Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Apple, Intel have reached preliminary chip-making deal

reuters.com

Post reply on HN
Print view View original

141–150 of 150 posts

Re: Apple, Intel have reached preliminary chip-making deal

#141

Post by deepskier »

aurareturn wrote: post #114
adrian_b wrote: post #113

Earlier quoted context omitted.

I do not know on what data you base your sentence "Intel's own Panther Lake CPU tile is on 18A and it's extremely disappointing in terms of perf/watt and raw perf." Panther Lake does not have great raw performance, because for now Intel has not succeeded to obtain in their new 18A CMOS process clock frequencies as high as they get in the older TSMC 3-nm process used for their previous Arrow Lake H CPU generation and…

All efficiency data can be found here: https://www.notebookcheck.net/Intel-Panther-Lake-Core-Ultra-... The most important one for efficiency is ST perf/watt. MT perf/watt is largely based on how many cores there are. You can achieve better MT perf/watt simply by having more cores (more transistors) and run them at lower clocks. Panther Lake also has an entirely new MT config with 3 tiers of cores vs 2 for Arrow Lake.…

ARL is the relevant comparison point. LNL was specifically designed and targeted for low power. It's not just the node but the design that is relevant. Even for single core perf, the design dictates which other parts of the chip can be turned off. LNL was optimized there at the expense of total performance. PTL has 1.7x-2x multi core performance vs LNL. These are apples and oranges.

Re: Apple, Intel have reached preliminary chip-making deal

#142

Post by aurareturn »

deepskier wrote: post #141
aurareturn wrote: post #114

Earlier quoted context omitted.

All efficiency data can be found here: https://www.notebookcheck.net/Intel-Panther-Lake-Core-Ultra-... The most important one for efficiency is ST perf/watt. MT perf/watt is largely based on how many cores there are. You can achieve better MT perf/watt simply by having more cores (more transistors) and run them at lower clocks. Panther Lake also has an entirely new MT config with 3 tiers of cores vs 2 for Arrow Lake.…

ARL is the relevant comparison point. LNL was specifically designed and targeted for low power. It's not just the node but the design that is relevant. Even for single core perf, the design dictates which other parts of the chip can be turned off. LNL was optimized there at the expense of total performance. PTL has 1.7x-2x multi core performance vs LNL. These are apples and oranges.

When a new node is introduced, it will usually say something like "+15% at ISO power" or "-30% at ISO performance".

The fact that Panther Lake could not beat LNL on ST perf/watt is telling of the node.

PTL most likely derived its design on LNL. Overall, Arrow Lake was a complete disaster of a design. PTL was a better high performance design for sure and uses different core config for MT.

Re: Apple, Intel have reached preliminary chip-making deal

#143

Post by ahartmetz »

01100011 wrote: post #68
ahartmetz wrote: post #33

Earlier quoted context omitted.

Yes, it's called industrial policy and it can work very well.

Yes and it can also reduce competitive forces which were driving Intel to innovate. The goal of a robust supply chain is not aligned with the goal of technical supremacy. Sure, the US did achieve technical supremacy in the past with government intervention and assistance, but the world was much different then. Now the US has to compete with East Asian innovation.

Asian innovation often is the result of industrial policy to get started. Once it gets going, the state can and should let go and let market forces work.

I don't think that the basic situation has changed in that regard. China is currently trying hard to catch up in semiconductor manufacturing and AI. It seems to be working somewhat for AI (Qwen), we need to wait and see about semi manufacturing. A curious and long-term failure so far is civil aviation.

Re: Apple, Intel have reached preliminary chip-making deal

#144

Post by ahartmetz »

modeless wrote: post #123
oezi wrote: post #111

Earlier quoted context omitted.

How does Global Foundries fit into this? Are they still a thing?

They stopped pursuing cutting edge fabrication processes many years ago. Always seemed like a short sighted business decision to me.

I read about it at the time. Apparently, every new node was twice as expensive to develop as the previous one, which killed off the smaller competitors one by one. Even today, I find it hard to tell whether Globalfoundries really made a mistake. They get to cash out on their existing fabs and a bit of sustaining innovation for a long time instead of potentially killing themselves swiftly by trying and failing to keep up with the leading edge.

Re: Apple, Intel have reached preliminary chip-making deal

#145

Post by modeless »

ahartmetz wrote: post #144
modeless wrote: post #123

Earlier quoted context omitted.

They stopped pursuing cutting edge fabrication processes many years ago. Always seemed like a short sighted business decision to me.

I read about it at the time. Apparently, every new node was twice as expensive to develop as the previous one, which killed off the smaller competitors one by one. Even today, I find it hard to tell whether Globalfoundries really made a mistake. They get to cash out on their existing fabs and a bit of sustaining innovation for a long time instead of potentially killing themselves swiftly by trying and failing to keep…

Considering the market cap of their competitors who kept going it seems foolish to have given up entirely.

Re: Apple, Intel have reached preliminary chip-making deal

#147

Post by aurareturn »

brookst wrote: post #137
aurareturn wrote: post #92

Earlier quoted context omitted.

Nvidia’s chips aren’t usually on the latest nodes. Not yet. The primary reason is because most AI chips are full reticle sized which means the first year yields likely won't be very cost effective. It takes a new node a few years to fully mature in terms of yield. Little iPhone A series and server CPU chiplets are perfect for new nodes. That said, Nvidia will certainly try to move smaller and lower volume chips in fu…

Agreed on all counts. Apple’s hedge with Intel makes a ton of sense, especially if they continue to pay a premium for first dibs on a significant chunk of TSMC’s newest nodes. So far Apple has been aggressive about stopping production of older processors, but as we see with MacBook Neo, even the lowest chips are increasingly overpowered for many users. I expect even if Intel can’t match TSMC’s latest, they’ll be able…

Going with Intel might mean higher cost. Intel has lower yields and making chips in the US is much costly than Taiwan. You can see this in TSMC Arizona chip costs vs Taiwan.

Re: Apple, Intel have reached preliminary chip-making deal

#148

Post by xnx »

dwaite wrote: post #139
bpoyner wrote: post #24

Earlier quoted context omitted.

I imagine it would be a big lift. Asahi linux is managing through reverse engineering the hardware support, without any official documentation. Even with official documentation it would be a significant change from other aarch64 hardware.

I'm curious why I would want to dual boot Windows when there's a perfectly good hypervisor and paravirtualization built into macOS? Arm-based windows support via Parallels does work, but AFAICT there's no official way to buy a Windows license due to a Microsoft/Qualcomm partnership.

No RAM, CPU, or disk overhead.

Re: Apple, Intel have reached preliminary chip-making deal

#149

Post by NoSalt »

bryceacc wrote: post #72
NoSalt wrote: post #50

What year is this? What happened to the M1, M2, ... MN chips of Apple's? Is Apple going to go with Motorola after Intel??? Obviously I couldn't read the article due to it being paywalled.

If you can't read, why comment? >It is unclear which Apple products Intel would make chips for, according to the report. Intel and Apple declined to comment.

It was a joke. Sad your sense of humor left you about the same time as your manners.

Re: Apple, Intel have reached preliminary chip-making deal

#150

Post by evancox100 »

dwaite wrote: post #139
bpoyner wrote: post #24

Earlier quoted context omitted.

I imagine it would be a big lift. Asahi linux is managing through reverse engineering the hardware support, without any official documentation. Even with official documentation it would be a significant change from other aarch64 hardware.

I'm curious why I would want to dual boot Windows when there's a perfectly good hypervisor and paravirtualization built into macOS? Arm-based windows support via Parallels does work, but AFAICT there's no official way to buy a Windows license due to a Microsoft/Qualcomm partnership.

You can buy an ARM Windows license for parallels on Mac now. It works quite well.

I think WoA used to be exclusive to Qualcomm, but that hasn’t been true for at least a couple years, maybe longer.

Post reply on HN