[blackwasp]

Nvidia Corp.’s market valuation topped Intel Corp.’s for the first time, powered by soaring demand for graphics chips in data centers and other fast-growing technology fields.

Nvidia gained 2.4% on Wednesday, giving it a market value of more than $248 billion. Shares of the graphics chipmaker are up 72% so far this year as investors bet the coronavirus pandemic has accelerated a shift to cloud-based digital services that use its technology. Intel shares have fallen 2% in 2020.

Nvidia was co-founded in 1993 by Jensen Huang, who’s still running the company. At the time, it was one of about two dozen graphics chip companies. It’s now the only independent maker of these components, after all of its rivals have been bought, folded or become part of larger companies.

Nvidia was more successful than its peers at developing chips that turn computer code into the realistic images computer gamers love. Under Huang, the company has pushed that technology into new markets, such as data center servers and artificial intelligence processing.

Investors have rewarded this fast expansion with a rich valuation. Since debuting on the Nasdaq in 1999, the stock has averaged an annual return of 33%. In the past five years, it has soared more than eightfold and trades at 75 times earnings, according to data compiled by Bloomberg. Intel shares trade at 12 times earnings.

Nvidia is now the third-largest chipmaker by market capitalization, behind Taiwan Semiconductor Manufacturing Co. and Samsung Electronics Co.

Intel also lost the title of the world’s largest chipmaker by revenue to Samsung Electronics Co. in 2017. It regained the title a year later, thanks to its resilient server chip business.

Source

[GutsyGibbon]

With the fruit company switching to Arm, expect a slew of Arm powered machines running windows. After Intel sold their 5G interests, they will have nothing that is a leading solution in the industry. Seems like a slow death spiral for Intel.

[locusjag]

Intel had failed Moore's law for the first time in all these years in 2018. They simply couldn't etch any finer on the chips, thus failing to double computing power. That's in my limited understanding.

Nvidia on the other hand has given us the current AI summer with their GPUs singlehandedly. Tesla had begun their first self-driven cars with their GPU. I don't know Tesla's current arrangement; but Nvidia has enabled the current spurt of interest in the world in Deep Learning with their GPUs. Of course most credit belongs to those who've devised a method to utilize GPUs for deep learning.

[Sebring]

Meanwhile AMD seems to be doing well with RYZEN. Performance and pricing are both above par. Apple moving away is a big sign too

Originally Posted by locusjag

Intel had failed Moore's law for the first time in all these years in 2018. They simply couldn't etch any finer on the chips, thus failing to double computing power

[srishiva]

Originally Posted by GutsyGibbon

With the fruit company switching to Arm, expect a slew of Arm powered machines running windows. After Intel sold their 5G interests, they will have nothing that is a leading solution in the industry. Seems like a slow death spiral for Intel.

Intel may have moved away from handset chipsets. But they are trying to do a major push on the network equipment with open alliances.
But then even in those wireless areas GPUs are kindling new interest in cloud deployments.

[Gsynch]

NVIDIA being a leading gaming platform company, also provides supercharged computing.

For 30 years, the dynamics of Moore’s Law held true as microprocessor performance grew at 50 percent per year. But the limits of semiconductor physics mean that CPU performance now only grows by 10 percent per year.

NVIDIA GPU computing has given the industry a path forward -- and will provide a 1,000X speed-up by 2025.

[GTO]

I can't for the life of me figure out how or why Intel missed out on the smartphone revolution (just like their longtime partner Microsoft).

Can any BHPian explain why in layman terms?

[V.Narayan]

Originally Posted by GTO

I can't for the life of me figure out how or why Intel missed out on the smartphone revolution (just like their longtime partner Microsoft).

Can any BHPian explain why in layman terms?

"Marketing Myopia" by Theodore Levitt. This is a blind siding caused by the frailness of human thinking influenced by super success with one formula. First you define your product by the market and then once you are super successful the market (in your mind) gets defined by the product!! As new modes of transportation come into play over the next 25 years this could (and most probably will) happen to some car manufacturers too where they will define their business as making ICE cars and not as providing a human transportation solution.

[Sutripta]

Originally Posted by GTO

I can't for the life of me figure out how or why Intel missed out on the smartphone revolution (just like their longtime partner Microsoft).

Can any BHPian explain why in layman terms?

Microsoft -
MS was facing a lot of regulatory pressure (and it never really recovered from it) and that both distracted its management, and made it excessively cautious, did not want to open new fronts, which would give further ammunition to its opponents.
Steve Ballamer did not believe in mobile computing. In fact was quite derisive about it. (See VNs explanation above)
Present - I think Nadella's opting out of the mobile OS space ALTOGETHER is a mistake. He had a different set of compunctions. Like with Tim Cook, shareholders might be happy now, but futurewatchers are somewhat apprehensive. Unless something else (Hololens? Quantum computing? Not cloud computing for corporates) clicks.

Intel -
Don't really know.
Much has been said about theirs being wedded to PCs. The 80x86 architecture was/ is notoriously bad when it came to computing power per watt. In their defense, that was never a consideration initially. Why they didn't design another product line without these problems is what I'd like to know. Both the market, and Intel's size should have made it a nobrainer.
And Intel had a huge advantage when it came to legacy code - emulating or actually implementing x86 instructions in microcode/ hardware without legal problems.
Their bigger failing has been twofold. Not being able to manufacture at productivity levels for which they were famous. Don't know why. Can't be something fundamental because other fabs are delivering.
Far more important in my view was their inability to deliver on 5G*, when Apple bet half the farm on it. Someone should take a deep dive into why Intel couldn't, but it will take time to get at facts. Now one can just speculate. It is too recent, and people who could shed light on it are bound by NDAs.

I must add that judging a company purely on market valuation (Tesla? now NVidia etc) leaves a lot to be desired. But then again not only am I not a finance guy, would not bet on my expertise in most other fields also!

Sutripta

* Unlikely to be the quality of people, because Apple essentially picked up the full Intel 5G team.

[StarrySky]

Intel was trying to enter the mobile phone market before the smartphone era (post iPhone). Back in 2001/2002, Intel was partnering with the company I used to work for, to develop a 2G/3G mobile phone platform (our company was probably one among a few of Intel's partners). So, this was before network operators were deploying 3G and there were practically no 3G phones in the market. I am not quite sure why the project stalled (way above my pay grade). Maybe they didn't think the mobile phone market would take off the way it did. Eventually another MNC partnered with my employer company to develop a mobile phone platform, which was then used by a Korean company to launch a few mobile phone models.

[Gsynch]

While there are many missed opportunities by Intel especially in the last decade, the mobile platform is one such. Another opportunity which is not known to many is the Automotive sector and within it, Autonomous driving.

NVIDIA placed a huge bet on this and kept investing for years till this major breakthrough: Nvidia announced a partnership with Mercedes-Benz to bring software-defined computing architecture to the Mercedes-Benz fleet starting in 2024

https://blogs.nvidia.com/blog/2020/0...ined-vehicles/

The price target for NVIDIA has been revised to 500 USD now, 20% upside on Friday close.

Disclaimer: I am invested in the company

[msdivy]

Intel with $75B in revenue & $23B in net income is now smaller than Nvidia, which has $12B in revenue & $3B in net income . That is the beauty of stock market.

Originally Posted by GTO

I can't for the life of me figure out how or why Intel missed out on the smartphone revolution (just like their longtime partner Microsoft).

Can any BHPian explain why in layman terms?

Intel designs x86 chips, which are CISC based and geared towards providing the highest computing power, with the goal of getting it done fast. Its primarily used in desktops, servers, laptops designs. In auto parlance, it is equivalent to the design of truck engines and the goal here is load-bearing capacity.

Arm, on the other hand, designs the RISC based chips, whose goal is power efficiency and used in embedded systems designs. An equivalent auto example is the engine of fuel-efficient cars.

You can use a truck to transport 5 boxes in a single trip or use a car to transport 1 box at a time and make 5 trips. At any time during the trip, the car would consume less fuel than the truck. The computing limitation of RISC is mitigated by serverless computing design, where all the load is processed on x86 servers and Arm mobiles show the results. For instance, for converting a video, you can upload the original video from your mobile to a website. The website will convert and you can download the converted video.

Arm was founded in 1985 and it was a steadily successful company until the mobile & handheld boom catapulted them into the stratosphere. BTW, Arm doesn't manufacture the chips. It only licenses it. It is akin to Fiat sharing its engine design with anybody.

In contrast, Intel's x86 design is a closely guarded secret. Intel has to design & manufacture them. BTW, Intel also has RISC design of x86 and it did get into mobile & hand-held business. But no OEM/ODM was willing to risk on maintaining 2 Android production lines - Arm & Intel. Hence no traction for Intel and it has lost big money here - $Bs.

The market also played its part. The main mobile players - Apple & Samsung are competitors to Intel. The Chinese want a cheap design and Intel likes to charge a premium. None of these manufactures would go to Intel. So it unlikely Intel can get into mobile & hand-helds until the landscape changes drastically.

(just like their longtime partner Microsoft).

Microsoft (with Windows CE & Windows Mobile OS) & Palm OS were ruling the handheld market prior to the introduction of the iPhone. With the iPhone, Apple dictated how a smartphone should look & behave. Android was able to catch up with Apple's philosophy. Microsoft was slow here and it finally came up with Windows Phone OS. It couldn't compete with Android, which was free and it couldn't charge premium like Apple, for their Lumia phones. Also, the developers' community had moved on to iOS & Android. They were not developing Windows Phone apps since there was very little usage.

With the change in CEO, Microsoft is focussing on cloud business. They are now a pureplay software company in mobile space. But there is every chance Microsoft might ride the next wave when there is one.

[locusjag]

Originally Posted by GTO

I can't for the life of me figure out how or why Intel missed out on the smartphone revolution (just like their longtime partner Microsoft).

Can any BHPian explain why in layman terms?

I've also read through lesser-known Management Guru Chris Argyris' account of how Intel very nearly missed the bus to the world of microprocessors, decades ago. In his book "Reasons and rationalizations: limitations to organizational learning", he narrated how he made those at the helm of Intel realize that violently putting down novel suggestions from insiders and employees only harms the organization. Intel used to make memory devices of some sort initially (ROMs I think). Whilst talk came up fom junior staff of moving on to microprocessors, those at the helm in Intel apparently would abuse the juniors vehemently. The bigwigs just couldn't stomach the newer thoughts presented by those under them.

But Chris Argyris helped Intel steer through an intense chapter of change management to transition to microprocessors....and the rest is history.

Did Intel need another Chris Argyris in recent times? Of course, I'm only speculating as to the recent shape of things inside Intel. I just remembered this book that I read a decade ago in B School. Chris Argyris is probably the most underrated management guru of years bygone IMO.

[Sutripta]

Originally Posted by msdivy

Intel designs x86 chips, which are CISC based and geared towards providing the highest computing power, with the goal of getting it done fast.
...
RISC based chips, whose goal is power efficiency and used in embedded systems designs.

Not to get into a debate about RISC/ CISC, and not really relevant to this thread, but RISC was conceived to be faster than CISC. Power consumption was important only in seeing to it that the chip did not selfdestruct.

The first commercially available RISC processor IIRC was the Intel i960. A commercial failure. Apparently still used by ISRO!

Sutripta

[ashokrajagopal]

Intel did try in the Smartphone market with its Atom series.

It did collaborate with many including Google, to directly design SoC for Android.
There were a few, but the most popular device IIRC was Asus Zenfone range. There were a few from Lenovo also I believe.

Atom could not survive in the Cortex-Snapdragon field.

Of course, Intel could have leveraged its market position and brand name to try on their own (remember Dell phones ?), but they did not.