility into a 2nm hub, giving it an edge over TSMC, which has limited its most advanced technologies from reaching U.S. soil. TSMC limiting its involvement to mass produce 2nm chips in the U.S. gives Samsung a massive opening with its Taylor plant, as it’s testing […]

Read full article at https://wccftech.com/samsung-taylor-plant-2nm-trial-production-begins/

The Apple Silicon continues to be efficient enough to the point that chipsets like the M5 can be used in portable Macs like the newly updated MacBook Air lineup for the majority of tasks without worrying about performance drops. However, when it comes to sustained workloads, which matter greatly, like gaming, an active cooling fan continues to tower over those silent passive solutions because the latest comparison shows that titles like Cyberpunk 2077, Baldur’s Gate 3, and others can lose up to 40 percent in framerates because of the cooling design difference alone. An active cooling fan gives Apple’s M5 sufficient […]

Read full article at https://wccftech.com/apple-m5-loses-up-to-40-percent-in-fps-due-to-poor-passive-cooling/

Taiwan foundry revenue growth expected to reach 30% 2026; strategic adjustments initiated for mature processes.

The Exynos 2600 is Samsung’s first 2nm GAA chipset, with the advanced lithography promising better performance and efficiency over the company’s older manufacturing processes. However, those claims are busted in the latest benchmark comparison, where the chipset ends up reaching 30W of peak power draw, making it equivalent to what some notebooks consume. It is also disappointing to see that the Snapdragon 8 Elite Gen 5 is significantly more efficient than its rival, proving that TSMC’s foundry is ahead of its rivals. A Decompression Test shows the Exynos 2600 requiring 63 percent more power while completing the task slightly longer than the […]

Read full article at https://wccftech.com/exynos-2600-efficiency-questioned-30w-peak-power-draw-in-geekbench-6/