Intel has been badly managed for decades, from a personnel management perspective. People put up with it because it was freaking Intel and you’ll put up with a lot to work for the best.
When they’re no longer the best, however...
Intel was fine when Apple first switched over to them. The beginning of Core (and especially Core 2) was when they had started to recover from the disaster that was NetBurst (which, honestly, isn't too dissimilar to the spot of bother they're in now with not being able to move past "Lake" microarchitectures). It all fell apart post-Haswell. Broadwell was a mess, Skylake had quality issues like crazy, and then the best that they've been able to do since then is shrink the die with Ice Lake to 10nm? I get that Tiger Lake is supposed to bring dramatic improvements in this department, but compared to the kind of innovation we were seeing from 2006-2014, it's not much at all.
It does risk sending the company's chip business into a tailspin though; less money from selling chips means less money to invest in improving their chips, means it's even harder to catch up, leading to less money from selling chips... the company overall probably won't disappear, but Nokia is still around too, just not in the main business that made them famous (excl. HMD licensing the name).
Again, I only proposed them shrinking down to just Xeon because that's a product that seems to still be okay on the whole. Nor is anyone particularly worried about thermal efficiency on a server CPU the way they might be on small form-factor desktops and notebooks of all sizes. Plus, it would give them the opportunity to focus on what they're good at, which seems to be their reasoning for ditching various arms of their business (though, I would've still imagined that they had a bunch of other business units that weren't x86 still making them money).
In 2019 the storage business reported a $1.2B operating loss. It’s there largely to help defray fixed costs on fabs, as far as I can tell. If you look at their businesses, just about all of their money is from processors or from devices that support their processors.
As far as not leaving Xeon any time soon, it’s not good when your customers are only buying your products until the consumer technology wave reaches their business too...
I'm not sure I agree there. Intel doesn't need to put in the kind of work to increase thermal efficiency with Xeons that they do with H/HQ, U, or even Y series mobile Core i9, Core i7, Core i5, or even Core i3 CPUs. I'm not saying that they need to be stagnant, but performance per watt isn't as critical of a concern for the kind of market that the Xeon is designed for by comparison. I don't doubt that Epyc isn't making headways, but it's not as clear-cut of a good idea as Ryzen Pro processors (when compared to their vPro Intel equivalents) and even that isn't as clear-cut of a no-brainer as consumer-focused Ryzen processors are for non-business-use consumer PCs and kits.
I do believe that Ryzen Pro is going to successfully take on vPro Intel CPUs in the business space. Dell is one hold-out and even they're taking the temperature of the room.
Epyc's take-over of the datacenter won't be for a good while. If it's a resources issue (and it really sounds like it is), Intel might be better off downscaling for what they're still good with. Again, it seems like they're already trying for that approach.
This article also leads me to believe that Intel isn't ceding the Server crown (like it's having to with the desktop and notebook crown currently) anytime soon:
What does Arm's new N1 architecture mean for Windows servers? - TechRepublic
Good riddance to x86 processors in Apple computers. Now Apple can finally make their products thin and not have thermal throttling issues. I'd still like to see Apple put large batteries in their laptops, so I'd gladly accept the additional weight and thicker case. Apple is definitely going to have the thinnest desktops available and they're going to be more powerful with Apple Silicon than any consumer-based Intel processor. Yay!
Apple has definitely made decent modification to ARM64 in Apple Silicon, but this is a byproduct of the ARM64 architecture being more efficient and of Apple's increasing desire for thinness. Note that no one was ever complaining about the performance of MacBook Airs from 2011-2015 because, they were pretty good for the technology that was out at the time. (Comparable ARM SoCs were not up to snuff at that point in time.) It isn't that x86 sucks, it's that ARM has gotten better while x86 has stayed stagnant (mainly because AMD was playing catchup and Intel was running around in circles dancing to the tune of their own hubris).
Sandy Bridge (2nd gen Core) was a pretty leap, they’ve really just been iterating ever since.
Sandy Bridge was huge. I'd say that original Core 2 was similarly huge. Similarly Haswell and Skylake were also big deals. But those four instances were it. Skylake was their "peaked in college" moment. Though, the parallels between it and NetBurst are striking. They got themselves out of that mess. Who's to say they won't get themselves out of this one?
If we're trying to make the point that ARM64 should succeed x86-64, we're not there yet because no other ARM64 processor manufacturer is achieving performance levels like Apple is. When the rest of the ARM64 processor market catches up, then it will be time to call time-of-death to x86 for the consumer market. I'd still argue that they're not going anywhere for the server market.
The 7 nm screwup is too big to hide. I think it's game over at this point. It's just about how quickly they disintegrate. I held INTC for a long time but sold in the middle of this year when the CEO said they were thinking of using TSMC. That basically sent the message of "we've screwed up so bad we have no idea how to fix it" or, in other words, the people that knew how to make it work are no longer at Intel.
Again, see NetBurst. AMD had a good lead with the Athlon 64 and the Athlon x2 from about 2002 to about 2006, which is when Intel was pretty much a sputtering mess with the Pentium 4. This is not dissimilar. The only scary thing is that this bad spell seems to have lasted for longer. But that could simply mean that Intel needs to stop being stubborn about their manufacturing.
Sadly Intel is hopelessly behind Apple. Look at the M1. When could we have realistically expected similar performance at the power usage of the M1 from Intel???
That's not Intel's fault as much as it is x86 shortcomings compared to ARM's room for growth. I'm not saying that Intel isn't screwing up, because they totally are. But it's not like AMD, being as amazing as they are right now with their x86 chips, would've been much better. ARM has just gotten good enough to switch to. Apple makes these transitions as much for roadmap reasons as it does for up-front benefits (if not much moreso). It is in this respect, more than all others, that this is similar to the PowerPC to Intel move.
re b: Oct'20 earnings statement reads: Announced agreement to sell Intel NAND memory and storage business to SK hynix for $9.0 billion
Right. I did forget about that one.
McAfee was sold in Sep 2016
Technically, Intel still has 49% ownership of McAfee.
re a: yes, but more and more Arm is showing up in the server space, recently Microsoft but also Google, Amazon etc are working on their own, and there are the likes of Ampere starting to dig into this space
I'm not saying Ampere and the like aren't still making headways, but Intel has time to focus on Xeon to at least keep the server business afloat in ways that they will eventually fail with the consumer PC market. More efficient laptop CPUs is always a demand for Intel. Apple was always pushing demands Intel would eventually struggle to meet. The rest of the laptop industry will follow. I do not believe the same phenomenon will occur in the same way in the server market, at least not for a long enough time for Intel to make headway. Furthermore, it says something when Microsoft, having previously planned an ARM64 version of Windows Server 2016 and 2019 have backpedaled and made that for internal use only. Not to say that focus isn't shifted to Azure, but they're still very much committed to the Windows Server product (as even in Azure, those are the VMs being spun up).
re c: pls read Intel's earnings announcements, not a whole lot of revenue outside of x86
Intel will continue to dominate x86 which is not going to go away any time soon, but their revenue has been mostly flat and it will further decline going forward ...
I did look at their website just now. I know that networking products are still a focus, but I concede that the Processor business seems very front-and-center. They're going to have to do something. I think they have two options. Abandon ship for everything that they don't have a fighting chance of saving (again, I believe Xeon is among this). Or do the dramatic soul-searching/restructuring that people are calling upon them to do. The latter just makes too much sense, but they appear to be stubborn. The former seems more likely given that this is what they've been doing so far.
For what it's worth, I do still see their site advertising their memory and storage business, though I guess the sale to Hynix is far from complete.
It does seem that Intel has lost focus. I was an Intel Product Dealer (IPD) and built a ton of Intel only computers for family and friends. I started the company I ran and started selling servers and lots of desktops. I thought they were making money, but they dropped the whole program. They dropped their own motherboards too. They seem to have pulled back from a lot of what made them an important company for me, and tons of people. Their conferences were usually quite well attended.
But AMD, and Apple are definitely eating their lunch. Intel failed at cellular modems, and, how much more.
Are they hoping for a buyout, or are they making more money through some other means? Government contracts perhaps?
I was saddened to find that they totally gave up on the IPD program, and on small businesses and home hobbyists. Part of me thinks they deserve to get a beating. Sorry. Dumping on those markets was not a good idea, IMO. All of the off-brand motherboards were a can of worms compared to theirs.
I had completely forgotten about Intel branded motherboards. You just triggered some happy nostalgia for those. They were always pretty decent in terms of quality (also Hackintosh support). Now, Asus is my current brand of choice. But if Intel returned to that space, I'd be all over 'em.
All this means for Mac users is a faster transition away from Intel and future macOS versions not supported anymore.
I really don't know how you figure that.
Apple is not producing new Intel Mac models at this point (especially since the previously "leaked" 2020 Intel 16" MacBook Pro has yet to surface and only has one more day to really do so). Apple is still going to support macOS on Intel for the foreseeable future (probably cutting it off after a similar 8 year interval has passed to the Catalina-able Macs that just got cut off from Big Sur).
If I had to guess, the next culling of support will be of Haswell Macs, mandating that Broadwell and newer are supported.
Then one or two versions after that, it'll be Skylake and newer Macs.
Then one or two versions after that, it'll be Coffee Lake and newer Macs (with the iMac Pro still being supported). After that, it's either done, or it'll be anything Intel 16" MacBook Pro or newer for the next version or two, and then it's done. But Intel going to hell isn't going to mean the end of Intel support on macOS.
