PC Ekspert Forum - View Single Post - Vijesti i novosti o Intel procesorima i čipsetima

The Exiled · 04.09.2024., 16:45

Citiraj:

Intel manufacturing business suffers setback as Broadcom tests disappoint

Citiraj:

Intel's contract manufacturing business has suffered a setback after tests with chipmaker Broadcom, dealing a blow to the company's turnaround efforts. The tests conducted by Broadcom involved sending silicon wafers - the foot-wide discs on which chips are printed - through Intel's most advanced manufacturing process known as 18A. Broadcom received the wafers back from Intel last month. After its engineers and executives studied the results, the company concluded the manufacturing process is not yet viable to move to high-volume production.

Intel 18A is powered on, healthy and yielding well, and we remain fully on track to begin high volume manufacturing next year," an Intel spokesperson said in a statement. "There is a great deal of interest in Intel 18A across the industry but, as a matter of policy, we do not comment on specific customer conversations." Broadcom spokesperson said the company is "evaluating the product and service offerings of Intel Foundry and have not concluded that evaluation."

Typically fabricating an advanced chip requires more than 1,000 separate steps inside a chip factory, or fab, and takes roughly three months to complete. Production success is determined by the number of working chips on each silicon wafer. Achieving a substantial yield is crucial to move to producing the tens of thousands or hundreds of thousands of wafers demanded by big chip designers. Broadcom's engineers had concerns with the viability of the process. Typically that refers to the number of defects on each wafer or the quality of the chips fabricated.

Betting on a new manufacturing process such as Intel's 18A is impossible for some smaller chip companies because doing so would require resources they do not have. Intel released its manufacturing tool kit for its 18A process to other chipmakers over the summer, Gelsinger said on an earnings call last month. The company plans to be "manufacturing-ready" by the end of this year for its own chips and begin high volume production for external customers in 2025, Gelsinger said. At an investor conference last week, he said there are a dozen customers "actively engaged" with the tool kit.

Izvor: Reuters

Citiraj:

Intel's poor stock performance could lead to its removal from the Dow Jones

Citiraj:

Back in the late 90s, Intel and Microsoft were the first major tech players to join the elite Dow Jones Industrial Average club. It was the heyday of the PC revolution that these two giants had largely ushered in. But the tides have turned over the last couple of decades. While Microsoft has soared to become the world's second-biggest company thanks to its booming cloud and AI businesses, Intel has been struggling. The chipmaker's market cap has now dipped below $100 billion for the first time since its peak in 2000. Its stock has plummeted nearly 60% this year alone, making it the worst performer among the 30 Dow components that make up the DJIA. The company also reported a $1.6 billion loss for the second quarter, causing shares to sink even further to their current $20 level.

With such a low stock price and a mere 0.32% weighted influence in the DJIA, analysts in a Reuters report are sounding the alarm that Intel's days in the index could be numbered. The Dow's selection committee keeps a close eye on the spread between its highest and lowest-priced components. When that gap exceeds 10x, they've historically given the bottom dweller the boot. Right now, healthcare behemoth UnitedHealth Group's lofty $580 share price is a whopping 29 times more expensive than Intel's. A few key missteps have contributed to this downfall, starting with the fact it failed to ride the AI wave. The company has also been bleeding market share in its bread-and-butter data center CPU business, and its massive investments in new manufacturing capacity look questionable given the cloudy prospects for its foundry efforts. In fact, an upcoming $32 billion factory in Germany, which is already facing delays, could be outright canceled.

Izvor: TechSpot

EDIT:

Citiraj:

Intel scraps 18A process for Arrow Lake, goes with 'external nodes' likely TSMC

Citiraj:

In what could easily be construed as a setback for Intel's foundry initiative and its vaunted plan to deliver five nodes in four years, the company announced today that it no longer plans to use its own 20A process node with its upcoming Arrow Lake processors for the consumer market. Instead, it will use external nodes, likely from partner TSMC, for all of Arrow Lake's chip components. Intel's only manufacturing responsibilities for the Arrow Lake processors will consist of packaging the externally produced chiplets together.

The announcement comes as Intel embarks on a vast restructuring in the wake of troubling financial results last quarter. The company continues to lay off 15,000 workers, among the largest workforce reductions in its 56-year history. The node change comes after Intel initially demoed a wafer of Arrow Lake processors fabbed on the 20A node at its Innovation 2023 event, which indicated the chips were already far along in the development cycle. At the time, Intel said Arrow Lake would come to market in 2024.

Intel says its crucial next-gen 18A node remains on track, and it has now shifted its engineering resources from Intel 20A to the newer 18A node. As such, it appears that Intel will now leapfrog over its 20A process entirely to avoid the capital expenditures required to bring the node to full production. Eliminating the always-eye-watering ramp costs of a new node, particularly one as advanced as 20A, will surely contribute to the company meeting its cost-cutting goals.

The Intel 20A node was never planned for many products due to the company's fast-track move to the more advanced 18A node, so building out an extensive network of fab equipment would have limited returns. However, Intel's 20A served as a vehicle for several new advances, like  RibbonFet Gate-All-Around (GAA) technology, which is Intel’s first new transistor design since FinFET arrived in 2011. It also marked the debut of the company's  PowerVia backside power delivery tech, which routes power for the transistors through the backside of the processor die.

Intel says the learnings it gained from its 20A node have contributed to the success of its 18A node, which makes sense given that 18A is a tighter refinement of the technologies invented for 20A. Intel again noted that it had reached a sub-0.40 D0 defect density (def/cm^2) for 18A, a critical measurement of the yield rate for a process node. A process node is usually considered production-worthy once the D0 reaches 0.5 or below.

Intel notes that it has chips built with the 18A process already powered on in the lab and booting operating systems and also touted that it has now delivered its critical PDK 1.0 to customers. This is the key framework that will allow external customers to build chips using Intel process nodes, a critical component of Intel's IDM 2.0 turnaround plan, which hinges on the company becoming an external foundry that produces chips for external customers.

Microsoft and the U.S. Department of Defense have already signed on to produce chips using the 18A node, and Intel plans to have eight tape-ins by the middle of 2025.

Izvor: Tom's Hardware

Jedino kaj fali u ovom Intel čušpajzu je da CEO Pat da/dobi otkaz i da onda država službeno krene u spašavanje Intela.

Intel 20A i 18A su zapravo dio iste nekadašnje 5nm obitelji, tak da, ako odustaju i otkazuju 20A, a pričaju da 18A ide bez problema (Broadcom se ne bi složil) - lažu sami sebe.

Tim više lažu sami sebe kroz ove PR budalaštine, jer se već više od pola godine (1 - 2) zna da budu Lunar i Arrow Lake na TSMC 3nm procesu, a sad kao odjednom otkazuju nekaj kaj u startu nije valjalo.