Apple Said to Work on Mac Chip That Would Lessen Intel Role • Hacker News Board

Re: Apple Said to Work on Mac Chip That Would Lessen Intel Role

#261

Post by r00fus » Fri, Feb 03, 2017, 9:41 PM UTC

scarface74 wrote: post #232

gcp wrote: post #223

Earlier quoted context omitted.

Only in theory. In practice the existing base of software in the App Store would cause compatibility issues. Intel based Android devices have an ARM emulator on board (libhoudini) and claim to be ARM when querying the store, in order to not have a software offering disadvantage.

I'm not saying it's a good idea, but we are talking about Apple allowing IOS apps run natively on Macs. iOS apps are already running natively on macOS - but only if you have Xcode. Apple could really just include the x86 based iOS framework that it already has in the next version of macOS and tell developers if you want to run on Macs, just modify your UI to support a third target - you are already probably targeting…

I'd be amazed if Apple ever gave up ARM for x86. Has Intel improved that much in comparison to ARM for power efficiency?

Re: Apple Said to Work on Mac Chip That Would Lessen Intel Role

#262

Post by roryisok » Sat, Feb 04, 2017, 1:02 AM UTC

__jal wrote: post #139

roryisok wrote: post #71

Earlier quoted context omitted.

No reason they couldn't put a touch screen on a MacBook or imac, though I guess they can't guarantee one on a mac pro

Well, one reason is that touch screens for desktop OSes suck. Holding your arm out screen operations for any length of time is awful, and wiping the prints off your screen on your pant leg doesn't work very well on a laptop. That's not to say Apple doesn't release terrible things at times (hockey puck mouse[1]?), but I have trouble imagining Apple releasing one of those. (Laplet? tabletop?) [1] I actually think that…

I have a hp spectre with a touch screen and windows 10, and it certainly doesn't suck. Great for sticking in tablet mode and browsing, organising photos etc.

Re: Apple Said to Work on Mac Chip That Would Lessen Intel Role

#263

Post by foobiekr » Wed, Feb 08, 2017, 4:45 PM UTC

wbl wrote: post #120

foobiekr wrote: post #100

Earlier quoted context omitted.

Why does intel need to do this? x86 decode is a tiny, tiny part of the overhead of modern x86 CPUs. Architecture for the most part at this point doesn't matter unless you are talking about truly novel approaches like the Mill. Arm certainly is not a significant improvement.

It's more then just decode. Segmentation, x87 emulation, and the interaction with other features are all pain points.

None of that stuff matters in reality.

You should think about ISA implementations as verification problems instead of problems in building a silicon implementation of the ISA. From that perspective it should be obvious that intel has the best, richest, deepest verification set that exists in the CPU space and since it is tied to x86, that is an advantage for the two x86 vendors. Verification is incredibly hard for complex CPUs and building up verification is time intensive. This is part of the reason that the brainiac end of the design spectrum has pretty much become just x86 and POWER which has nearly the same history.