Home News Forum Articles
  Welcome back Join CF
You are here You are here: Home | Forum | Apple to transition Macs to their own chipset

You are currently viewing our boards as a guest which gives you limited access to view most of the discussions, articles and other free features. By joining our Virgin Media community you will have full access to all discussions, be able to view and post threads, communicate privately with other members (PM), respond to polls, upload your own images/photos, and access many other special features. Registration is fast, simple and absolutely free so please join our community today.


Welcome to Cable Forum
Go Back   Cable Forum > General Discussion > Science & Technology

Apple to transition Macs to their own chipset
Reply
 
Thread Tools
Old 24-06-2020, 22:43   #16
downquark1
cf.mega poster
 
Join Date: Jun 2003
Age: 33
Services: Plusnet FFTC
Posts: 4,649
downquark1 has a bronzed appealdownquark1 has a bronzed appeal
downquark1 has a bronzed appealdownquark1 has a bronzed appealdownquark1 has a bronzed appealdownquark1 has a bronzed appealdownquark1 has a bronzed appealdownquark1 has a bronzed appealdownquark1 has a bronzed appealdownquark1 has a bronzed appealdownquark1 has a bronzed appealdownquark1 has a bronzed appeal
Re: Apple to transition Macs to their own chipset

You have to understand if an ARM chip exceeds the x86 in floating point performance without gpu offloading then the x86 is defacto obsolete. Which means AMD will be brought to it's knees and intel will be on life support.

The only people who will be buying x86 will be retro gamers.
__________________
"Knowledge is Power. Power Corrupts. Study Hard. Be Evil."
downquark1 is offline   Reply With Quote
Advertisement
Old 24-06-2020, 22:56   #17
Damien
Remoaner
Cable Forum Team
 
Damien's Avatar
 
Join Date: Mar 2004
Posts: 29,715
Damien has a lot of silver blingDamien has a lot of silver blingDamien has a lot of silver blingDamien has a lot of silver bling
Damien has a lot of silver blingDamien has a lot of silver blingDamien has a lot of silver blingDamien has a lot of silver blingDamien has a lot of silver blingDamien has a lot of silver blingDamien has a lot of silver bling
Re: Apple to transition Macs to their own chipset

Quote:
Originally Posted by downquark1 View Post
You have to understand if an ARM chip exceeds the x86 in floating point performance without gpu offloading then the x86 is defacto obsolete. Which means AMD will be brought to it's knees and intel will be on life support.

The only people who will be buying x86 will be retro gamers.
We'll see but people are excited about ARM so Intel should be concerned. The warning signs have been there for a while now, first when the iPhone started really started getting impressive numbers and then when the iPad Pro started dominating in multi-score tests as well as single-core ones.

The iPad Pro is genuinely more powerful than most consumer laptops.

If Apple prove ARM to work very well on proper laptops then Intel will be in serious trouble for the consumer market.

They might be saved by compatibility issues with the software. Apple is well-suited for such transitions because they don't have too many businesses running 20-year-old software on modern macs. Their control over the platform for the software that is run means they can get developers over quickly to the new platform. Microsoft will find this transition harder.

I don't know enough about servers but my understanding is x86 will survive there for a while yet according to people who do know. Xeon chips just seem to be well suited to that task.
Damien is offline   Reply With Quote
Old 01-07-2020, 11:45   #18
Kushan
FORMER Virgin Media Staff
 
Kushan's Avatar
 
Join Date: Dec 2010
Location: Warrington
Posts: 4,638
Kushan has a bronzed appealKushan has a bronzed appeal
Kushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appeal
Re: Apple to transition Macs to their own chipset

Quote:
Originally Posted by Damien View Post
Developers are under an NDA to not benchmark or report benchmarks from the development units they're getting.

It's possible they are very capable processors and the reason Apple do not want raw numbers out there is they're good enough to depress sales of the existing Mac range.

It's also possible they're very good and Apple doesn't want competitors to react yet. Let's say they clock in at 50%+ improvements over their equivalent predecessors, that would shake the laptop industry and Microsoft will make a much bigger push into ARM than they have done so far.
I feel there's a 3rd, much more obvious (and in my mind, likely) possibility - they're simply not as fast as x86.

x86 vs ARM is not a new thing, it's something that has been debated for years and years. They're ultimately both very good at different things but apples to apples, ARM is way more efficient at lower power draw and x86 is way more powerful (And expensive!) at higher power draws.

We've seen both sides of this - we've seen x86 in embedded devices, like phones and got great performance with awful battery life. We've also now got ARM in the server space, which (shock horror) isn't quite as fast as its x86 counterparts but works out much cheaper overall as they're cheaper to run (And presumably cheaper to buy). Keep in mind those benchmarks were against slightly older-gen x86, but even then it's clear where ARM shines and where it doesn't.

I would not expect to see performance gains here at all, that's not what ARM has ever been about , but..comparable performance is still nothing to be sniffed at.

Apple tends to have some pretty awful thermal designs in their machines, so you might even see better performance just from sheer efficiency, but Apple aren't going to be looking to knock Intel or AMD's performance crown off any time soon.

Keep expectations in check, competition is always a good thing.
__________________
My views are my own and in no way reflect those of Virgin Media.
Kushan is offline   Reply With Quote
Old 01-07-2020, 12:18   #19
downquark1
cf.mega poster
 
Join Date: Jun 2003
Age: 33
Services: Plusnet FFTC
Posts: 4,649
downquark1 has a bronzed appealdownquark1 has a bronzed appeal
downquark1 has a bronzed appealdownquark1 has a bronzed appealdownquark1 has a bronzed appealdownquark1 has a bronzed appealdownquark1 has a bronzed appealdownquark1 has a bronzed appealdownquark1 has a bronzed appealdownquark1 has a bronzed appealdownquark1 has a bronzed appealdownquark1 has a bronzed appeal
Re: Apple to transition Macs to their own chipset

What I foresee happening is Apple implementing APIs for offloading things to the GPU, adobe will quickly adapt to establish market share (as it seems they already have high performing software). Initially open source software will be frozen out by lack of manpower leaving them with crap performance for a while.

Maybe after a few years it will have evened out depending on how open the apple standard is.
__________________
"Knowledge is Power. Power Corrupts. Study Hard. Be Evil."
downquark1 is offline   Reply With Quote
Old 01-07-2020, 12:20   #20
Damien
Remoaner
Cable Forum Team
 
Damien's Avatar
 
Join Date: Mar 2004
Posts: 29,715
Damien has a lot of silver blingDamien has a lot of silver blingDamien has a lot of silver blingDamien has a lot of silver bling
Damien has a lot of silver blingDamien has a lot of silver blingDamien has a lot of silver blingDamien has a lot of silver blingDamien has a lot of silver blingDamien has a lot of silver blingDamien has a lot of silver bling
Re: Apple to transition Macs to their own chipset

Quote:
Originally Posted by Kushan View Post
I feel there's a 3rd, much more obvious (and in my mind, likely) possibility - they're simply not as fast as x86.

x86 vs ARM is not a new thing, it's something that has been debated for years and years. They're ultimately both very good at different things but apples to apples, ARM is way more efficient at lower power draw and x86 is way more powerful (And expensive!) at higher power draws.

We've seen both sides of this - we've seen x86 in embedded devices, like phones and got great performance with awful battery life. We've also now got ARM in the server space, which (shock horror) isn't quite as fast as its x86 counterparts but works out much cheaper overall as they're cheaper to run (And presumably cheaper to buy). Keep in mind those benchmarks were against slightly older-gen x86, but even then it's clear where ARM shines and where it doesn't.

I would not expect to see performance gains here at all, that's not what ARM has ever been about , but..comparable performance is still nothing to be sniffed at.

Apple tends to have some pretty awful thermal designs in their machines, so you might even see better performance just from sheer efficiency, but Apple aren't going to be looking to knock Intel or AMD's performance crown off any time soon.

Keep expectations in check, competition is always a good thing.
I think there would be surprise if Apple doesn't outperform the equivalent intel chips when they first come out. After all they're posting very competitive scores with their phones and iPads at the moment and those devices have to consume less power and generate less heat than their laptop counterparts.

The advantages may only come from efficiency but that's sort of the point. In laptops and anything that doesn't have the luxury of being able to generate a lot of heat and a lot of power it may be that ARM can get better performance than x86 for the same heat/power draw.
Damien is offline   Reply With Quote
Old 02-07-2020, 16:43   #21
Kushan
FORMER Virgin Media Staff
 
Kushan's Avatar
 
Join Date: Dec 2010
Location: Warrington
Posts: 4,638
Kushan has a bronzed appealKushan has a bronzed appeal
Kushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appeal
Re: Apple to transition Macs to their own chipset

Quote:
Originally Posted by Damien View Post
I think there would be surprise if Apple doesn't outperform the equivalent intel chips when they first come out. After all they're posting very competitive scores with their phones and iPads at the moment and those devices have to consume less power and generate less heat than their laptop counterparts.

The advantages may only come from efficiency but that's sort of the point. In laptops and anything that doesn't have the luxury of being able to generate a lot of heat and a lot of power it may be that ARM can get better performance than x86 for the same heat/power draw.
I think there's going to be a big distinction between Apple's ARM chips compared to the intel equivalent and Apple's ARM chips compared to previous Apple implementations of those chips.

This is something I watched recently and is related to the subject. Essentially, Apple's recent devices haven't been particularly great at thermal management of those chips. I think at one point in the video Linus speculates if it's deliberately bad so that when it comes to the ARM chip switch, it doesn't look like you're loosing much (any?) ground. While that's a throwaway comment and not meant to be taken seriously, I do think there's possibly a sliver of truth to it.

So you'll have this situation where you'll see comparable or even better performance than the previous gen Apple machines, but realistically the performance won't actually beat x86 if x86 has adequate cooling. In other words Apple will use murky comparisons to skew things in their favour. And maybe that's fine, because you can argue that it's a strength of ARM to require less cooling, but you can tell from the Video that Apple really half-arsed their cooling.
__________________
My views are my own and in no way reflect those of Virgin Media.
Kushan is offline   Reply With Quote
Old 02-07-2020, 18:29   #22
Uncle Peter
81-82-83-84
 
Uncle Peter's Avatar
 
Join Date: Nov 2006
Location: on holiday by mistake
Age: 50
Services: Vivid 200, Full House, V6 x2
Posts: 5,967
Uncle Peter has a nice shiny starUncle Peter has a nice shiny starUncle Peter has a nice shiny star
Uncle Peter has a nice shiny starUncle Peter has a nice shiny starUncle Peter has a nice shiny starUncle Peter has a nice shiny starUncle Peter has a nice shiny starUncle Peter has a nice shiny starUncle Peter has a nice shiny star
Re: Apple to transition Macs to their own chipset

Quote:
Originally Posted by Kushan View Post
I think there's going to be a big distinction between Apple's ARM chips compared to the intel equivalent and Apple's ARM chips compared to previous Apple implementations of those chips.

This is something I watched recently and is related to the subject. Essentially, Apple's recent devices haven't been particularly great at thermal management of those chips. I think at one point in the video Linus speculates if it's deliberately bad so that when it comes to the ARM chip switch, it doesn't look like you're loosing much (any?) ground. While that's a throwaway comment and not meant to be taken seriously, I do think there's possibly a sliver of truth to it.

So you'll have this situation where you'll see comparable or even better performance than the previous gen Apple machines, but realistically the performance won't actually beat x86 if x86 has adequate cooling. In other words Apple will use murky comparisons to skew things in their favour. And maybe that's fine, because you can argue that it's a strength of ARM to require less cooling, but you can tell from the Video that Apple really half-arsed their cooling.
The thermal advantage only exists if Apple don't cram too much stuff on the SOC, ie: Co-processors, encoding chips etc etc. Cramming more components onto a tiny surface area doesn't really bode well for dissipating heat.

We don't really know much at this point what is going into the new machines. The only information which seems to be out there are some supposed leaked benchmarks from the developer kits Apple have been sending out. This is probably the most informative article:

https://arstechnica.com/gadgets/2020...ransition-kit/

Main things to note:

- The developer kit is a Mac Mini running a hobbled A12X SOC from the iPad Pro (half of the cores disabled)

- Benchmarks are from non-native (x86) Geekbench tests via the Rosetta 2 interpreter

- Results: 20% slower benchmarks than an entry level i3 Macbook Air on single thread performance

- 38% faster for multi threading

Which is not too bad really if you factor in the fact there's a performance penalty associated with not running native ARM code.
__________________
Nero fiddles while Gordon burns... in his Joy Division oven gloves
Uncle Peter is offline   Reply With Quote
Old 02-07-2020, 19:10   #23
Kushan
FORMER Virgin Media Staff
 
Kushan's Avatar
 
Join Date: Dec 2010
Location: Warrington
Posts: 4,638
Kushan has a bronzed appealKushan has a bronzed appeal
Kushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appealKushan has a bronzed appeal
Re: Apple to transition Macs to their own chipset

Yeah I remember reading that article a few days ago. It's certainly looking like Rosetta is at least decent at what it does, but there's so many caveats and questions it's hard to really draw many conclusions.

I'm going to try anyway, though!

Assuming that it's legitimate and somewhat indicative of performance, assuming overhead for the Rosetta2 implementation, etc. I still think it's in line with what I'm saying - these processors aren't going to be faster than x86, apples to apples. The single-threaded performance is a dead giveaway, even accounting for the Rosetta overhead. If it was emulation we were talking about then that would be one thing, but recompiled code shouldn't be 30% slower.

I do wonder why they've disabled 4 of the 8 cores of the A12, presumably the disabled cores are the little low-power cores, maybe they just don't have a fully working OSX scheduler that can handle big.LITTLE CPU's so it's easier to just fudge a homogenous chip for now.

You could have 8 beefy full cores instead, plus with general manufacturing process improvements, etc. that could be quite the beefy chip when all is said and done, but again even accounting for all that, there's nothing that suggests Apple has suddenly beaten the likes of Intel and AMD at things like IPC. Core to Core, I just don't see Apple beating big blue or AMD any time soon.

The only thing that could possibly swing it is accounting for Rosetta's overhead (Which as I said, I doubt is that big), or Apple going for multithreaded performance with a 12 or 16 core SoC - but even still, Intel and AMD especially have gone there and beyond and I would expect 8 core CPUs to be pretty mainstream in 2 years time so again, it's not really an area I see Apple competing in.

I genuinely believe Apple is doing this to give themselves more control over their ecosystem. They'll be able to make thinner, lighter devices with better battery life than the competition, they'll be able to push a universal app ecosystem that more easily allows developers to develop native apps for iOS and desktop with the same codebase (Something that others have tried and so far failed to do well), that will perform well enough but isn't going to be the top performer out there - and it doesn't need to be.
__________________
My views are my own and in no way reflect those of Virgin Media.
Kushan is offline   Reply With Quote
Old 02-07-2020, 19:27   #24
downquark1
cf.mega poster
 
Join Date: Jun 2003
Age: 33
Services: Plusnet FFTC
Posts: 4,649
downquark1 has a bronzed appealdownquark1 has a bronzed appeal
downquark1 has a bronzed appealdownquark1 has a bronzed appealdownquark1 has a bronzed appealdownquark1 has a bronzed appealdownquark1 has a bronzed appealdownquark1 has a bronzed appealdownquark1 has a bronzed appealdownquark1 has a bronzed appealdownquark1 has a bronzed appealdownquark1 has a bronzed appeal
Re: Apple to transition Macs to their own chipset

There's reports that apple did this because intel faffed them around with skylake. I imagine the heterogeneous bandwagon is also instrumental. Since they have so much control over their eco system they are in the best position to leverage it.
__________________
"Knowledge is Power. Power Corrupts. Study Hard. Be Evil."
downquark1 is offline   Reply With Quote
Old 03-07-2020, 09:06   #25
Damien
Remoaner
Cable Forum Team
 
Damien's Avatar
 
Join Date: Mar 2004
Posts: 29,715
Damien has a lot of silver blingDamien has a lot of silver blingDamien has a lot of silver blingDamien has a lot of silver bling
Damien has a lot of silver blingDamien has a lot of silver blingDamien has a lot of silver blingDamien has a lot of silver blingDamien has a lot of silver blingDamien has a lot of silver blingDamien has a lot of silver bling
Re: Apple to transition Macs to their own chipset

Quote:
Originally Posted by Uncle Peter View Post
https://arstechnica.com/gadgets/2020...ransition-kit/

Main things to note:

- The developer kit is a Mac Mini running a hobbled A12X SOC from the iPad Pro (half of the cores disabled)

- Benchmarks are from non-native (x86) Geekbench tests via the Rosetta 2 interpreter

- Results: 20% slower benchmarks than an entry level i3 Macbook Air on single thread performance

- 38% faster for multi threading

Which is not too bad really if you factor in the fact there's a performance penalty associated with not running native ARM code.
That's Apple clearly capping the developer unit because they knew people would benchmark it. They didn't just disable the cores, they seem to have underclocked it as well
Damien is offline   Reply With Quote
Old 03-07-2020, 10:18   #26
Uncle Peter
81-82-83-84
 
Uncle Peter's Avatar
 
Join Date: Nov 2006
Location: on holiday by mistake
Age: 50
Services: Vivid 200, Full House, V6 x2
Posts: 5,967
Uncle Peter has a nice shiny starUncle Peter has a nice shiny starUncle Peter has a nice shiny star
Uncle Peter has a nice shiny starUncle Peter has a nice shiny starUncle Peter has a nice shiny starUncle Peter has a nice shiny starUncle Peter has a nice shiny starUncle Peter has a nice shiny starUncle Peter has a nice shiny star
Re: Apple to transition Macs to their own chipset

Quote:
Originally Posted by Damien View Post
That's Apple clearly capping the developer unit because they knew people would benchmark it. They didn't just disable the cores, they seem to have underclocked it as well
Yes it seems likely and the fact that the low power cores to which the article relates have been disabled this could be a nod to the fact that these will not be used in the chips which will power the RTM units when they finally surface. Also, perhaps some of the embedded image processing stuff which is used by mobile device cameras won't be required.

Looking at the bigger picture though, some of the tech which goes into these new generations of processing engines is mind bending stuff so it's difficult to make meaningful comparisons with the olde worlde tech we used to use in our systems to play Doom and Sim City.
__________________
Nero fiddles while Gordon burns... in his Joy Division oven gloves
Uncle Peter is offline   Reply With Quote
Reply


Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +1. The time now is 12:29.


Server: curium.zmnt.uk
Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2020, vBulletin Solutions Inc.