Intel's disruption is now complete • Hacker News Board

Re: Intel's disruption is now complete

#71

Post by p_l » Sat, Nov 14, 2020, 5:15 PM UTC

whatever1 wrote: post #40

My question is why Qualcomm while using the same arch, has not produced a chip as fast as the M1?

You're missing the wildly different target that Apple can follow.

Apple designs chips and devices in lockstep. When designers were working on various A-series chips, they knew the details of the device, and could optimise on that.

Qualcomm doesn't get that kind of view. They make a generic CPU design for given market segment, and have to "please" much wider buyer range. As such, if they tried to apply some of the things that are publicly known about Apple designs, they would probably end up with a chip that would be hard to sell due to taking more space or having higher TDP etc. And before advent of multicore, Qualcomm was using the available space budget heavily to integrate the radio functionality internally (then it oscillated a bit over time depending on chip and market segment).

Re: Intel's disruption is now complete

#72

Post by eecc » Sat, Nov 14, 2020, 5:16 PM UTC

lgessler wrote: post #58

agumonkey wrote: post #48

Earlier quoted context omitted.

It's somehow incredible how this infiltrates even the most advanced classes of company in terms of technical knowledge. You'd expect a bit more objectivity

Indeed, for we computer geeks are known for our lack of ego...

Absolutely, not. Seeing my 4 years old son experiment with social skills I’m learning soo much about the most inexplicable conflict situations I’ve been through in 15 years of IT

Re: Intel's disruption is now complete

#73

Post by JumpCrisscross » Sat, Nov 14, 2020, 5:16 PM UTC

ThrowawayR2 wrote: post #69

agumonkey wrote: post #48

Earlier quoted context omitted.

It's somehow incredible how this infiltrates even the most advanced classes of company in terms of technical knowledge. You'd expect a bit more objectivity

The thing about the top tech companies, particularly Intel and other chip companies where postgraduate degrees are more or less required to do anything interesting, is that people there have literally gone though over a decade of intense competition in the educational system to qualify to get their foot in the door. And then, to stay there and rise through the ranks is another decade or more of intense competition, b…

> people there have literally gone though over a decade of intense competition in the educational system to qualify to get their foot in the door. And then, to stay there and rise through the ranks is another decade or more of intense competition

Two decades of requisite grunt work will also encourage your best junior talent to seek out start-up opportunities.

Re: Intel's disruption is now complete

#74

Post by eecc » Sat, Nov 14, 2020, 5:22 PM UTC

baybal2 wrote: post #62

SoSoRoCoCo wrote: post #43

Earlier quoted context omitted.

I think they have been a commodity for a long time. Rephrased, the discussion is similar to, "Will GE make a comeback in the refrigerator market?" Intel isn't going anywhere, they have many, many big customers to satisfy who simply cannot shift to ARM due to ROIs <=0. And after 30+ years of this, I also don' think there's anything new or revolutionary in silicon fabrication on the horizon. I stopped getting excited a…

> And after 30+ years of this, I also don' think there's anything new or revolutionary in silicon fabrication on the horizon. Why so? Few more material, and device generations are on the way. Litho has went through more generation changes in the last 1 decade, than in any previous one. Whole new device classes are on the way: optics on silicon, new memory, logic in the backend, and logic on package. Semiconductor eng…

> Semiconductor engineering is still a career suicide though, that haven't changed in the last few decades.

Wow, so I need to thank myself for accidentally bumbling towards ops and dev? Meh, for me it’s just that in EU there’s a total dearth of opportunities in the industry

Re: Intel's disruption is now complete

#75

Post by skohan » Sat, Nov 14, 2020, 5:22 PM UTC

mortenjorck wrote: post #70

usrusr wrote: post #34

Earlier quoted context omitted.

Purely for themselves, or also for entering the cloud infrastructure market. In this day and age if you are big enough (Apple surely is) and happen to have superior server hardware it wouldn't really make sense to sell that hardware, you'd rather keep it and rent it out.

An Apple alternative to AWS would be interesting, but it would probably be heavily geared toward iOS and Mac app backends, with little flexibility outside those use cases. The Xcode integration would probably be amazing, though.

It's hard to imagine Apple really getting the cloud market, or their product offering would look like. Cloud by necessity has to be pretty agnostic and flexible to meet the diverse needs of customers. Apple hasn't really demonstrated an ability to succeed with this type of product, and leans heavily toward very tailored, limited, "apple knows best" types of products.

Re: Intel's disruption is now complete

#76

Post by zarkov99 » Sat, Nov 14, 2020, 5:22 PM UTC

SoSoRoCoCo wrote: post #20

As someone who worked on Intel's phone chip: we definitely didn't win it. We fucked it up twelve ways to Sunday. Why: giant egos. There were turf wars between Austin, Santa Clara and Israel over who would design it, and the team that won out had long since lost its best principle engineers and had no clue how to spin the architecture to meet the design win. Otellini's hindsight hedge is pure spin: we knew the landing…

How much of the blame do you think is Otellini's? I got a really bad feeling about Intel when he became CEO.

Re: Intel's disruption is now complete

#77

Post by Kapura » Sat, Nov 14, 2020, 5:22 PM UTC

I found this very interesting, and it seems a similar thing is happening to Intel in the DIY/gaming space right now. For a long time, the Intel i5/i7/i9 were the processors of choice for gaming, but AMD's Ryzen line has been making remarkable inroads, with the recently released Ryzen 5000 series boasting performance that competes or tops Intel's top chips in the space, and generally with lower power draw to boot.

I'm not a tech journalist, but the way they seem to have done this was exactly what was described in this essay. AMD brought out "chiplet" designs, initially at the low end, but with significant performance gains generation over generation. Combined with simpler motherboard slot conventions, delay's in Intel's next-generation processes, and overperforming Intel in areas like streaming, the value case for buying AMD processors in this space has grown to the point where you would find very few enthusiasts who would strongly recommend Intel over AMD if you're building a new system and wanted the best performance at your price point.

Seems the infighting issues described by another poster have not only affected Intel's competitiveness in the mobile space.

Re: Intel's disruption is now complete

#78

Post by baybal2 » Sat, Nov 14, 2020, 5:22 PM UTC

whatever1 wrote: post #40

My question is why Qualcomm while using the same arch, has not produced a chip as fast as the M1?

They don't need to, they are in a much more down to earth market niche.

Re: Intel's disruption is now complete

#79

Post by jeffbee » Sat, Nov 14, 2020, 5:23 PM UTC

raverbashing wrote: post #52

jeffbee wrote: post #26

Earlier quoted context omitted.

I'm still waiting for my M1 to show up but from studying the details that other people have commented on, it's clear that Apple is slathering on the microarchitectural resources that both Intel and AMD are stingy with. Their core has a gigantic L1 instruction cache, rather a lot of L1 TLB entries (4x more than Intel), large(r) pages by default, and many other aspects. L1 TLB entries in particular are a well-known pro…

Interesting that you mention page size, IMHO 4k was small even 10 years ago. Though of course the next bump is 4M which I think might be too big for most apps Apparently ARM supports 4KB, 64KB and 1MB (I just googled that) Also maybe all the x86 crud might be finally taking its toll (and not just the ISA but all the cruft that got on top of it for 40yrs+ like legacy buses and functionalities, ACPI, or even the overen…

All the legacy junk that's actually on the CPU is tiny. Even the x86 instruction decode you could lose in a corner of a 512x512 FMA unit.

Re: Intel's disruption is now complete

#80

Post by redvenom » Sat, Nov 14, 2020, 5:24 PM UTC

dboreham wrote: post #2

What's the real reason Apple can design a higher performance device? Better fab process? ARM ISA is fundamentally faster? Graph conveniently omits AMD who now beat Intel using the same ISA but different process.

It's also the fact that they are putting everything on one chip. With an Intel chip it's just the CPU.