As part of the show, we’ve had the opportunity to interview Alex Katouzian, Qualcomm’s SVP and GM of the mobile, compute and infrastructure business – including Handsets, XR, Compute, Edge/AI Cloud, 5G/4G businesses.
Similar to last year’s interview, we were able to talk to Alex about this year’s new announcements, the 5G ecosystem, and the new technologies which enable these new generation products.
Alex Katouzian
Qualcomm
Andrei Frumusanu
AnandTech
Andrei F. : Let’s first start with an introduction for the readers out there, which are going to be reading this during the launch event.
Can you describe in a few brief sentences, what the new Snapdragon 888 is all about?
Alex K. : So as you know, the whole Tech Summit is about showcasing our technology. And what happens, is this premium tier technology, including the modem, and the system solution, start to dictate what we do not only in all the other tiers of our products and for mobile, but actually a lot of the adjacent businesses that we’re going after as well, they use all of this technology. In fact, I can’t see any business that we have, whether it being IoT, wearables, automotive, or even cloud edge, mobile broadband PCs, or XR – all of them use almost the exact same technology across the board.
And so this is the this premium tier announcement, like every year, which really dictates what we do across multiple businesses, and across multiple tiers of our products. Those features and those capabilities start to waterfall all the way down into into lower and lower tiers. I would say, somewhere in the neighborhood of three to six months, you see a lot of these features water falling down. So we’re very, very excited about this.
I think 5G now is very well established, there’s like 40 countries, a hundred plus operators with 5G around. It is the case where, we introduced the new chipset and the end-unit solutions before all the networks were ready to go.
It’s a really good thing to have, because now many of the service providers, so many of the app developers, so many of the ISVs, they have the vehicles to start to develop stuff from the ground up for 5G.
I think we have we’re expecting somewhere in the neighborhood of, about 180 million to maybe 220 million 5G phones coming through in 2020, and then, about 450 to maybe 550 million units of 5G devices in 2021.
I think I think China is going full speed ahead. Big OEMs like Samsung, Apple, LG, Motorola, Vivo, Oppo and Xiaomi. I don’t know what’s going to happen to to to Huawei, it’s still up in the air. But I don’t know if you saw the news – there was a recent sale of their sub-brand called Honor. So, Honor is now like a company owned by the Shenzen government. And so they’re gonna start to get into the market, they want to get active in 5G very fast. So I think we have super amounts of momentum.
Getting back to your question on the Snapdragon 888. It is really probably the the most advanced part we’ve done, with the most advanced technologies across the board.
Including 5G, we have a third generation 5G modem going inside that. We have multiple different carrier aggregation modes between mmWave and Sub-6 between TDD and FDD.
We have, if I look at the pillars, if I look at AI, we have a 26 TOPS AI engine that we’re putting in, this is the sixth generation engine. We have a tensor processor really designed from the ground up for for mobile AI capabilities embedded in there.
We have our graphics core, that is latest graphics core where we’re seeing huge improvement from the 865 mobile processor that came out last year. Our elite gaming is advancing in a big way. We have multiple different partners that we’re working with now.
AF: So because this is your third generation 5G product, and you could pretty much say that 2020 has been the year where it’s gone mainstream, have you gained any additional learnings this year?
It has been such a wide deployment across so many companies, so many countries. What have you actually improved this year, which might be new, compared to the previous two generations?
AK: So absolutely, I think we’ve done a hell of a lot more testing with the infrastructure vendors. You know, we’ve been able to optimize power on multiple different use cases. We’ve been able to, like I said, work with so many service providers try to figure out how to have their services be built with 5G in mind versus and agnostic approach.
The handsets have gotten more slick. Power dissipation has grown and battery life has gotten longer. You know, there were some reports, if you remember, in the beginning, like the phones would heat up – none of that stuff is an issue anymore, in our opinion.
AF: So on that point, I remember early this year when the first 5G devices came out, especially in the US, those with the mmWave, that there was some mixed reaction. I mean, the speeds were fantastic. But also many people said that the devices heating up, that battery life is not as good. Is that something solved in this generation?
Absolutely solved. And let me tell you what the problem was. They would take the phones, and they would run full speed mmWave wave for two hours straight on something. Of course, the phone would heat up, you know, even even running a game without mmWave, if you’re going for like two hours straight, full blast, the phone heats up.
So I think they put it through some extreme types of testing. But now, we have optimizations across the networks that have been put up, we have multiple different networks that we’re working with. We have developed different methods by which, we can reduce not only the die area of the modem itself, but then methods to try to figure out how to optimize all the channel optimizations and all the data optimizations that we’re going through. All of those things are big learnings for us over the past two years,
AF: For networks where you have both mmWave wave and sub-6. How do you work to balance out the traffic between the two the two sides of the network? Is that something which you have to work closely with the carrier, or is that something which you on your site can control more?
AK: Well, both. So so for example, the carrier lays out their networks, to try to figure out how to best serve the most amount of traffic. When you go into dense urban areas, that’s where mmWave kicks in the most, because they want to have the maximum amount of bandwidth available to most amount of people. And so, based on their layouts, and based on their maps, we can figure out with the infrastructure vendor and the carrier how to switch over from one to the other.
AF: Okay, but what I mean is for example, if you’re browsing your Twitter, maybe you don’t need the bandwidth from mmWave, do you have some smart mechanisms to direct the traffic over to Sub-6 or LTE in those use cases?
AK: Not particularly from Qualcomm, but we do work with the carriers to try to accommodate what they want to do. And if there’s, if there’s any technical issues with the infrastructure, we work with the vendors there to figure that out.
AF: But the chipsets themselves would be able to support such a thing, right?
AK: Correct. Yeah. Anytime we get a signal from the carrier to switch to something else we we accommodate right away.
AF: So you mentioned the 888 represents the pinnacle of the newest technologies. One big aspect of that is the new process node, the new chip is your first flagship SoC on an EUV node. I know last generation, you had the 765 on EUV, but this one is like the big prestigious design.
In the past, we talked about process nodes, and how Moore’s Law is slowing down. How do you see, in the broad sense of terms, how the technology is evolving? Is it getting harder, is getting more expensive? What’s your view?
AK: Very good question. It is slowing down. And by when I say slowing down, we are also slowing down in moving into advanced nodes. And the reason for it is exactly what you described; Moore’s Law is not as it was before, you’re not getting the shrinks as you did before, you do get some power dissipation advantages, and you get some shrink advantages, but not as big returns as is adding functionality every year. And the functionality gets added every year.
So here’s the cool part. When you have a process technology, and it’s leading edge, what happens is when you try to squeeze the maximum amount of performance out of it, and you try to maximize your power dissipation, your yields in the beginning are not very good. And so what happens is, you’re on a learning curve on these yields. It looks like a bell shaped curve. So, as the bell shaped curve tightens up, you can increase performance and decrease power.
But every process, nowadays, we, and fabrication facilities like TSMC, and Samsung and others, we can squeeze those curves down over like a two to three year period.
So a process, when it first comes out, it’s great, because it gives you some advantages, but it’s not at its maximum capability. So over two or three years, they introduce best known methods, better transistor capabilitiees, they’re just slightly better transistor capabilities, but you’re not actually shrinking the transistor.
So our designs start to flow with that. So maybe you’ll see one or two generations at least of a similar process. So like 5 nanometer is not exactly the same as 7 nanometre, but it reuses a lot of the technology learning curves of 7 into 5. So not only is the transistor getting slightly smaller, but we’re squeezing more and more performance out of it. And so, as the migration from process to process starts to slow down, we also learn how to get better performance and better power out of the same, or very similar process.
AF: At which point do you see that the return on investment wouldn’t be enough to just rely on process improvements, and when do you go to look into more exotic solutions, things like new packaging technology, or disaggregated designs like chiplets? Is that something which is on the radar for the mobile? Or do you still try to keep everything together in a monolithic design because of power efficiency?
AK: Very good question. Look, I think we have a few years. And the reason for it is, let me give you an analogy, you know, how you put up solar panels. Say you have 100% efficiency, or you’re close to whatever you’re spending. You could add more, but you get diminishing returns. On the process technology, I think we have a few years left until we say that we’ll no longer going to integrate an SoC, with application processor and modem and a bunch of other things.
Obviously, the die area starts to grow. But the shrinking of the transistor can’t keep up as much as the functionality is growing. But, like I said, the efficiency of reusing a process or a tweaked process is also yielding us quite a bit of capability. So I think we have a few years before we have to think about separating the die in such a way that it makes sense for us to partition.
The other thing you have to take into consideration is, you’re talking about a heterogeneous system on a chip, which means parts of parts of the die depend on other parts of the die to operate. So it’s not as simple as saying, I’m going to cut graphics out and put it someplace else, or I’m just going to cut CPU and put it someplace else. We are in the process of looking at that – I would say in the next three or four years, our architecture and partitioning will change. But I think we have a few years before that happens.
AF: This generation, you’re using the Cortex-X1 which is basically like a new category within Arm’s CPU IP offering. From your side, in terms of a product segmentation and marketing aspect, does that allow you to go higher up in product range?
We obviously might not see it go into lower-range Snapdragon. How do you see that differentiation with new CPU technology?
AK: I think that core does help, and especially on the premium tier. Let me also explain; the premium tiers life doesn’t end in the first year. So, people use that, people look to use those premium tier parts in the following year for designs that are inbetween high tier and premium tier.
The feedback from our customers in our ecosystem has been consistent, there are four or five technologies that they hone in on: CPU, GPU, camera, AI, modem. But, top of that list it’s always, always CPU and GPU. And so, the performance allows you to extend the life of those designs, and it helps you set a precedent. So it absolutely helps us, and it also helps us for the length of the life of the device.
AF: Moving on from mobile to the laptop side of Snapdragon, such as last year’s 8CX.
It’s been a year now, and pick-up and adoption hasn’t been that great. We haven’t seen that many designs, the ones that are available aren’t that popular. What’s your view of that segment, and is Qualcomm as motivated as you were one year ago in terms of engaging that ecosystem?
AK: For sure one hundred percent behind it. Let me let me also explain – the laptops these days are really moving towards mobile. The camera is super important. The audio is super important. The battery life is super important. Not having a fan is super important. Portability, thinness, connectivity, always-on always-connected, all those traits of mobile are moving to the PC.
And people say, imitation is the best form of flattery. Look at look what happened with the [Apple] M1. Their product pitch is almost a duplicate of what we’ve been saying for the past two or three years.
It’s almost exactly the same; like I have an SoC, and you don’t have to have anything on the outside, everything’s integrated, multimedia capability, AI capability, camera capability, you know, great battery life, the whole thing.
And now, I think the momentum of having so many different apps, and so many different capabilities running on an Arm-based solution that has multiple traits associated with it, and low power capability, and being able to do video conferencing and have great audio clarity.
I’m talking to you from a Samsung Book S (with 8CX processor), I’ve been using it for a year and a half, one of the best laptops ever had. You’re correct in saying that the consumer hasn’t yet caught up. Enterprises are catching up. Our channel push has been stronger than ever, carriers are starting to realize that this is a good thing to have.
And then here’s something else that really is going to catch. We’re going to have small cells based on 5G going inside enterprises, we’re going to have access points that have 5G backhaul going inside of enterprises. Having the capability of that 5G PC with the small cells to go into an enterprise is a very, very attractive formula for all of the carriers. So that channel is going to start to open up, Microsoft is 100% behind us as a partner to try to get all of the apps, all of the 64 bit emulations up and running and resolve all these issues. And on top of that, there’s Apple who’s now in the market and everyone wants to react to it.
AF: And so do you think that Apple’s announcement now has pushed an ecosystem forward to be more motivated? There’s kind of a few things happening at the same time right now, such as Microsoft bringing 64 bit emulation to Windows, and that’s going to open up a huge swath of software compatibility. Do these different things coming in at the same, do you think that it’s going to pick up the ecosystem for Windows on Snapdragon?
AK: Yes, huge momentum, huge momentum. And the important thing is we’re continuing to put together roadmaps of devices that are going to be competitive in the market, they’re going to be higher performance, they’re going to be better and better every year. And you will see our upcoming SoCs in the future.
AF: So you’re saying every year, but you’re still on a different cadence than on the mobile SoC side, because the 8CX Gen2, the refresh, it’s the same chip, right? So you’re on a different cadence – when would we see something way higher performance to enable more use cases?
AK: I can’t tell you exact dates, but I can tell you, it’s coming.
And, and I’ll say this, too. We are 100% dedicated to this market. I think Microsoft is 100% dedicated to us, to make sure that this is going to happen. And I think that, you have a $2 trillion company coming into the market and saying, this is the way to go.
We’re only $170 billion, but you know, it helps to have a $2 trillion company saying, “Yeah, this is the right thing to do”. And guess what, we’re inundated with calls to make sure that this is going to happen. So we’re 100% behind this stuff.
AF: Speaking of Microsoft, when do you see you might be designing SoCs with the new security IP from Microsoft, Pluton?
AK: We’re definitely working with them on that. I can’t predict, but you know, our working relationship with them is super tight. So we’re working with them on that, but I can’t predict to you when that’s going to happen.
AF: On the x86-64 emulation side, is there some collaboration you have with them in terms of optimizing the performance for your chips?
AK: 100%. And you saw SQ1 and SQ2, we work with them super tightly. Even though everything is based on Qualcomm technology, our optimizations with them is second to none. Our working relationship with them, and trying to figure out how we fit our hardware layers in their stack, and how the APIs are optimized that way, to present a full system solution to a consumer. And I think it’s gonna get even better from now.
AF: So, from your view, what’s the biggest growth market? On one side, we have 5G. On the other side, we have this upcoming Windows on Snapdragon market, maybe, hopefully, that’s going to pick up way more now. What’s the biggest path for opportunity right now?
Well, I think the PC market for us as supplementary market. Because if I can’t grow in mobile, and show technologies in mobile, then I think the PC side will suffer. So I think volume wise, hundred percent of the mobile market is the biggest market by far.
But I think, the most lucrative part of the mobile market, just like almost any other market is, in fact, the premium tier and the high tier. As technologies become better and better, that high tier part is also getting better and better. Like, the high tier part we have this year is better than our premium part tier we had two years ago. So that improvement by itself is kicking it up a notch. And you know, those high tier equivalent type of devices, they’re also going to lower tier PCs.
So I look at the PC market as a supplementary market. And think of it this way, if I can capture 10% of that market, 15% of that market, it’s still a small player, but it’s a huge growth for us. So I definitely see the opportunity in both, but absolutely the mobile market is the bigger one.
AF: What about on the low end side? How cheap can we get 5G devices in 2021? Are we going to see a $250 5G phone?
AK: Absolutely. $250 is is definitely a target. If you look at the 400-series part that we announced with 5G – one hundred percent going into that market. I also think you’re probably gonna see $300 PCs with LTE.
AF: $300 PCs with LTE? That would be quite something.
AK: Education type PCs, and, you know, lower lower end, affordable PCs. Yes.
AF: Last question, what happened to the Snapdragon 875? Explain that!
AK: (laughs) Look, the number eight is the representation for premium tier for the past 10 years, there’s there’s no doubt behind that.
And that has gained so much momentum with consumers, just purely by marketing, by us, marketing by our OEM partners. So the eight has got a big attraction. And since we have, I think, probably the best part we’ve done this year, we wanted to, assign a premium tier number and name to it, and 888 is a good one.
We thank Alex Katouzian and the Qualcomm team for their time and answers.
SAN FRANCISCO (AP) — Artificial intelligence‘s recent rise to the forefront of business has left most office workers wondering how often they should use the technology and whether a computer will eventually replace them.
Those were among the highlights of a recent study conducted by the workplace communications platform Slack. After conducting in-depth interviews with 5,000 desktop workers, Slack concluded there are five types of AI personalities in the workplace: “The Maximalist” who regularly uses AI on their jobs; “The Underground” who covertly uses AI; “The Rebel,” who abhors AI; “The Superfan” who is excited about AI but still hasn’t used it; and “The Observer” who is taking a wait-and-see approach.
Only 50% of the respondents fell under the Maximalist or Underground categories, posing a challenge for businesses that want their workers to embrace AI technology. The Associated Press recently discussed the excitement and tension surrounding AI at work with Christina Janzer, Slack’s senior vice president of research and analytics.
Q: What do you make about the wide range of perceptions about AI at work?
A: It shows people are experiencing AI in very different ways, so they have very different emotions about it. Understanding those emotions will help understand what is going to drive usage of AI. If people are feeling guilty or nervous about it, they are not going to use it. So we have to understand where people are, then point them toward learning to value this new technology.
Q: The Maximalist and The Underground both seem to be early adopters of AI at work, but what is different about their attitudes?
A: Maximalists are all in on AI. They are getting value out of it, they are excited about it, and they are actively sharing that they are using it, which is a really big driver for usage among others.
The Underground is the one that is really interesting to me because they are using it, but they are hiding it. There are different reasons for that. They are worried they are going to be seen as incompetent. They are worried that AI is going to be seen as cheating. And so with them, we have an opportunity to provide clear guidelines to help them know that AI usage is celebrated and encouraged. But right now they don’t have guidelines from their companies and they don’t feel particularly encouraged to use it.
Overall, there is more excitement about AI than not, so I think that’s great We just need to figure out how to harness that.
Q: What about the 19% of workers who fell under the Rebel description in Slack’s study?
A: Rebels tend to be women, which is really interesting. Three out of five rebels are women, which I obviously don’t like to see. Also, rebels tend to be older. At a high level, men are adopting the technology at higher rates than women.
Q: Why do you think more women than men are resisting AI?
A: Women are more likely to see AI as a threat, more likely to worry that AI is going to take over their jobs. To me, that points to women not feeling as trusted in the workplace as men do. If you feel trusted by your manager, you are more likely to experiment with AI. Women are reluctant to adopt a technology that might be seen as a replacement for them whereas men may have more confidence that isn’t going to happen because they feel more trusted.
Q: What are some of the things employers should be doing if they want their workers to embrace AI on the job?
A: We are seeing three out of five desk workers don’t even have clear guidelines with AI, because their companies just aren’t telling them anything, so that’s a huge opportunity.
Another opportunity to encourage AI usage in the open. If we can create a culture where it’s celebrated, where people can see the way people are using it, then they can know that it’s accepted and celebrated. Then they can be inspired.
The third thing is we have to create a culture of experimentation where people feel comfortable trying it out, testing it, getting comfortable with it because a lot of people just don’t know where to start. The reality is you can start small, you don’t have to completely change your job. Having AI write an email or summarize content is a great place to start so you can start to understand what this technology can do.
Q: Do you think the fears about people losing their jobs because of AI are warranted?
A: People with AI are going to replace people without AI.
WASHINGTON (AP) — The Biden administration said Tuesday that it would provide up to $325 million to Hemlock Semiconductor for a new factory, a move that could help give Democrats a political edge in the swing state of Michigan ahead of election day.
The funding would support 180 manufacturing jobs in Saginaw County, where Republicans and Democrats were neck-in-neck for the past two presidential elections. There would also be construction jobs tied to the factory that would produce hyper-pure polysilicon, a building block for electronics and solar panels, among other technologies.
Commerce Secretary Gina Raimondo said on a call with reporters that the funding came from the CHIPS and Science Act, which President Joe Biden signed into law in 2022. It’s part of a broader industrial strategy that the campaign of Vice President Kamala Harris, the Democratic nominee, supports, while Republican nominee Donald Trump, the former president, sees tariff hikes and income tax cuts as better to support manufacturing.
“What we’ve been able to do with the CHIPS Act is not just build a few new factories, but fundamentally revitalize the semiconductor ecosystem in our country with American workers,” Raimondo said. “All of this is because of the vision of the Biden-Harris administration.”
A senior administration official said the timing of the announcement reflected the negotiating process for reaching terms on the grant, rather than any political considerations. The official insisted on anonymity to discuss the process.
After site work, Hemlock Semiconductor plans to begin construction in 2026 and then start production in 2028, the official said.
Running in 2016, Trump narrowly won Saginaw County and Michigan as a whole. But in 2020 against Biden, both Saginaw County and Michigan flipped to the Democrats.
Although no one likes a know-it-all, they dominate the Internet.
The Internet began as a vast repository of information. It quickly became a breeding ground for self-proclaimed experts seeking what most people desire: recognition and money.
Today, anyone with an Internet connection and some typing skills can position themselves, regardless of their education or experience, as a subject matter expert (SME). From relationship advice, career coaching, and health and nutrition tips to citizen journalists practicing pseudo-journalism, the Internet is awash with individuals—Internet talking heads—sharing their “insights,” which are, in large part, essentially educated guesses without the education or experience.
The Internet has become a 24/7/365 sitcom where armchair experts think they’re the star.
Not long ago, years, sometimes decades, of dedicated work and acquiring education in one’s field was once required to be recognized as an expert. The knowledge and opinions of doctors, scientists, historians, et al. were respected due to their education and experience. Today, a social media account and a knack for hyperbole are all it takes to present oneself as an “expert” to achieve Internet fame that can be monetized.
On the Internet, nearly every piece of content is self-serving in some way.
The line between actual expertise and self-professed knowledge has become blurry as an out-of-focus selfie. Inadvertently, social media platforms have created an informal degree program where likes and shares are equivalent to degrees. After reading selective articles, they’ve found via and watching some TikTok videos, a person can post a video claiming they’re an herbal medicine expert. Their new “knowledge,” which their followers will absorb, claims that Panda dung tea—one of the most expensive teas in the world and isn’t what its name implies—cures everything from hypertension to existential crisis. Meanwhile, registered dietitians are shaking their heads, wondering how to compete against all the misinformation their clients are exposed to.
More disturbing are individuals obsessed with evangelizing their beliefs or conspiracy theories. These people write in-depth blog posts, such as Elvis Is Alive and the Moon Landings Were Staged, with links to obscure YouTube videos, websites, social media accounts, and blogs. Regardless of your beliefs, someone or a group on the Internet shares them, thus confirming your beliefs.
Misinformation is the Internet’s currency used to get likes, shares, and engagement; thus, it often spreads like a cosmic joke. Consider the prevalence of clickbait headlines:
You Won’t Believe What Taylor Swift Says About Climate Change!
This Bedtime Drink Melts Belly Fat While You Sleep!
In One Week, I Turned $10 Into $1 Million!
Titles that make outrageous claims are how the content creator gets reads and views, which generates revenue via affiliate marketing, product placement, and pay-per-click (PPC) ads. Clickbait headlines are how you end up watching a TikTok video by a purported nutrition expert adamantly asserting you can lose belly fat while you sleep by drinking, for 14 consecutive days, a concoction of raw eggs, cinnamon, and apple cider vinegar 15 minutes before going to bed.
Our constant search for answers that’ll explain our convoluted world and our desire for shortcuts to success is how Internet talking heads achieve influencer status. Because we tend to seek low-hanging fruits, we listen to those with little experience or knowledge of the topics they discuss yet are astute enough to know what most people want to hear.
There’s a trend, more disturbing than spreading misinformation, that needs to be called out: individuals who’ve never achieved significant wealth or traded stocks giving how-to-make-easy-money advice, the appeal of which is undeniable. Several people I know have lost substantial money by following the “advice” of Internet talking heads.
Anyone on social media claiming to have a foolproof money-making strategy is lying. They wouldn’t be peddling their money-making strategy if they could make easy money.
Successful people tend to be secretive.
Social media companies design their respective algorithms to serve their advertisers—their source of revenue—interest; hence, content from Internet talking heads appears most prominent in your feeds. When a video of a self-professed expert goes viral, likely because it pressed an emotional button, the more people see it, the more engagement it receives, such as likes, shares and comments, creating a cycle akin to a tornado.
Imagine scrolling through your TikTok feed and stumbling upon a “scientist” who claims they can predict the weather using only aluminum foil, copper wire, sea salt and baking soda. You chuckle, but you notice his video got over 7,000 likes, has been shared over 600 times and received over 400 comments. You think to yourself, “Maybe this guy is onto something.” What started as a quest to achieve Internet fame evolved into an Internet-wide belief that weather forecasting can be as easy as DIY crafts.
Since anyone can call themselves “an expert,” you must cultivate critical thinking skills to distinguish genuine expertise from self-professed experts’ self-promoting nonsense. While the absurdity of the Internet can be entertaining, misinformation has serious consequences. The next time you read a headline that sounds too good to be true, it’s probably an Internet talking head making an educated guess; without the education seeking Internet fame, they can monetize.
