Big tech is in the kind of business where the 3rd, 4th, or 5th item in the chain of Toyota's "Five Whys" can have such far reaching effects. In a way, this is nothing new. The auto industry has been in that kind of business for a very long time now. This is why Toyota's "Five Whys" exists.
I don't think the root of the problem here is technical, however. I suspect it's ultimately cultural. Didn't Andrew Grove say, "Only the paranoid survive?" How hard would it be to maintain a culture of paranoia at an industry leading company with huge margins? That's the kind of position that fosters overconfidence, not paranoia.
AMD's Epyc Master plan: (Or how AMD let Intel paint itself into a corner)
That statement is beyond hyperbolic...
If the chipsets are simpler, less critical and easier to design than CPUs, shouldn't they start new processes with the least critical stuff?
You can design around unreliable processes by increasing redundancy and these simpler chips seem excellent as technology validation.
Also, easier on the RTL isn't really easier on the process. The digital guys and the process guys have very different constraints, so there's probably not even a benefit to having "simpler" chips on a newer process first.
I assume a 10nm Quark X would be hard to see with the naked eye ;-)
Unfortunately, your response doesn’t really deal with the point I made: not only does it apply equally to both companies, but also I cannot think of any economically significant use case where Intel will beat AMD while costing more ($ or watts) per gflop.
Can you provide one example of where Intel products beat AMD's in latency in a way that prospective customers are willing to pay a premium for that performance delta?
With that kind of problem, you may want to abandon x86 altogether.
At this end of the spectrum, that means abandoning commodity hardware, with everything it implies, not just a change in binary architecture, as it might for ARM vs. x86 in end user devices.
Recently, I looked up the max memory on IBM's system z, and it was only 32TB, but it looks like their Power Systems E980 can take 64TB. It appears that, like with their mainframe line, going that route means significant vendor lock-in, including proprietary memory modules.
It's certainly interesting that the current generation of high-end x86 CPUs took a step backwards from the last generation, in that that their max is 12TB and just catching back up to 24TB isn't expected to happen until next year.
For some reason, while I was in college, I got a complimentary subscription to a magazine called Supercomputing Review. It was one of the most interesting reads of that time.
It's a bit like x86 vs SPARC vs POWER vs ARM. Each has workloads that run better on them for the same amount of upfront and operational costs.
AMD is less than 50% the price at large core count models, relative to comparable to Intel, so even with half the SIMD pipeline they can still win on $/gflops there.
It isn't really the same as considering entirely different ISAs. The switching cost Intel -> AMD is far cheaper than switching Intel -> PPC. Sparc and ARM are not competitive at all.
AMD full-year 2017 revenue: $5.3 billion
You see, Intel is gasping for breath, drowning in the $18 billion of profit they'll generate this year. How they'll survive another year at this rate is anyone's guess.
http://brucefwebster.com/2008/04/15/the-wetware-crisis-the-t...
Are you sure the problem couldn't be interpreted as, "We've set the die sizes too large?" Because that's how AMD is approaching it, and they're winning.