Tag: chip design

OM in the News: Apple’s Spectacular Failure

In seeking to build a key part for its new iPhones, Apple set out to design a chip that would allow it to cut ties with Qualcomm, a longtime supplier and bitter foe. But the new iPhone models just unveiled are missing a proprietary silicon chip that Apple had spent several years and billions of dollars trying to develop in time for the rollout, reports the Wall Street Journal (Sept. 25, 2023). 

Apple iPhone 15 Pro Max and a Qualcomm modem chip.

The 2018 marching orders from CEO Cook to design and build a modem chip—a part that connects iPhones to wireless carriers—led to the hiring of thousands of engineers. The goal was to sever Apple’s dependence on Qualcomm, which dominates the modem market. (Apple paid more than $7.2 billion to Qualcomm for chips last year). But recent tests found Apple’s chip was too slow and prone to overheating. And its circuit board was so big it would take up half an iPhone, making it unusable.

Apple hasn’t publicly acknowledged its modem project, much less its shortcomings.  Engineering teams working on Apple’s modem chip had been slowed by technical challenges, poor communication and managers split over designing the chips rather than buying them. Teams were siloed in separate groups across the globe. Bad news from engineers about delays or setbacks– leading to unrealistic goals and blown deadlines–were concealed.

Apple believed, first,  it could replicate the success of the microprocessor chips it designed for iPhones. Adoption of those chips fattened profit margins and improved performance for billions of devices. Second, Apple and Qualcomm had bickered and swapped accusations of lying and theft.

Apple had found that designing a microprocessor, a tiny computer to run software, was easy by comparison. Modem chips, which transmit and receive wireless data, must comply with strict connectivity standards to serve global wireless carriers. A brute force of thousands of engineers, a strategy successful for designing the computer of its smartphones and laptops, wasn’t enough to quickly produce a superior modem chip. Three years ago, Apple began replacing processor chips from Intel, used for years in Macs, with a proprietary chip that allowed its laptops to run faster and generate less heat. That Apple chip saved the company $75 to $150 on every computer.

Classroom discussion questions:

  1. Why was it harder to design a modem chip than a microprocessor chip?
  2. What did Apple do wrong?

OM in the News: Apple Gains Control by “Insourcing”

“Apple built its gadget empire by outsourcing production to a vast ecosystem of chip makers and other component specialists. It is now taking a lot of that business back,” writes The Wall Street Journal (June 24, 2020). The company, which released its first iPhone processor in 2010, plans to ship Macs this year with custom chips, a move that ends a 15-year technology partnership with Intel. (Intel stands to lose about $2 billion in laptop chip sales annually). Apple said the custom-designed chips are more efficient and offer higher-performance graphics.

The plan fits into Apple’s broader strategy of replacing many third-party parts with components designed in house. Apple’s built-for-purpose parts now account for 42% of the costs of core iPhone components, up from 8% five years ago. Custom components have cut costs, boosted performance and increased Apple’s control over future releases. The new Mac processors will shave $75 to $150 off the cost of that computer.

The strategy springs from Apple’s philosophy—fostered by Steve Jobs—that owning core technologies provides a competitive edge. Customized chips and sensors can help its iPhone, iPads and Macs leapfrog rivals in battery performance and features. It also can protect Apple from Chinese rivals that buy universally available parts. Apple relied on third-party components for years while it built the engineering depth and expertise it needed to design more components itself. Apple’s chip division has mushroomed over the past decade to thousands of engineers.

The initiative—called insourcing—can give Apple a 2-year jump on competitors in device performance because Apple can plan how multiple chips work together to limit power consumption and free up space inside iPhones and iPads for other components. Many companies continue to supply Apple, which provides substantial revenue, even as they fear Apple will start making the very components they provide it.

Classroom discussion questions:

  1. What are the reasons Apple chose to “insource?”
  2. How is Apple achieving competitive advantage through OM? (Hint: See pp. 36-39 in your Heizer/Render/Munson text).

OM in the News: Sharing the Same Production Process at Samsung and Globalfoundries

Two Globalfoundries workers in Albany, NY
Two Globalfoundries workers in Albany, NY

Samsung and Globalfoundries just announced (see The Wall Street Journal-April 18, 2014) that they have agreed to adopt the same production process as they upgrade their chip-manufacturing services, an unusual alliance with implications for many designers of computer chips and other devices, notably Apple. With the agreement, chips produced by Samsung and Globalfoundries will be essentially identical; companies that design chips could have their products produced in factories operated by either company with no extra effort.  Companies generally prefer to reduce their reliance on a single supplier for components. In this case, the pact between Globalfoundries and Samsung provides a new selling point as the two companies try to woo customers away from Taiwan Semiconductor, the biggest chip maker.

The new pact could allow Apple in the future to shift chip orders between Samsung’s Austin plant and a Globalfoundries factory near Albany, N.Y.  “The idea of doing business with multiple suppliers is built right into Apple’s DNA,”  says one industry expert.

The pact also reflects the intense financial pressures associated with pursuing Moore’s Law, Silicon Valley’s shorthand for shrinking semiconductor circuitry to improve chips’ speed and data storage capability. With individual production tools priced at tens of millions of dollars—and complete chip factories costing $5 billion or more—fewer and fewer companies still develop new production processes. In response, companies are now working together to share costs of developing new production recipes.

But the deal goes much further. Globalfoundries agreed to abandon a technology it had been developing for creating chips with circuitry measured at 14 nanometers, or billionths of a meter. It will instead license Samsung’s 14 nanometer process, which has technical benefits, and uses common production tools and materials.

Classroom discussion questions:

1. What are the benefits of shared production?

2. Why is Apple encouraging this concept?

                                         

OM in the News: New Product Design Keeps Intel a Generation Ahead

Chapter 5 discusses the critical importance of new product design in OM and the impact it can have in the technology arena. Indeed, as The New York Times (May 4,2011) breaks the news about Intel’s success in a 3-D transistor design, we see how important staying ahead of the fierce competition can be. Breaking away from a design (called the planar transistor) that has been a constant in the chip industry since 1959, Intel has found a way to make smaller, faster, and lower-powered chips.

Its designers are turning from 2-dimensional transistor switches (the basic building block of the information age) to a third dimension to find more room. Intel said yesterday that the new process allows chips to run 37% faster in low voltage applications (like iPhones and iPads), while power consumption drops 50%. 

This exciting news is significant because it means the world’s largest chip maker (see yesterday’s blog about the industry) is not slipping from the pace of doubling the number of transistors etched onto a sliver of silicon every 2 years, a phenomenon known as Moore’s Law. This “law”, named after Intel’s co-founder, has set the computer industry apart from other manufacturing because it continued to improve at an accelerating rate.

Intel currently uses  a photolithographic process to make a chip and the current generation is called a 32 nanometer chip. (By comparison a human red blood cell is 7,500 nanometers in width). “Intel is on track for  22 nanometer manufacturing later this year”, says the scientist in charge of the project. By 2015, the firm is on target to make chips in a 10 nanometer generation.

Being first out with the 3-D chip technology gives Intel a full generation lead over competitors, but it does not guarantee success in a fast-changing consumer products market where “ultra-low power” chips are critical in consumer products.

Discussion questions:

1. Why is Moore’s Law a critical part of the chip industry?

2. What is Intel’s gamble in developing the 3-D chip?