Beginning a Spatial Journey

Composing this article from Mt. Hood on Apple Vision Pro

My mind was racing as I assertively tugged on the large glass door of my local Apple Store. Despite arriving ten minutes early to my appointment, a friendly employee welcomed me to a demonstration area. The store was nearly empty, and the early morning light was beaming through the south-facing glass facade onto tables full of Apple hardware. As I was seated, a second employee sat a small tray containing my Apple Vision Pro on the table in front of me. The purpose-built tray satisfyingly cradled the contours of the headset. After learning how to pick it up without smudging the glass front, I gently lowered the headset into place. It was oddly anticlimactic, as if nothing happened. I could see the store just as it was a moment ago. Then I remembered that I had a computer in front of my face. I wasn't seeing the room – I was seeing a real-time representation of the room. The illusion worked!

Over 30 minutes, I littered the store with windows, browsed the web on Mt. Hood, and swam with sharks. At one point, a butterfly gracefully flew in front of me and briefly landed on my outstretched hand. As it did, I thought I felt the gentle tickle of its spindly legs. Of course, I didn't actually feel it. It's all a clever illusion, but it's convincing enough that my brain failed to register a difference. As expected for a first-generation product, there were occasionally small cracks in the illusion, but I left impressed.

The demo was a tightly curated experience that matched the device's strengths. How does Apple Vision Pro feel in actual use? Is it a productivity tool or merely a fancy media consumption device?

An Infinite Canvas

In a previous article, I mentioned that Apple Vision Pro did not appear to have a “killer app” at launch; however, I’m starting to think the killer app may be the VisionOS interface itself. The ability to effortlessly scale work beyond the boundaries of traditional displays is profoundly liberating. Apple Vision Pro frees the user from the constraints of the physical world. An infinite canvas is more than just a spatial characteristic; it's a metaphor for these technologies' seemingly boundless potential. Apple Vision Pro forces us to rethink our relationship with computers fundamentally and breaks new ground in workflow management.

Retina Resolution?

One of my primary concerns was resolution. Can Apple Vision Pro resolve text well enough to read for significant periods? While text doesn't appear quite as razor-sharp as it presents on the 5K Studio Display, it’s more than enough for a strain-free reading experience.

It's too early to determine how successful this product will be, but I can't help but feel that we're at the beginning of another pivotal moment in tech, where we fundamentally rethink its role in our lives. Today’s Vision Pro is a preview – a window into the future. The device is slightly heavy, the battery life isn’t what we’ve become accustomed to with Apple products, and visionOS contains a few irritating bugs. Despite these flaws, Apple Vision Pro offers a unique, visceral experience. My excitement for this product transcends the enthusiasm I felt at the release of the original iPhone. Apple Vision Pro reminds me of the wonder I experienced using my first computer, our IBM Personal System/2, as a child. My, how far we’ve come.

Preparing the Way

AI-generated image of a futuristic VR/AR headset using DALL-E 3.

In Fall 2015, Tim Cook proclaimed, "We believe the future of TV is apps." As it turns out, the future of television wasn't apps. Will the future of computing be spatial? On Friday, February 2nd, we will begin to find out when Apple Vision Pro, the company's first augmented reality headset, goes on sale.

Like most new platforms, Apple Vision Pro does not appear to have a "killer app." The future of Apple's new headset is far from assured. Apple needs third-party developers' cooperation to help secure the success of its new platform; however, the company's introduction of a Core Technology Fee (CTF) in response to the European Union's Digital Markets Act alienated a large swath of the developer community. Many popular streaming apps like Netflix, YouTube, and Spotify appear uninterested in developing a native VisionOS app.

Apple Vision Pro is the first headset that excites me as a user and developer. Despite the uncertainty, I ordered the new device and am actively developing a VisionOS app. Its emphasis on augmenting reality rather than escaping it provides a compelling user story and ultimate flexibility for developing productivity-focused apps.

I'm looking forward to documenting my initial experience this Friday and can't wait to see what we collectively do with it.

Scaling Silicon

The Apple M1 Ultra - Two M1 Max dies connected via an interposer

Apple raised eyebrows in 2020 when the company announced plans to transition from Intel processors to chips designed in-house, marking the end of a 15-year partnership with Intel.¹ For long-time followers of technology, it was reminiscent of Steve Jobs' announcement at the 2005 Worldwide Developers Conference (WWDC), where he revealed Apple's plan to transition from PowerPC to the x86 architecture from Intel. Like the x86 transition fifteen years earlier, the rollout of Apple silicon went astonishingly smoothly despite the fundamental incompatibility between x86 and ARM instruction sets. For the first time in the recent past, Intel, Advanced Micro Devices (AMD), and Apple have taken divergent strategies in microarchitecture design. Each strategy has its own strengths and weaknesses, so it will be fascinating to see how well each approach scales to the future's cost, efficiency, and performance demands. AMD's chiplet design offers pricing advantages over Intel at the expense of bandwidth constraints and increased latency. Apple's system-on-a-chip (SoC) strategy requires larger dies but offers complete integration; however, we may be seeing the first cracks in Apple's ARM SoC strategy after scaling back plans for a high-end Mac Pro.² According to Mark Gurman's reporting, a Mac Pro with an SoC larger than the M1 Ultra would likely have a starting cost of $10,000. To get a better perspective on the pricing challenges Apple may be facing when designing an SoC for the Mac Pro, let's explore how yield and cost change as die size increases.

For example, we can calculate the number of rectangular dies per circular wafer for Apple's basic M1 SoC and the M1 Max using basic geometry:

M1

Die Dimensions	10.9 mm x 10.9 mm
Die Size	118.81 mm2
Scribe Width	200 µm
Wafer Diameter	300 mm
Edge Loss	5.00 mm
Die Per Wafer	478

M1 Max

Die Dimensions	22 mm x 20 mm
Die Size	440 mm2
Scribe Width	200 µm
Wafer Diameter	300 mm
Edge Loss	5.00 mm
Die Per Wafer	117

The smaller M1 dies give us four times the quantity per wafer over the M1 Max. This is one factor influencing the cost of physical materials, but things get really interesting once we begin calculating yields. The process of fabricating working silicon wafers is delicate and rife with the opportunity to produce imperfections. Defects can be caused by contamination, design margin, process variation, photolithography errors, and various other factors. Yield is a quantitative measure of the quality of the semiconductor process and is one of the most important factors in wafer cost. The measure used for defect density is the number of defects per square centimeter. Assuming a standard defect density of 0.1/cm² using a variable defect size yield model for TSMC's N5 node, our two wafers possess vastly different yields:³

M1

Die Dimensions	10.9 mm x 10.9 mm
Die Size	118.81 mm2
Scribe Width	200 µm
Wafer Diameter	300 mm
Edge Loss	5.00 mm
Die Per Wafer	478
Defect Density 3	0.1 cm2
Yield	88.9%
Good Dies Per Wafer	425

M1 Max

Die Dimensions	22 mm x 20 mm
Die Size	440 mm2
Scribe Width	200 µm
Wafer Diameter	300 mm
Edge Loss	5.00 mm
Die Per Wafer	117
Defect Density 3	0.1 cm2
Yield	65.4%
Good Dies Per Wafer	76

This yield disparity increasing from 4x to just over 5.5x further inflates our larger die's already higher manufacturing cost. Just how much of a price difference? Extrapolating data from the Center for Security and Emerging Technology, we can estimate that a 300 mm wafer created using TSMC's N5 node costs just under $17,000.⁴ Therefore, our M1 has a theoretical materials cost of $40 while our M1 Max has a cost of $223. Given that an M1 Ultra is two M1 Max dies connected via an interposer, the raw silicon cost of the Ultra is likely around $450. While all these figures are nothing more than conjecture, they clearly illustrate how quickly costs skyrocket and yields shrink as die size increases.

Where does this leave the Mac Pro and the Apple silicon roadmap? Cost-effective silicon capable of performance significantly higher than the current top-tier SoC will likely require a more advanced lithography process to decrease transistor size. A likely candidate is TSMC's N3 node, which is where Apple is headed over the subsequent few product cycles. However, the rate at which manufacturers are able to decrease transistor size is rapidly slowing, as evidenced in the chart below, so a more fundamental rethinking of chip manufacturing is on the horizon.

TSMC. (2021, June). 3nm Technology. https://www.tsmc.com/english/dedicatedFoundry/technology/logic/l_3nm

One certainty is that we are entering an exciting period of technological advancement that is beginning to disrupt the market. The ability of technology companies to adapt quickly is shifting from a mere competitive advantage to a requirement for survival. The future belongs to those who dare to think without boundaries.

References:

¹ Gurman, M., & King, I. (2020, June 22). Apple-made computer chips coming to Mac, in split from Intel. Bloomberg.com. Retrieved December 26, 2022, from https://www.bloomberg.com/news/articles/2020-06-22/apple-made-computer-chips-are-coming-to-macs-in-split-from-intel?sref=9hGJlFio

² Gurman, M. (2022, December 18). Apple scales back high-end Mac Pro plans, weighs production move to Asia. Bloomberg.com. Retrieved December 30, 2022, from https://www.bloomberg.com/news/newsletters/2022-12-18/when-will-apple-aapl-release-the-apple-silicon-mac-pro-with-m2-ultra-chip-lbthco9u

³ Cutress, I. (2020, August 25). ‘Better yield on 5nm than 7nm’: TSMC update on defect rates for N5. AnandTech. https://www.anandtech.com/show/16028/better-yield-on-5nm-than-7nm-tsmc-update-on-defect-rates-for-n5

⁴ Kahn, S., & Mann, A. (2022, June 13). AI chips: what they are and why They Matter. Center for Security and Emerging Technology. Retrieved December 30, 2022, from https://cset.georgetown.edu/publication/ai-chips-what-they-are-and-why-they-matter

Further Reading:

Agrawal, V. D. (1994). A tale of two designs: the cheapest and the most economic. Journal of Electronic Testing, 5(2–3), 131–135. https://doi.org/10.1007/bf00972074