By now, it’s clear that Apple is at the cutting edge of a revolution in chips. Every week, “Odd Lots” hosts Joe Weisenthal and Tracy Alloway take you on a not-so-random walk through hot topics in markets, finance, and economics. This week is about Apple.
M1 is optimized for Mac systems in which small size and power efficiency are critically important. As a system on a chip (SoC), M1 combines numerous powerful technologies into a single chip, and features a unified memory architecture for dramatically improved performance and efficiency. It features the world’s fastest CPU core in low-power silicon, the world’s best CPU performance per watt, the world’s fastest integrated graphics in a personal computer, and breakthrough machine learning performance with the Apple Neural Engine. As a result, M1 delivers up to 3.5x faster CPU performance, up to 6x faster GPU performance, and up to 15x faster machine learning, all while enabling battery life up to 2x longer than previous-generation Macs. With its profound increase in performance and efficiency, M1 delivers the biggest leap ever for the Mac.
In this week’s “Odd Lots,” Weisenthal and Alloway talked about stumbling, rather stagnant Intel and how the world is witnessing a dramatic rethink of chip architecture and new chip capabilities with more emphasis on specialized semiconductors that are focused on performing specific tasks.
Apple is leading the way, as usual, and blazing new ground with their M1 chip that’s earning rave reviews.
Weisenthal and Alloway speak with Doug O’Laughlin, a former buy-sider, who now writes the newsletter Mule’s Musings, on the industry and other things in tech.
MacDailyNews Take: Certainly, Apple is not only at the cutting edge of a revolution in chips, they’re leading it.
You can subscribe to “Odd Lots” via Apple Podcasts here.
From what I understand of the way Apple’s M1 risc-type processor works, if AMD, Intel and the like develop cisc-type processors that also use a single size instruction/data byte size, the overhead involved with that ‘layer’ can be eliminated from the current x86 family moving the PC-chip makers closer to M-series speeds. Definitely a rethink for the industry.
Just came across an article about a processor Nvidia announced a year ago (DRIVE AGX Orin) that could be a competitor to the M1, Imagine how much further they may have developed it in the past year.
https://www.anandtech.com/show/15245/nvidia-details-drive-agx-orin-a-herculean-arm-automotive-soc-for-2022
Apple may not have as much leeway to ‘relax’ as we may have thought.
We often see technology within shipping Apple products being compared to technology being announced by other companies. What matters is what you can buy and take home.
Nvidia may be working on all sorts of great things and others too, but Apple isn’t sitting back and congratulating itself. The successors to M1 must surely already be in mass production ready for the next product and work on the chips after that will also be nearing completion.
Yes, but the article I linked was an announcement last year and just last week I came across an article about Kia vehicles including Nvidia’s chips as the core processor in their new lineup of vehicles’ systems, outwardly primarily infotainment. The point is Apple is not as far ahead of the pack as it first seemed with the M1’s announcement which made it seems years ahead.
I’m sure Apple is following Intel’s model of having 2-3 tech teams leapfrogging each other. I believe the current i3-i9 series Core CPUs are 9th generation. This means Intel has 2 teams each working on the 10th and 11th generations for a year or more already.
Apple are not following Intel’s model. If Intel were doing things right, Apple would not need to create their own chips in order to get the results they want.
Apple has always worked in such a way that when one product is launched, it’s successor has already been fully developed and the generation after that has been designed in detail. Now that they make their own CPUs, they need to get their silicon designs ready well ahead of their hardware or software plans, so need to be working to a detailed plan stretching several years into the future.
The NVIDIA comparison is moot. Apple does not make embedded chips for other devices except their own. And designing a chip for a single purpose, such as a vehicle component is infinitely simpler than one that will run a complex operating system that supports an infinite number of devices and applications.
The crazy thing is, even the base M1 chip Mac notebook can run the ARM version Windows far faster than ARM based Surface devices.
The problem in the PC world is that you have three major forces (developers, operating system, computer mfg’s) in a co-dependent ecosystem that are highly resistant to big, bold moves.
I believe that the next 16″ MacBook Pro and iMac Pro with these chips will be equally stunning.
I take it you didn’t actually read the article linked nor how much broader Nvidia is beyond GPUs and graphics boards. The Nvidia chip described there is a full blown SoC very much in line with what the M1 represents. Apple will ignore Nvidia at its peril.
I’m just hoping Apple is able to take advantage of Apple Silicon by producing more powerful chips each year. Once the competition sees Apple slacking off, they’ll do everything in their power to design or implement better chips to stop Apple from growing market share. Apple needs to keep pressing with greater performance core counts and getting a discrete GPU running in its high-end computers. Apple will have the economies of scale on its side and not have to pay any other chip manufacturer for processors, or have to wait on any chip makers timeline.
Apple has plenty of cash to spend and can tie its own hardware and software together which should help keep Apple ahead of the competition. I think consumer x86 processors are going to fade away for the most part if Apple shows that its own ARM designs are superior to x86 chips. I’d also really hate to see Apple lose out some other ARM chip maker like Qualcomm. Apple should try to offer Apple Silicon Macs at some reasonable price points to gain market share and turn Windows users into macOS users.
So far, I haven’t heard any rival chip makers say, “There’s no way Apple can come in with Apple Silicon and offer a superior processor after we’ve been designing processors for decades.” I’ve been waiting to hear some stupid statement like that but maybe Intel has been caught by surprise.
Very good points Magnificent. Apple has so many past projects that were initial leaders and then they sat back and watched them wither on the vine. Let’s hope that they keep the momentum rolling on these chips and leave competition blinking in the dust they leave behind.
Wow! The battery life on my one month old MacBook Pro is phenomenal when compared with my 2 year old MacBook Air or my older (6 years) Pro. It lasts easily for 12 hours and even more. I would have preferred the old arrangement of Function Keys on the top row rather than the Touch Bar but you can’t have everything. The M1 chip and the Terrabyte HDD make life sweet. The fingerprint login works fast and has been failproof.
I have an iPhone 12 Pro and the SE from last April and they are snappy and efficient although the fingerprint feature on the SE is often glitchy for some reason despite my having re-done all the fingerprints. Oh well, it still works quite well in all other aspects.
Apple has a clear advantage at the moment for sure. For a first gen PC SoC to perform so well it is amazing. So that lead over others will be large for some time. I’m sure others will try to copy Apple’s approach (Samsung of course and probably Google) but Apple has now built a lot of skill set within the company and the design of the chips is a critical part in getting speed and efficiency. Also tying up with TMSC means that Samscum can no longer copy the designs and create rip offs.
As we have seen with the iPhone, Apple continue to lead the way after 10+ years. They own the profit share of the market and force others to scrap of the low end for the most part.
I feel that we will see this with Apple Silicon Macs. These will become vastly superior to Intel PCs very quickly and the rest of the industry will have catch up with their own version of SoC.
In the end I think this is the beginning of the end of Intel X86 chips in PCs.
Perhaps the PCs will start using Nvidia’s chips. Considering that they now own ARM and also use TMSC for fabrication and are also the cutting edge for creating AI SoCs they’ll be the new big competitor for Apple in the chip ‘race’.
Another possibility may be Nvidia DPU processors working in tandem with x86 in future PCs.
https://www.infoworld.com/article/3587769/nvidia-pushes-into-a-wider-application-ecosystem.html?utm_source=Adestra&utm_medium=email&utm_content=Title%3A%20Nvidia%20pushes%20into%20a%20wider%20application%20ecosystem&utm_campaign=InfoWorld%20US%20%20All%20Things%20Cloud&utm_term=Infoworld%20US%20Editorial%20Newsletters&utm_date=20201224020457&huid=818de676-0b15-4ffd-8f87-39b227f7a286
Apple should create a server version of the Mac Mini. With the recent discovery of foreign hacking of vital cyber systems in government and wariness of Chinese made computer and telecom products, I believe a big market will develop for US made systems. Apple could claim a large share given the superiority of its Mx chips.
In the server space at cloud and supercomputer levels they will certainly meet stiff competition with Nvidia’s established footprint. Nvidia is also an international corporation based in the U.S.
But no one speaks of the Apple Tax for their RAM (and likewise their storage). The architecture is definitely a winner BUT nothing warrants the exorbitant premium for their extra storage. Alas, for RAM, there is no way around it, unlike with the old architecture.
And it is much harder to anticipate what RAM needs might be over the life of the product, as needs can and do change. So you are forced to predict, and pay a premium, well in advance.