Tech.pinions

Thanks to the Covid-19 boost, demand for PCs and laptops has been surprisingly strong in 2020. At the beginning of this year, most PC forecasters suggested that unit sales of PC’s and laptops in 2020 would be down anywhere from -3% to -5%.

It now looks like sales of PCs and laptops could be in the positive next year slightly and total 270-280 million PCs While this demand is welcome, the big question is, will this type of growth continue?

Most of the primary industry players believe that demand will continue to be strong. Businesses will continue to buy updated PCs and laptops for those forced to work at home at least through mid-year. However, demand for major IT refreshes seems to be soft for 2021, making accurate forecasting of 2021 PC demand difficult at this time.

Thanks to some research I am doing with a former CRM consulting colleague, I was able to talk to three high-level IT executives over the last ten days. Our calls were related to the research we are doing, and being the opportunist I am, I snuck in a question about their IT spending in 2021 and thoughts about 2022 PC purchases.

All three confirmed that Covid had changed a lot of their PC buying strategies since they have had to buy and manage new laptops purchased for work-at-home staff. A lot of those purchases were driven by how old the work-at-home user’s laptop is, and if it was over five years old, they upgraded them.

They also upgraded some younger laptops based on the quality of that laptop’s camera, but the lion’s share was to replace older ones with lower speed processors that could not handle more complex workloads from home.

Two of them stated that they wanted to be clear that these purchases were upgrades to meet current work from home needs and not corporate IT refreshes as in the past. These two IT decision-makers said that broader refreshes of older PCs and laptops were not in the cards for 2021.

Due to the economic instability and lagging effects of Covid-19 in the broader business world, any aggressive PC refreshes in 2021 are slim. On the other hand, all three suggested that if the economy stabilizes and starts growing by the end of 2021, they could begin to pursue more aggressive PC upgrades to mainstream IT staff the following year.

Currently, ~450 million PCs are still in use worldwide that are four years or older. That is a huge market opportunity for PC vendors to tap into in the future.

I found it interesting that these IT directors I spoke with are already looking at 2022. 2020 was a difficult time for them, and they seem to have a clear idea of their needs in 2021. Like many other IT directors, I suspect that they see a future need to do some aggressive personal computer refreshes in the not too distant future.

That said, a lot could happen with Covid-19 recovery and the world economy in 2021. If Covid-19 vaccines work and the world starts to get back to some sense of normal business rhythms by 2022, the PC makers and our industry need to be sure they are ready for possible robust enterprise demand in 2022.

Apple’s December Announcements
Early this morning, Apple made a few new announcements, keeping them on their monthly announcement cycle since late summer. Today’s news was AirPods Max and the availability of Fitness+. When I first heard the rumor about AirPods Max being announced on December 8th, I was initially skeptical because of the late release, mostly missing the holiday shopping cycle. The more I thought about it, the more it made sense they would be super high-end, which would not be the same type of holiday gifts as AirPods or AirPods pro would be. Given the ultra-end high creator, audiophile, target audience, this announcement coming in December will not make an impact but getting them to market matters most. $549 is a pricey solution, but audiophiles would know it’s hard to get a great pair of studio headphones for less than $500, and some of the top ones used by DJs and producers can cost over $2,000.

Commenters are going back and forth on the design, but as with all of Apple’s products, and their wearables, in particular, it is clear their design is meant to stand out and call attention to itself and the wearer. I keep this in mind when I think about Apple’s future glasses solutions and how they will likely follow this pattern and shoot for an iconic look.

As interesting as the AirPods Max headphones are, they will likely be reserved for the premium customer base where Fitness+, Apple’s fitness service now available December 14th, will likely see more widespread adoption. Fitness+ is one of the more interesting services I think Apple is pushing, given some of the philosophical similarities to a Peloton workout. The other thing that stands out with this service is Apple’s integration with Apple Music, iPhones, iPads, AppleTV, and Apple Watch. With this latest service, Apple is creating services in more areas of consumer’s lives and possibly working for a solution to touch all the potential parts of an Apple customer’s life.

Antitrust and Startups
You may have noticed a lot of news about different startups starting to explore or file to go public. There is a range of factors playing into this and one that is starting to cause the startup world to think long and hard about their investment strategy.

First, it is important to note most VCs prefer their startups to go public than get bought. This generally has a longer-tail of reward and upside. But many investors know exits are often their quickest way to return ROI on investment and sometimes the best thing for the startup and its team. The range of IPOs we are seeing is an indication of many of these businesses feeling the need to go beyond private capital to scale their company, and some have a great chance of succeeding in the public market. But the angle that got me thinking recently was how some startups and investors are starting to consider the regulatory challenges around antitrust as a potential barrier for more exits in their portfolio.

With antitrust scrutiny now circling, it does seem it will be harder for large-scale acquisitions to take place, and most startups are raising the kind of capital that will require a large scale acquisition in order for it to provide a return to their investors. This makes for a challenging new dynamic as investors will now have to take this into consideration as they are looking to deploy capital.

All of this to say, a healthier, more balanced venture system is one that I think will do both investors and startups well. I’ve never liked the need for a grand slam for a winning approach but rather an approach that favors singles, doubles, and triples more consistently. Simply meaning, the freedom to invest in a business that is at heart a good business, even if not “venture scale,” is a much more healthy investing philosophy. The challenge is, up to this point, those businesses didn’t do well in the public market. If that changes, and I hope it does then the entire definition of what a “venture scale” company is could change, and that would be a good thing, in my opinion.

Over the last two months, I have had the opportunity to work with and test Lenovo’s ThinkPad X1 Fold, the first truly folding laptop in the market. I actually got to see a prototype of the ThinkPad X1 Fold at a Lenovo customer event in Orlando, Florida in May of 2019. What I saw there was perhaps one of the most fascinating portable computing designs I have ever seen. The notebook has always held a specific passion for me as one of my earliest projects was helping IBM work on the first clamshell laptop they brought to market in 1986.

While laptops, especially over the last 10 years, strived to be faster, smaller, lighter, and with great battery life, the actual clamshell design has had very few fundamental design changes. We did get 2-in-1’s, where the screen is detachable so it could be used as a tablet, but most still end up being some type of a clamshell design in the end.

With the Lenovo X1 Fold, this China-based PC maker has pushed the laws of physics with its clever hinge designs that are patented. They integrated a folding screen in a mobile form factor that is more like carrying a book, than any type of portable computer.

After two months of using the Lenovo ThinkPad X1 Fold, here are some of my observations about this product, as well as my thoughts about the future of folding laptops.

1-Impressive design. One cannot look at the Lenovo ThinkPad X1 Fold and not be impressed with its design. When you open it up from its book-like form factor, the screen folds out to become a 13 ” screen that has a stand behind it allowing it to sit upright in front of a person as a normal laptop screen does today. The keyboard sits in the center of the fold when the screen is folded. Once the screen is in place, it charges while sitting in the center of the fold and can be taken out and used for typing input. One big issue with this keyboard is that it is half the size of a full travel keyboard; so, getting used to typing on a smaller keyboard was problematic for me with my chubby fingers.

2-The screen. The screen itself is not a high-resolution screen but offers crisp text and images and is very readable. The key reason for a lower screen resolution has to do with the folding screens being made by companies like BOE and others who perfected folding screens using lower resolution screen technology. They promise the next versions could support higher resolution screens, but the first generation of folding smartphones and laptops use the best possible folding screens available at the time.

4-It uses standard Windows 10. One of the bigger promises of any folding device is having an OS that can support folding dual screens and is optimized for the kind of multitasking that should come with folding designs. Microsoft is working on a version of Windows called Windows 10X for foldable devices, but it was not ready for Lenovo to use in their first folding laptop.

When the screen is unfolded and placed to use as a normal Windows laptop, the ThinkPad X1 Fold works exactly like a normal laptop, albeit with a smaller keyboard. This is the way I used the X1 most of the time and, at least for me, this was more like a normal laptop experience I am used to using daily.

When in the slightly folded mode where you can have dedicated screens and apps to use, Windows 10 is not designed for this type of function. Yes, you can do multitasking and have multiples screens open on a Windows 10 computer, but in a foldable device, Windows 10 is not optimized yet for this type of form factor.

5-Battery life. The ThinkPad X1 fold’s battery is quite limited. At best I got about three hours of continual use. If I wanted to watch a streaming movie, battery life was just over two hours.

6- The X1 Fold has a SIM card slot. Lenovo believes that this type of device will be one you carry with you everywhere and made sure to include a sim card slot for wide-area networking. It only supports 4G in this version, but future models will support 5G modems too.

The Lenovo ThinkPad X1 is a marvel of innovation and design. I found it to be highly portable and most of my experiences with it were positive. I am anxious to try it with Windows 10X, but for now, and by using it more as a laptop, than an optimized folding portable, I was pleased and surprised at its ability to meet my mobile computing needs.

The bigger question that eventually needs to be answered is if there is a market for folding portable computers?

I don’t think we can answer this question based on just one folding portable computer available today. Over the next 12 months, we should see at least two other big-name PC companies release some type of folding portable that would compete with the ThinkPad X1 Fold.

We are asking a similar question about folding smartphones. We are also too early in the folding smartphone market to answer this question too. Both folding laptops and folding phones are in the highest price range in each of their categories. That means only high-end enthusiasts and ultra-early adopters will even buy these first versions.

The biggest takeaway from my experience with the ThinkPad X1 Fold is that since laptops hit the market back in 1985, we are seeing radical designs that break the clamshell mold. Thanks to folding screens, breakthroughs in hinge design, new battery technology being created for foldable of all types, PC makers are gaining a new toolbox of components that make it possible for them to innovate in mobile devices of all types.

While we may not have the proper business cases yet for folding portable computers, I truly hope that the PC makers keep the drumbeat of innovation going in mobile computing.

Salesforce + Slack
The news is official that Salesforce is acquiring Slack for ~$28b, which is roughly a ~25x multiple of Slack annual revenue projection for 2021. What has been extremely interesting to observe with this deal is how negative financial analysts have been on the deal. I’ve read four notes from large institutional banks, and all of them don’t like the deal. My hunch is they all think ~25x revenue multiple is too rich, which is part of the negativity, but there is another common thread of negative sentiment that strikes me.

Overwhelmingly, Salesforce financial analysts (meaning financial analysts whose job is to provide notes on Salesforce) seem to be fuzzy on the synergy Slack brings to Salesforce. And, to go one step further, many of them are not grasping Salesforce moving further from its CRM roots. It was around 2016 and subsequent years with acquisitions like Quip, Mulesoft, Tableau, etc., that the street began to notice Salesforce pull away from their CRM roots, and this seems to cause some confusion. Looking at it this way, with Slack, Salesforce would now become a market leader for software in sales, customer support, marketing, eCommerce, data analytics, application integration, and collaboration software.

My take on the negative read from the street is they are looking solely for the synergy of these acquisitions to CRM (where little synergy exists) but are not taking into account parallel revenue streams to offer Salesforce’s large customer base. From my perspective, Salesforce is smart to not only want to diversify their revenue streams but to be able to become as close to a complete solutions provider for their customers. Salesforce has not only an extremely large customer base but also a very good, well Salesforce, and the customer engagement opportunity to upsell value-added services that their customers need to run effectively internally and externally puts them in an entirely different competitive solution than those companies just offering one piece of the pie.

Many readers would know I’ve commented before how Slack was unlikely to be able to survive on its own long-term in a world where enterprise software is increasingly becoming a bundled world. I always believed Slack would be acquired. I thought Google made the most sense, but understanding how Salesforce wants to become full suite this deal makes sense for Salesforce as well, even if the acquisition number is quite high.

The last point I’ll make here is how I find it interesting that Salesforce’s investor relations team is not doing a great job helping investors understand the companies long-term vision. If most of their large institutional analyst followers are sending notes to investor clients saying they worry the company is going too far away from CRM roots, and this is clearly the case, then someone internally is not communicating this strategy and working to sell investors on the vision and upside. I track a lot of companies who do investor relations remarkably well, and I sometimes forget how poor some companies can be as well as this practice.

Amazon AWS Trend Setting
Amazon has had a busy week throwing a number of announcements out in relation to Re:invent. When I looked through a range of the announcements it was hard to not conclude that not only is Amazon the market leader when it comes to cloud providers but they also set the trends for the market. Essentially, where AWS goes in terms of features and technology, others will follow. Amazon is also extremely aggressive from a technology standpoint. Note the specific bullets from what I consider their most telling slide.

• Fastest networking with P4d instances (400 Gbps)
• Largest high memory instances for SAP (24TB)
• Largest local storage instances with D3en (336TB)
• Most powerful ML training instances with P4d
• Most powerful ML inference instances with Inf1
• Best pirce/performanice for graphics-intensibe workloads with G4ad
• Only cloud provider with on-demand macOS instances
• Only cloud provider that supports Intel, AMD, and Arm processors

I emphasized in bold the fastest, largest, most, best, and only to make the point of where Amazon is looking to lead but also trendsetting. Some of where Amazon goes is feature decisions, these can be copied, others are entirely strategic like macOS support and their own efforts with their Graviton processor for Arm instances. But it is hard to not see Amazon being extremely aggressive and extremely customer-centric with AWS.

It is interesting when reading Jeff Bezos note to investors and the language he uses to convey Amazon’s relentless focus on customer experience and the customer overall. Often when I read this I think about Amazon e-commerce and forget that this culture extends to AWS customers as well. If I was Microsoft and Google, I think this element would worry me the most when competing with Amazon.

The other point I want to make here, and this point is actually extremely relevant to both data center/cloud and client computing devices. What Amazon is emphasizing is a clear shift from general-purpose technology to specialized technology. This shows up in the diversity of underlying computing components AWS is adopting. The data center is less about just CPU/GPU and now more about specialized processors for specialized workloads. A good example of this is Amazon’s adoption of Intel’s Habana Gaudi AI Processors.

Nvidia has long argued the GPU is the ideal general-purpose computing tool for all non-CPU related tasks. For a long time, this was true but we are now seeing the move away from the GPU’s general-purpose back to specific purpose silicon for many of the tasks previously done on the GPU like AI training, inference, AI modeling, etc. I’m not saying there are not still good reasons to use the GPU for some of this modeling, but AWS adopting Habana Gaudi AI Processors which are specific purpose accelerators for AI training demonstrates how some tasks will move of the GPU because they will do these critical tasks better.

Broadly, the trend I’m talking about is the move away from general-purpose solutions to specific purpose technology, and as I said it applies to the data center/cloud as well as client devices. In client, Apple is the biggest champion and adopter of this trend as all their A-series chips and the new M-series chips for Mac contain a range of specific purpose silicon designs. One of Apple’s key performance differentiators is because they focus more on the total silicon solution and specific purpose components rather than just try architect a CPU/GPU to handle a broader range of tasks. Apple designs its chips with specific accelerators or co-processors that handle specific workloads better and let the CPU and GPU do what they are best at.

Honestly, I feel like this approach is the only way forward to compete in the modern digital world. This is a fundamentally different approach to computing solution design than the predominant one of the past 30 years. Many companies will struggle to adopt this approach or understand its full merits, but as I said, ones ability to compete will require embracing this move from general-purpose computing solutions to specific purpose ones.

Before I dive too deep into the point I want to make here about the challenges Intel faces, and it is worth calling out that my initial theory about Intel’s long-term position turned out to not play out. In 2016, I wrote this article titled Intel and the last foundry standing theory. My intro paragraph laid out what this theory was in a nutshell.

Coming out of the Intel Developer Forum (IDF), I thought I would share more on what I believe is Intel’s long-term thesis about themselves. Quite simply, I believe they feel they will be the last foundry standing with their leading-edge process technology. If you follow this as closely I do, you will note Intel and other foundries like TSMC and Samsung are in a race to get to the next process node. Today, these foundries manufacture at many different nodes, but 14 nanometers and 16 nanometers are the leading edge processes of today. With each jump to new process nodes (10nm being next and 7nm after that), designers can pack more processing power onto a single chip while still keeping a low power profile. The industry calls this “performance per watt” and the amount of performance increases with each new process while still maintaining a lower power voltage.

At the time of my writing of the article, Intel was not yet behind in process technology. And although they struggled to get to 14nm on the timetable they proposed, they were still the leading foundry. But moving from 14 to 10nm proved to be the challenge that sent Intel from leader to underdog. In the time it took Intel to get from 14nm to 10nm TSMC has successfully offered 16nm, 7nm, and 5nm process technologies. In the same time frame, Samsung has transitioned from 14nm, to 10nm, and currently 7nm EUV. Essentially, since 2016, each competing foundry has moved two full process generations ahead of Intel.

Intel should be fully moved to 10nm in 2021 but their process at that density has around 100 million transistors per millimeter square. Which is roughly the same amount of transistors per millimeter square as both Samsung and TSMC 7nm process.

This chart, which UBS created in a recent note on Intel, creates a compelling visual both in terms of process generation timelines and transistor density at each process.

This all ties into my initial thesis that Intel had always believed they would lead in process technology, and that is clearly no longer the case. That being said, the above chart highlights both the reason why Intel struggled for so long to get from 14nm to 10nm. The main reason always stated by Intel for the delay was because of their aggressive targets of transistor density for 10nm. As you can see, Intel’s 14nm had ~40m transistors per mm square, and both TSMC and Samsung’s 10nm had ~55 transistors per mm square. Intel’s target for 10nm was 100m transistors per mm square, which backs up the overly aggressive density target Intel was shooting for. Had they not been so aggressive, it is likely they would not be that far behind TSMC and Samsung.

Intel’s management and manufacturing group has been clear they will not be as aggressive on future notes, which bodes well for Intel executing on a more predictable timeline. But I do not believe they will catch up or surpass TSMC or Samsung.

One of the other elements of my theory was an acknowledgment of how difficult it would be and how much money in research, development, and patent innovation would be required to keep advancing manufacturing to new process technology. With that in mind, another chart I found in the aforementioned UBS report is a bit of a surprise.

The below chart maps the CapEx spend of Intel and TSMC over time.

One of the indicators most of us watched for was Intel’s CapEx spending as it related to buying manufacturing equipment in order to scale up for the next process development and manufacturing. As you can see, Intel had been spending the most in Capex and now is trailing TSMC, although both have been spending less than their peaks. While both are slowing slightly their annual CapEx spend, it is noticeable Intel is behind TSMC, which is telling in my opinion.

While the lower spend in CapEx could signal the less aggressive density targets or the long delays in process technology transition, at this point, Intel committing less and less to CapEx each year further convinces me that Intel will not catch up to other foundries, and further worries me that Intel can sustain their foundry business in the long-term.

Intel’s management, and the company as a whole, find themselves in a situation they have never been in before. They went from undisputed industry leader to now chasing the field as the underdog. In many ways, Intel was a dynasty that is facing a cross roads where decisions will need to be made that could dramatically alter its trajectory and some scenarios would leave us with an Intel so different from its roots for better or worse.

But the great thing about an underdog story is the chance of the upset. Intel does have a rolodex of great technology. They remain competitive still despite the shortcomings they have had the last five years. Process technology is not everything, but it still matters more than some companies want to admit. AMD has been gaining ground on Intel every quarter and in every category. Apple just released silicon for PCs that will likely set a new bar of expectations for notebooks and desktops in the years to come. In some ways, there has never been more excitement, or more ambition, or more validation for a computing architecture other than Intel’s standard of x86.

A well known venture capitalist Josh Wolf coined one of my favorite sayings “chips on shoulders put chips in pockets.” We can only hope that Intel has a massive chip on its shoulder with everyone counting them out. Because while we may not see Intel put chips in pockets, a competitive Intel is good for the industry and honestly good for the US as a country. It would be a terrible shame for no leading semiconductor company to be based in the US. Intel is down but they are still in the game and I do hope the adjustments they make at halftime will keep them competitive in the second half of the game.

Economic Recovery Update
I’ve been pouring through a number of economist reports that are modeling in the impact of a vaccine on the US economy in 2021 as well as different scenarios regarding the presidency and senate races. I’m not going to dive deep into all the different scenarios but I will take a look at the few that seem most likely.

At a high-level, every economist note establishes the baseline that this slump in the economy and GDP is dramatically different than past recessions even though numbers look similar. I made similar points early on when I was writing about the economic issues and weighing in on the debate of whether economic recovery would be quick or slow. What’s important to note about this recession/economic downturn from prior ones in the last 20 years is that average personal savings are much higher than before. So while people aren’t spending, it doesn’t appear in masse consumers don’t have the money they are simply not spending it. Yes, there are some job losses, and a key part of the recovery, stimulus, and vaccine will hopefully have an impact on employers hiring again to bring the unemployment rate down. But in general, personal savings among US consumers in this environment is much higher, and that gives economists hope that when they feel comfortable spending again they will, and perhaps spending quite a bit.

The scenarios I want to weigh in briefly are the US Presidential and Senate races, and a timeline for a vaccine. Biden has won the presidency and now the scenarios turn to the senate race. Economists universally agree the best scenario for a quick economic recovery is a Democratic sweep of the senate. The belief is a unified government will yield the largest economic stimulus package, multi-year spending plan, and the best chance of getting the virus past us in a timely fashion. Under this scenario, economists estimate real-GDP growth of 2.5-4% in 2021. In most models I’ve seen, a democratic sweep is the best scenario in terms of economic growth, government spending, investment confidence by businesses, etc.

The second scenario is a split senate and there is less confidence in a fast recovery under this scenario. Models in this scenario assume a smaller fiscal stimulus for COVID relief and impacts in spending commitments from governmental and business which would push any pre-COVID level recovery into 2022. The concern in this scenario is the continued falling of the unemployment rate which will only hurt GDP, consumer confidence, and spending and thus dramatically delay any type of recovery.

The other scenario working into these models is the early vs. late vaccine. Headlines provided positive work toward an early vaccine yesterday but there is still a long way to go. Consumer research being done by economists has continually shown a vaccine, not necessarily the elimination of the virus, is the single biggest thing holding consumers back from being confident to return to life as normal. Given the US GDP is driven by 70% of consumer spending any recovery has to assume consumers get back to life as usual.

In an early vaccine scenario, where distribution and availability take place in 2H21, estimates are consumer spending could accelerate and see a 3.4-4.5% GDP increase in the second half of 2021. In this scenario, businesses begin to have more confidence and the unemployment rate drops as businesses open and hires again, thus creating economic positives all around.

A late vaccine scenario is obviously a negative, with deepening unemployment losses, and a much later economic recovery being pushed into late 2023, and worries of lingering issues of consumer behavior, business declines, and more that may last even longer.

That being said, what seems obvious, is the best-case scenario we can hope for, with respect to the economy, is a Democratic sweep resulting in a unified government and an early vaccine in 2021.

Biden and China
A lot of questions have been raised about what the Biden win means for US relations with China. Most questions center on the trade war and the competitive nature of China and the US. Consensus thinking is Biden will not wind down the current trade war with China but may ease some restrictions in order to establish a better path to negotiate going forward. It also seems advisors to the Biden organization have a reputation for better dealing strategically with foreign relations and while we may still have a trade-dispute there is optimism collaboration between China and the US for things like battling COVID and climate change can help build bridges and ease tensions. The positive takeaway here for China-US relations is the new organization is likely to treat this more like a negotiation than a war.

No doubt Biden will be better than Trump in dealing with China, but it is still a tightrope walk given the tensions as a whole but even more so with Taiwan.

Overall, there is optimism international relations under Biden will improve and there is optimism that will create better long-term benefits for the economy.

As I covered last week, the Mac has room to grow as a business and take more share in the premium PC category where computers with an ASP of $800 or more sell approx 90m units a year. In contrast, Apple sells ~20-25m Macs a year. The Mac is on a growth trajectory this year, setting record sales last quarter at just over 6m units according to IDC, which is the first time the Mac has ever sold 6m or more in a single quarter. The Mac has been partially helped by COVID, and many individuals and families working and learning from home, and a greater need for each person to have a notebook arose.

As I’m sure, many of you followed Apple’s One More Thing event yesterday and saw Apple’s aggressive rollout for Apple Silicon in Macs and a new chip called the M1. Interestingly when I first saw the invite go out for the event and it was called One More Thing, I was intrigued. This phrase, made famous by Steve Jobs, typically was reserved for launching or highlighting something Apple thought was a big deal. When Steve Jobs would stay on stage, “and there one more thing,” you knew what was about to come was not something small or trivial but something significant. It is for this reason, and I think Apple’s ambitions for the Mac may be greater than many realize, and this is a much bigger deal. I think this visual from Apple on the Mac page says it all about their belief in how big this transition for the Mac is in actuality.

As we think about the transitions, the Mac has gone through from Motorola 68k to PowerPC to Intel, and now to Apple Silicon, each platform provided growth along the way. Arguably, shifting to Intel was highly strategic and a key reason Apple was able to grow the Mac business and be a viable consumer and corporate option. But even with Intel X86 designs, Macs still never crossed 10% of PC sales. The Mac is coming off its best year, which makes for an interesting time to begin to transition the platform over the next few years.

I will talk about performance and how that fits into Apple’s narrative in a bit, but I first want to acknowledge how most average consumers do not care so much about performance and specs or whose processor brand is in their notebook. Most consumers care about the features and functions of a product, and what it will allow them to do they can’t do with other products. For the Apple Silicon Powered Macs, I think the consumer value proposition will boil down to two things. The large app ecosystem, and better battery life.

Apple is bringing the whole of the iOS app ecosystem to Macs powered by the M1. This means day one, Macs running the M1 will have the largest collection of apps of any notebook/desktop platform on the market. Battery life is another key story, all though while M1 powered Macs will have the best battery life of any Mac, there are Qualcomm based PCs and some new Intel Evo powered Intel devices also touting 16-18 hours of battery life. Despite that, it is time the entire PC industry moved to better than all-day battery life and it seems we are one step closer to that reality.

Questions remain about how the new M1 Macs will work with the vast iOS ecosystem which had generally been designed with touch screens in mind, and how legacy software works using Rosetta 2, but I think Apple has to be confident in their own real-world tests to have brought out three new products running the M1.

The Purpose of Performance
During the presentation, Apple made some claims about the total computing performance of the M1 as well as the power efficiency of the architecture. Everyone understands why being power efficient matters, it means you can do more and get better battery life. The performance aspect is often overlooked in a world where notebook performance has been largely stagnant. The writers at Anandtech did a fascinating preview of the launch of Apple Silicon benchmarking Apple’s A14 chip to the current processors from Intel and AMD.

They showed a chart comparing the A14 to the latest Intel core I-7 chips and the latest release from AMD on the Ryzen 3 platform. Below that chart, which I encourage you to look at, they make the following conclusion.

The performance numbers of the A14 on this chart is relatively mind-blogging. If I were to release this data with the label of the A14 hidden, one would guess that the data-points came from some other x86 SKU from either AMD or Intel. The fact that the A14 currently competes with the very best top-performance designs that the x86 vendors have on the market today is just an astonishing feat.

The fact that Apple is able to achieve this in a total device power consumption of 5W including the SoC, DRAM, and regulators, versus +21W (1185G7) and 49W (5950X) package power figures, without DRAM or regulation, is absolutely mind-blowing.

With the A14 benchmarking as good, and in many cases better than the current offerings from Intel and AMD, the article seems to suggest that Apple can certainly achieve what they said as the fastest CPU in the world.

The article goes onto what is an incredible observation of Apple’s designs and architecture with their own custom silicon. This graph highlights the point the author then emphasizes.

During the release of the A7, people were pretty dismissive of the fact that Apple had called their microarchitecture a desktop-class design. People were also very dismissive of us calling the A11 and A12 reaching near desktop-level performance figures a few years back, and today marks an important moment in time for the industry as Apple’s A14 now clearly is able to showcase performance that’s beyond the best that Intel can offer. It’s been a performance trajectory that’s been steadily executing and progressing for years:

Whilst in the past 5 years Intel has managed to increase their best single-thread performance by about 28%, Apple has managed to improve their designs by 198%, or 2.98x (let’s call it 3x) the performance of the Apple A9 of late 2015.

Apple’s performance trajectory and unquestioned execution over these years are what has made Apple Silicon a reality today. Anybody looking at the absurdness of that graph will realize that there simply was no other choice but for Apple to ditch Intel and x86 in favor of their own in-house microarchitecture – staying par for the course would have meant stagnation and worse consumer products.

In this case, the performance benchmarks speak for themselves as to why this transition makes sense within Apple’s grand ambitions to truly separate the Mac from the rest of the PC industry. There is an important saying that if you give developers performance they will always use it. Apple has the most robust and thriving third-party development community in the world and their hope is that as they bring industry-leading performance to the Mac, those developers will create software that can only run on the M1 platform. And in the vein of one of Tim Cook’s favorite sayings, the Mac truly becomes something only Apple could create.

Word came down last week that Dish was going to discontinue Slingbox, the great technology created to watch your home TV while away from your home. Slingbox was the brainchild of Blake and Jason Krikorian, real visionaries who saw the streaming media future 13 years before it became mainstream.

Slingbox, like all great ideas, was born out of a pain point. It came about because Blake and Jason were avid SF Giant fans who were frustrated because they could not watch their beloved Giants when on the road. They decided that they should solve this problem and began in 2002 to try and figure out how to “sling” the Giants TV from their local TV station and make it available in a Web Browser or dedicated app.

Ben and I were very privileged to be able to witness the birth of Slingbox from the beginning. Very early in the Slingbox project, Blake, who I had known from his days at Microsoft, contacted us and shared his vision for what became Slingbox.

I, too, am a big SF Giants fan and loved their idea and was glad to give them advice on the project. While we never actually worked with them, we were brought into many early discussions and saw a seed of an idea grow into a successful technology launch that made Slingbox a big hit with techies and a broader consumer audience alike.

Slingbox came to market in 2005, and I was an early beta tester. That summer, I was in Beijing for meetings with a client and Chinese commerce officials. While back at my hotel, I decided to check out the score of the Giant’s game that was on back home at the time. I was not sure it would work given the restrictions on Chinese broadband networks, but to my surprise, I logged into my Slingbox app and began watching the Giants game in real-time. It was at that moment I realized that Slingbox was a game-changer and began to see that Blake and Jason had fundamentally invented the concept of TV anywhere.

I know that by 2005, we had seen early versions of user-created content given YouTube’s launch at about the same time. However, Slingbox introduced the world to the big promise of streaming commercial content that had much broader appeal. Slingbox was quite successful on their own but needed help for greater distribution. So in 2007, Dish Network purchased Sling Media, the company behind the Slingbox, for $380 million. This gave Dish the underlying technology they needed to build their more advanced streaming media platform they have today.

It was not a surprise to me that Dish has decided to, over time, discontinue Slingbox. There are still hundreds of thousands of users that use Slingbox, so their move to eventually close Slingbox servers will be gradually phased out over the next 24 months. Given our long association with Slingbox, Dish’s decision to kill Slingbox felt like a gut-punch to me.

One of their founders, Blake Krikorian, sadly passed away in 2016 at the young age of 49. However, his legacy as an industry pioneer will forever show that Blake and his brother Jason saw the future of streaming media and had a hand in shaping the streaming media market we all enjoy today.

In a meeting I once had with former Intel CEO Paul Otellini, he told me that Intel had to have all of their fabs running at full capacity for Intel to stay profitable. Intel is still profitable, largely due to solid ASPs and margins, but it appears not all the fabs are running at full capacity. When Otellini mentioned this data point to me, the PC demand was very strong. That year, the industry sold close to 400 million PCs. Today we sell only around 280 million PCs per year at best.

The good news for Intel is that while they are selling fewer PC chips, they are selling millions of chips for servers, which ultimately helps to keep the fabs humming even all fabs are not running at 100% capacity. One of the reasons I, and many others, think Intel should be more willing to help fabless companies produce their chips in Intel Fabs is that demand for non X86 processors is running high. Existing fabs, like the ones from TSMC, and Samsung, are almost at capacity, and they are getting new orders every day.

Every report I have read by semiconductor industry experts suggests that demand for new and advanced processors will continue to rise. Add dedicated AI chips, IoT chips, and new types of sensors and camera processors, and one can see how our current fabs could struggle to keep up with the demand. Then the is one other political concern that could be problematic with existing fabs that already exist.

I wrote an article a few weeks ago about the concerns around China and its aggressive interest in Taiwan. I mentioned that for the first time, I had heard real concerns from major Taiwanese execs that China could be moving faster than expected to bring Taiwan under its control. Since I wrote that piece, one other concern from top Taiwan execs is related to TSMC. Should China invade Taiwan, or at the very least, try and create a Hong Kong like international trade environment but under stricter China-based security rules, as they have recently done in Hong Kong, how would that impact Taiwan based business?

My sources in Taiwan have even asked what happens if China makes TSMC a Chinese company and asserts more control over it. While that is highly unlikely, these are the hard questions I am hearing from top manufacturing execs in Taiwan, who are more concerned than ever about China asserting their rule inside Taiwan. This increased demand should be seen by Intel as a major opportunity to bolster its long-term profitability. While they do have a strong position in PC’s and servers, to put all of their future growth eggs in this basket could, at some point, limit their long term potential. I understand they have other areas of focus that could help their growth, which includes smart cars, AI, and other markets for focused chips. But would those chips keep their factories at full capacity?

Ironically, Intel’s struggle to move to 7nm processes has forced them to think about outsourcing advanced processors to competitors such as TSMC. In an article in the Oregonian, it reports that Intel has confirmed that it is looking at outsourcing advanced processors.

CEO Bob Swan told Wall Street analysts on a conference call earlier this month that it may outsource advanced production to its rivals – he named Taiwan Semiconductor Manufacturing Co., specifically – to ensure “a predictable cadence of leadership products.” Swan told investors to expect a decision by late January.

On this month’s analyst call, Swan said Intel believes it can have it both ways – sending advanced production overseas while retaining internal production for components and older products that don’t require the most sophisticated technology. And Swan said Intel believes it could restore advanced manufacturing to its own factories sometime in the future if it chooses to.

This type of move would leave Intel’s own factories open to make chips for others, especially some Arm-based processors that don’t need advanced manufacturing processes. The challenge for Intel, and the tough pill to swallow, would be that in order to do this they would have to allow their fabs to use other companies process technology. If Intel was to offer TSMC and Samsung space in their fabs, both companies would use their proprietary process and Intel would basically be leasing space and equipment.

Intel still has fabs all over the world that could be utilized for making processors for Fabless semi-conductors firms and keep existing fabs going strong. Although this would upend Intel’s historical business model, this might be the time for them to seize the moment and go down a path that keeps all of their fabs humming.

Arm Based Macs and Mac Growth
Apple yesterday announced the news of their November event, where Arm-based Macs are likely to debut. My conviction is Apple has multiple agenda’s in moving from Intel to their own Arm-based Apple Silicon, but one of those agenda items is to grow the Macs share of the premium PC market. Proprietary estimates I’ve seen from IDC and Morgan Stanley have pegged the premium PC (defined as $1000+) at approximately 20-25% of WW notebook and desktop shipments. That equates to around 55-65m units a year annually. Apple averages about 20m Mac sales annually for reference or 33% of the premium PC market.

If my assumption is correct, and Apple believes there is share to gain for Macs in premium, then how much growth is realistic for Apple? The first thing we have to recognize is Mac share is unlikely to grow if current price points stay the same. Most financial analyst notes categorize Mac as ex-growth, and up to this point, that has been true. What’s fascinating, in my opinion, is how Apple can sell hundreds of millions of iPhones priced above $800 but only 20m Macs priced above ~$800. I don’t believe Apple will all of a sudden sell a hundred million Macs, but this point goes to show you that a product, when valued appropriately, even if expensive, can still move in volume.

This is why I think Tim Cook’s commentary about how the work-from-home trend gives them optimism for Mac and iPad is based on a potential reassessment of the value of notebook, desktop, and iPad form-factor. For this reason, I do believe Apple has share to gain for Mac in the premium price notebook and desktop segment, but again the question is how much room to grow.

Apple’s Annual Mac sales are in the 20m range and that is with current premium pricing and ASPs of well over $1000. So with the assumption Apple will at some point, bring an Arm-based Mac into the $799-$899 price range, I think Apple has the potential to grow Mac sales somwhere between 8-10m units a year. The unit growth may not seem like much but that is a nice bump in revenue for the Mac category.

In some talks with investors on the subject of Arm-based Macs I was asked if there are scenarios where Apple’s could grab even more share of the PC market, and the premium category specifically. In all honesty, consumers relationships with PCs would have to dramatically change for that to happen. Sure it is possible that Arm-based Macs with industry leading battery life, security/FaceID, incredible power, a new and thriving app ecosystem, and luxurious industrial design could do this, but it is not the scenario I’m betting on primarily.

I’m personally excited about Arm-based Macs and Apple’s likelihood to shake up the PC market. If anything, I’m predicting an interesting battle between Intel and Apple as Intel is likely going to go on the offensive with benchmarks to try and downplay any performance advantage Apple’s touts with Arm-based Macs. I anticipate a fierce battle that may get ugly but it will be fun to watch.

Apple Updates Risk Factors For Services
I wanted to briefly touch on the observation that Apple has updated their 10-K filing to include some new risk factors relevant to the potential regulatory issues around App store and their services business.

Regarding App Store, Apple notes the potential risk of reduced, narrowed or eliminated App Store take rates, which could have material adverse impacts on Apple’s financial condition and operating results. This seems to be alluding to either a mandated lower rate of commission for Apple or the forcing of alternate payment methods in which case Apple would get nothing. Interestingly, the prior risk factor, which was replaced by this one, was developers not using the App Store.

For services, a risk was updated related to gross margins varying and changing over time. The belief is this is alluding to the Google search deal, which if regulations impact, could materially hurt Apple’s services business as this deal with Google is estimated to be 15-20% of Apple’s annual services revenue. The Google deal falls under the licensing and other category and the belief is the Google payment represents the majority of this line item which has been 22% of services revenue over the last three years.

There is debate if this payment if fixed or variable. The argument that it is variable seems to hold water due to the fluctuation and weakness of the segment revenue which closely followed weakness in Google ad revenue. If a variable payment is likely the bulk of the deal’s revenue then that is a normal and fair business deal which should not be impacted. However, if the deal is largely a fixed sum, then it could come under any impact of regulations brought on Google.

The search angle, and revenue from a Google licensing deal is interesting in light of the recent rumors of Apple working on a search engine. While this could simply be Apple improving search as a whole on their devices related to Siri, some speculate Apple is looking to replace Google as a whole like they did with Apple Maps. From their privacy angle, this would make a lot of sense, and while the revenue from Google is hefty, it is certainly not something Apple truly needs or relies on.

New of the deal closing between AMD and Xilinx broke this week, and AMD will acquire Xilinx for an all-stock $35 deal. I’ve written quite a bit through the years about how the semiconductor would continue to consolidate, and Xilinx was a natural target, although I did not think AMD would be the buyer.

I found this tweet interesting from David Scor

It gets even more interesting when you look at the revenue multiples these deals closed for. Nvidia buys Arm for 20X revenue, AMD buys Xilinx for 11X revenue, Marvell buys Inphi for 13X revenue. In all these cases, the acquirer makes more per quarter than the total yearly revenue of the company they purchased. In all these deals, the acquirer is hoping to expand their market and leverage their technology to grow their piece of the pie in markets they currently compete in and those they want to compete in.

For AMD, Xilinx allows them to expand into automotive markets, acquire IP for FPGA’s, and possibly some elements of 5G. What stood out to me in CEO Lisa Su’s comments on the acquisition was how much they liked Xilinx because of their commitment to chiplets in their architecture. Chiplets are something AMD is heavily invested in from an architecture standpoint. In the broad context, Xilinx is extremely complementary to AMD and their philosophy on semiconductor architecture design.

Lisa Su also said, “the era of monolithic integration is over,” which was a direct shot at Intel given that it has been their mantra for many years. I actually believe Intel is moving closer to the embracing of chiplets, even though they are doing it in their own integrated way.

I have had several discussions with financial analysts about this acquisition, and most are not entirely sure how this deal helps AMD compete more broadly. Especially when AMD roughly makes a quarter how much Xilinx makes annually. But my basic point on why integration is inevitable in the industry is because supplies prefer to buy technology from as few sources as possible, which means the more they can buy from a preferred vendor, the better. This acquisition will let AMD get more share of wallet from customers but also keep those customers more loyal and committed to AMD technology.

This acquisition alone won’t immediately change AMD’s market share, and this is a long gameplay that we will see fruits from in 3-5 years. But, it has to be said, two or three years ago, AMD could not have made an acquisition like this. The fact they have executed and competed so well against Intel has gained them favor in the stock market and thus presented them with this opportunity. Being a public company can often be a curse, but in this case, it is a tremendous blessing for AMD.

Apple’s Bullish on Q4
Don’t call it a supercycle! That’s my favorite theme from many investor notes I’ve read. Everyone points to a number of fundamentals from 5G, expanded iPhone lineup, and elongated refresh rates globally as reasons Apple is well-positioned for a very good Q4. Yet no one wants to call it a supercycle, which is something I agree with. But the tone of these notes and Apple’s commentary on earnings was so positive on the iPhone cycle, yet no one wants to mention the supercycle phrase. It’s almost as if they feel if they do say it, it will jinx the quarter.

During the earnings call, Tim Cook kept emphasizing how bullish they are on iPhone going into Q4. Despite iPhone revenue being down in Q3, and markets like China slumping heavily, the new product launch is strong and new products always impact Q3 as a slump in sales. The reality is demand appears to be extremely heavy in all of Apple’s major markets. Many analysts are estimating Apple’s Q4 iPhone shipments to be well north of 70m with a consensus range in the 73-74m range.

The iPhone feeds the ecosystem, which is why a growing and loyal base is critical to so many adjacent products and services from Apple. While many in the media have been quick to point out how Apple’s reliance on iPhone revenue is lessening, it doesn’t change the reality that the iPhone sits in the center of Apple’s ecosystem and everything else is a branch from that tree.

While a strong Q4 is anticipated, Apple provided no guidance but did mention they expect growth across the board. On top of that commentary, they mentioned they are currently supply constrained across a range of product categories. Supply chain reports I’ve read mention Apple ramping manufacturing to meet demand and barring any more disruptions from COVID 19, I’m not sure this will be an issue impacting sales in Q4.

Lastly, one bit of commentary from earnings stuck out to me. A question came in from an analyst about why not offering a hardware bundle like they offer subscription bundles. Tim Cook did not allude to their interest in a hardware bundle but he did say they know a good portion of their customers likes to pay for products on a monthly play. Both iPhone and iPad can be paid as installments on Apple Card, and I fully expect this to Macs and likely Arm-based Macs to start. This single move to monthly payments for Macs could very well tip the market more in Apple’s favor than just lowering the price of a new Mac. How Apple can leverage Apple Card, or some other medium so their hardware can be paid in monthly installments is a luxury they have most of their competition does not. I expect them to leverage it more and more going forward.

Brief Update: As I read investor notes that hit my inbox this morning it does appear the supercycle language is now back (facepalm emoji). The notable difference is this cycle is being measured on a fiscal year basis, not as a single quarter, which I think is more reasonable. I have no doubt iPhone sales will grow in Apple’s fiscal 2021 and I think growth will continue into 2022.

Quibi has announced they have decided to shut down the service. I’m sure this is not a shock to many, although the implications are worth teasing out since I feel some conclusions are worth making.

Quibi’s commentary from management about the decision to shut down the service was largely blamed on timing. As many have pointed out, the timing should not have been better to launch a new service. It has been argued that people, during this pandemic, prefer long-form content over the short-term content Quibi was offering, but my take is good content will get watched and be in demand no matter the length.

What is interesting is the decision to simply shut the door and move-on. It is curious that Quibi, with quite a bit of cash still in the bank, is not attempting at least one pivot to see if they can turn the tide. It must have been concluded the cost to continue to compete, even in a pivot, would have required even more money they had access to, and with raising more money not an option, the conclusion was to cut their losses. This point leads to the main observation that sticks out to me.

Content is hard, and consumer attention is minimal. This was the root of Quibi’s challenge, but as we observe the significant amount of money raised by the company, it is another reminder of how expensive content is. The cost to get in and compete in the content market is borderline astronomical. Both of these points lead to the conclusion that the content industry is basically settled. There will not be content startups for a long-tine if ever that leads with high production value content.

This observation felt mostly obvious given the cost of content and the dynamic of competition favoring companies with existing distribution and a load of cash. There could still be new entrants into the content business, but they will be large established companies, not upstarts who require significant amounts of private financing.

The other implication, and this is a bit of speculation on my part, but I think it is logical, is I feel more consolidation is still to come for the content industry. Given the anti-trust scrutiny Apple, Amazon, Google, and Facebook are under, and I’m not sure if those companies are candidates in the near future to acquire content companies or a broadcast network. Still, I do think the NBC and CBS companies of the world may find themselves needing a partner with more cash and seek to sell.

The point goes back to the scarce attention consumers have and how the platform companies, in many regards, eat up a significant amount of that consumer attention with their core products.

The last implication I want to mention is how celebrities and or content owners/producers may be forced to rethink, where they prioritize their efforts. Quibi had a strong collaboration with top name actors/actresses, and while I’m sure many of them were paid upfront, I’ve heard from my friends who work in talent management that many of these actors/actresses are looking to new media opportunities to not just get paid upfront but also share in revenue growth in the long-term. I wonder if these owners/producers/and talent will start to look for new opportunities or just stick with the known success routes by working with Netflix, Apple, Amazon, etc. Again, the implications here, in my view, are what Quibi’s failure does to future innovative opportunities from the entertainment/content industry.