News
CAPACITY TEST:

Intel Dunlow Manifests Point To 28-Core Nova Lake-S Xeon Platform

Newsroom brief

Tom's Hardware reported that NBD shipment manifests point to an Intel Dunlow workstation and entry-server platform with up to 28 cores, LGA1954 packaging, dual-channel memory and 95W processor base power. The details remain pre-launch documentation, and Intel has not confirmed launch dates, pricing, core mix or customers.

Verified against source materialEdited by SendTech Times Chips & Compute Desk
Intel Dunlow Manifests Point To 28-Core Nova Lake-S Xeon Platform
Image source: Tom’s Hardware

Shipment manifests are pointing to an Intel Nova Lake-S workstation and entry-server platform called Dunlow, but the material is still pre-launch supply-chain documentation rather than a final Intel product announcement.

Tom's Hardware reported that NBD database entries located by @x86deadandback list a platform with up to 28 cores, an LGA1954 socket, dual-channel memory and a 95W processor base power.

The article described the parts as Xeon E-class Nova Lake-S processors for entry-level server and workstation applications.

NBD Manifests List A 28-Core Dunlow Platform

Tom's Hardware said the manifests describe Dunlow as a successor to Intel's Catlow platform, which used Xeon 6300P-series processors.

The NBD-listed Nova Lake-S workstation/server configuration has up to 28 cores and dual-channel memory rather than the wider memory subsystem associated with higher-end Xeon platforms.

Tom's Hardware said the shipment data also lists an LGA1954 form factor.

The LGA1954 package separates the alleged Dunlow configuration from existing desktop and server sockets, while the available evidence still comes from shipment manifests rather than Intel launch material.

Desktop Nova Lake-S Claims Reach 52 Cores

Tom's Hardware contrasted Dunlow with rumoured desktop Nova Lake-S platforms.

Tom's Hardware said those desktop chips allegedly reach up to 52 cores, including up to 16 Coyote Cove performance cores, up to 32 Arctic Wolf efficiency cores and four low-power Arctic Wolf cores in the system-on-chip tile.

Tom's Hardware also said the enthusiast desktop versions reportedly pull up to 474W.

The NBD-listed 95W processor base power points to a different target: lower-cost workstations and light server systems rather than maximum desktop performance.

Core Mix Is Still Unconfirmed

Tom's Hardware treated the internal core mix as unresolved.

The report said a 28-core Nova Lake-S processor for Dunlow could be an unusual all-performance-core design, a derivative of another Xeon die adapted to LGA1954, or a different configuration using Intel's tiled architecture.

Tom's Hardware noted that current Xeon 6300P-series Raptor Lake-E processors top out at 12 performance cores.

Tom's Hardware also said 28 cores do not naturally follow from a 16P+32E desktop design or an 8P+16E notebook design, though those details remain part of the source's pre-launch analysis.

Two Memory Channels Would Lower Platform Cost

Tom's Hardware said two DDR5 memory channels could be enough for storage, web-hosting and other server applications that do not require very high memory bandwidth.

That would place Dunlow between high-end desktop CPUs and more expensive Xeon platforms with wider memory subsystems.

The article also linked the possible product gap to Intel's broader workstation and server roadmap.

Tom's Hardware said Xeon 6 server and workstation CPUs can start at 16 cores with octa-channel memory, while cancelled Diamond Rapids processors would have used an octa-channel memory subsystem.

Intel has not announced Dunlow, confirmed the Nova Lake-S Xeon E-class configuration, named launch dates, disclosed pricing, published official core mixes, or identified workstation and server customers for the platform.

Share this article
inXf

Related articles

More
Samsung Starts PM1763 PCIe 6.0 SSD Production For AI Servers
Chips & Semiconductors

Samsung Starts PM1763 PCIe 6.0 SSD Production For AI Servers

Samsung has started mass production of the PM1763, a PCIe 6.0 enterprise SSD for AI and HPC servers with 28,400 MB/s sequential reads in a 16TB configuration, according to ServeTheHome. The report did not name customers, pricing, shipment volumes or server qualification dates.

AMD Linux Patch Adds Low-Power CPU Core Support For Future Chips
Chips & Semiconductors

AMD Linux Patch Adds Low-Power CPU Core Support For Future Chips

AMD has submitted Linux kernel patches that add a low-power CPU core classification beside performance and efficiency cores, but the company has not disclosed the first processor, launch date or product line that will use the new core type.

Arm and Supermicro Put Agentic AI Servers to a CPU Test
Chips & Semiconductors

Arm and Supermicro Put Agentic AI Servers to a CPU Test

Supermicro has introduced new server platforms built around Arm’s AGI CPU for inference-heavy and agentic AI workloads across cloud, enterprise and edge deployments. Arm says the AGI CPU includes up to 136 Arm Neoverse V3 cores, 12 DDR5 memory channels running at up to 8800 MT/s and PCIe Gen6 connectivity within a 300W power envelope. The key test is whether operators can use these CPU-heavy designs to add inference capacity without creating new pressure on power and cooling.

Intel 18A-P Enters Risk Production, but Foundry Proof Still Depends on Yield
Chips & Semiconductors

Intel 18A-P Enters Risk Production, but Foundry Proof Still Depends on Yield

Intel has started risk production of its 18A-P node, adding performance and power claims to its foundry pitch, while outside-customer commitments, Arm manufacturing proof and packaging capacity remain the next tests.

Memory Prices Push US PC Shipments Down 7%
Chips & Semiconductors

Memory Prices Push US PC Shipments Down 7%

Omdia data cited by Tom Hardware showed US PC shipments fell to 15.8 million units in the first quarter of 2026, as memory and storage chip shortages hit entry-level laptops and pushed the market toward a forecast 14.4% contraction.

FuriosaAI and Broadcom Target the Next Layer of AI Inference Infrastructure
Chips & Semiconductors

FuriosaAI and Broadcom Target the Next Layer of AI Inference Infrastructure

FuriosaAI said it will work with Broadcom on a next-generation AI inference platform built around its TCP architecture and Broadcom networking and packaging technologies. The planned third-generation accelerator will use a 2-nanometer compute die, HBM4/HBM4E memory and multi-die packaging, with sampling planned for the first half of 2028. The deal points to AI infrastructure competition shifting from single-chip performance toward memory, networking, power efficiency and rack-level system design.

Keep Reading

More Stories

Latest
MAS Proposes Three-Week Route For Repeat Singapore Retail FundsCapital & PolicyJul 10, 2026MAS Proposes Three-Week Route For Repeat Singapore Retail FundsMAS is consulting on an Alternative Funds Appendix for Singapore retail fund products. The regulator said most new fund categories could take about three months to assess, while later funds of the same type could be authorised within three weeks.AWS GraphRAG Deployment Claims 87 Percent Drug Research Cycle CutAIJul 10, 2026AWS GraphRAG Deployment Claims 87 Percent Drug Research Cycle CutAI News reported an AWS GraphRAG deployment that cut drug research cycles by 87 percent and moved six-month discovery work to three weeks. The account did not name customers, publish benchmark methodology or disclose Bedrock token costs.Humain Sets Aside 50 Megawatts Of Saudi Compute For Cohere ModelsCloud & Data CentersJul 10, 2026Humain Sets Aside 50 Megawatts Of Saudi Compute For Cohere ModelsHumain will designate at least 50 megawatts of dedicated compute capacity for Cohere foundational models, the companies said. The Saudi-backed AI company expects the expansion to be live by late 2027, but the announcement did not disclose the site, power supplier, pricing or hardware mix.Cerebras Plans 8x To 10x Manufacturing Scale-Up For AI InferenceChips & SemiconductorsJul 10, 2026Cerebras Plans 8x To 10x Manufacturing Scale-Up For AI InferenceCerebras chief executive Andrew Feldman said the company plans to scale manufacturing capacity by 8x to 10x this year and claimed its systems can run inference 10, 15, 20 or 30 times faster than GPUs. The interview-led source named customers including AlphaSense, Cognition AI, OpenAI, Block and GlaxoSmithKline, but did not include third-party benchmark methodology.Swift Lines Up 17 Banks For Tokenised Deposit Ledger TestsFintech & Digital PaymentsJul 10, 2026Swift Lines Up 17 Banks For Tokenised Deposit Ledger TestsSwift said 17 banks are preparing to test live transactions on a blockchain-based shared ledger for tokenised deposits. The system is meant to support round-the-clock cross-border payments, while final settlement remains tied to existing payment systems.OpenAI GPT-5.6 Rollout Meets White House Denial After Clearance AccountCapital & PolicyJul 10, 2026OpenAI GPT-5.6 Rollout Meets White House Denial After Clearance AccountA Commerce Department clearance account put OpenAI's GPT-5.6 on track for broader release, but the White House told CNBC it had not given any approval or green light and OpenAI has not announced a general-availability date.DP World Adds 700 Trucks To GCC Freight Network Without EV TimetableEconomyJul 9, 2026DP World Adds 700 Trucks To GCC Freight Network Without EV TimetableDP World said it has acquired 700 trucks for its GCC road freight network and said the fleet can support up to 35,000 truck trips a month. The expansion adds regional freight capacity, but the company did not give a timetable for green-energy vehicles or disclose the investment value.Malaysia Chip Outlook Rides AI Demand But Advanced Packaging Gap RemainsChips & SemiconductorsJul 9, 2026Malaysia Chip Outlook Rides AI Demand But Advanced Packaging Gap RemainsMaybank, Kenanga, MBSB and TA Securities said AI and data centre demand is supporting Malaysia’s technology sector, while Kenanga said exposure to higher-value advanced packaging remains limited. TA cited MYR 59.9 billion in semiconductor-related investments under the National Semiconductor Strategy.Google Adds Gemini Agent To Search Ads In India BetaAIJul 9, 2026Google Adds Gemini Agent To Search Ads In India BetaGoogle launched Business Agent for Leads in India as a Gemini-powered Search ad format that can chat with users on the search page. The company cited $77.25 billion in first-quarter ad revenue and several India ad metrics, but did not disclose pricing, wider rollout dates or independent lead-quality validation.Bitcoin Suisse Wins ADGM Licence For Institutional Crypto ServicesCapital & PolicyJul 9, 2026Bitcoin Suisse Wins ADGM Licence For Institutional Crypto ServicesBitcoin Suisse has received an ADGM Financial Services Permission for its Abu Dhabi subsidiary BTCS (Middle East) Ltd., allowing custody, approved virtual-asset trading and portfolio management for institutional and professional investors. The firm cited $3.7 billion in crypto assets under custody, but the announcement did not name UAE clients, launch volumes or retail availability.Planno Takes Solar GeoAI To 16 Markets Without Investment AmountAIJul 9, 2026Planno Takes Solar GeoAI To 16 Markets Without Investment AmountDubai-based Planno says Incubayt backing helped expand its solar GeoAI platform into 16 markets, while its Dubai rooftop analysis lists 4 GW of unbuilt capacity and no disclosed investment amount.Open USD Stablecoin Draws 140 Backers But Launch Timing Stays OpenFintech & Digital PaymentsJul 9, 2026Open USD Stablecoin Draws 140 Backers But Launch Timing Stays OpenOpen Standard launched Open USD with backing from banks, card networks and fintech firms, while the consortium said the stablecoin will go live later this year without naming commercial volumes or merchant adoption targets.