Read more.Custom ARM IP integration is going ahead smoothly, A6X is a sign.
Read more.Custom ARM IP integration is going ahead smoothly, A6X is a sign.
I think ARM has a long way to go before that happens and it will depend on what the rest of the market does first. The moves away from 68K and PPC were basically forced weren't they? 68K and PPC could no longer offer competitive performance / cost rather than Apple deciding to go with another architecture "because they could"?
(edit: for core 'PC' hardware anyway - from which the domestic market will probably move increasingly away from, towards tablets, etc which are much more open at the moment and will probably stay that way - though I have a feeling Intel will get their strategy right in that space in the next few years)
Last edited by malfunction; 06-11-2012 at 01:26 PM.
Nobody's going to argue that x86 isn't a bit long in the tooth and was probably due being thrown out at least a decade ago, but is ARM really there yet? On the one hand the move does sound crazy, but there's no doubt that people are moving towards tablets and other ARM devices instead of the humble PC/laptop so maybe it really isn't so crazy.
Given that most mac users just browse facebook before looking at their latest pictures from instagram, I think this will be great for them.
Small thinner lighter. Obviously the hundred odd pounds in CPU, Chipset, Battery, HSF cost saving will be past on to them right.... Right?
The other thing is when I first used a StrongARM RISC PC, it was demonstratably faster than the Intel PC I had at the time.
throw new ArgumentException (String, String, Exception)
That would probably end gaming on Macs.
At least at the moment they have bootcamp, once they switch to ARM they will only have Modern-UI apps available via dualboot (if and when WinRT works on an ARM Mac)
Main PC: Asus P8Z77 WS / 3570k @ 4.4GHz / 8GB Vengeance Black / GTX 780 Ti / Areca 1680 / HX 850 / 600T / K60 / M60 / 2x Dell 3007 / 2 x 256GB Samsung 830 (RAID0) / 2 x 240GB Corsair Force 3 (RAID0) / Windows 8.1
HTPC: AsRock Z77 Pro 4 / E3-1230v2 / 8GB XMS3 / GTX 780 / Tevii S480 / SST-LC20 / Antec TP-550 / PS50C6900 / 128GB Kingston V200 SSD + 3 x 1.5TB + 1 x 3TB / Windows 8.1 x64 Pro with WMC
HTPC2: Asus AM1I-A / 5150 / 4GB DDR3 / Corsair Force 3 240GB / Silverstone SST-ML05B + ST30SF / Windows 8.1 x64 Pro with WMC
Spare/Loaner: Gigabyte EX58-UD5 / i950 / 12GB RAM / GTS 450 / Corsair 300R / Silverpower 700W modular
Server Setup: HP DL160 G6 / 2 x E5620 / 64GB RAM / 2 x 300GB SAS (RAID1) / 6 NICs / ESX 5.5
2 x ESX 5.5 Nodes: Asus M5A78L-M/USB3 / AMD FX 6100 / 16GB XMS3 / 160GB SATA HDD / 5 NICs
NAS 1: HP N40L / 10GB RAM / 2x 2 x 3TB + 80GB Intel SSD (Hybrid) || NAS 2: HP N40L / 10GB RAM / 2x 2 x 3TB + 80GB Intel SSD (Hybrid) || Network: TL-WR1043ND w/DD-WRT + HP ProCurve 1800-24G
Laptop: Thinkpad T510 / 4GB RAM / 240GB Corsair Force 3
I'd be surprised if they did move away from x86 until at least Windows is supported on ARM. I use a lot of Apple kit, but this would be a game changer for me. I need to be able to use some Windows apps for work for which there are no OSX compatibles (Visio etc), and I also like to be able to boot into Windows for a bit of gaming.
I like OSX, it gives me a lot of the functionality of Linux out of the box and a tightly integrated core set of apps on top too. Stability is pretty good, although so is Windows 7 now (I think I've only had one or two bluescreens in the 3 years I've been using it).
A move away from x86, however, would likely mean a wholesale move back to Windows for me. Despite the hyperbole I still don't like Linux on the desktop.
I now expect this thread to descend into all Mac users are 'tards etc as per usual.
Even the most badass ARM architectures are slow compared to the modern Intel CPUs in Macs. Judging by the Chromebook results an A15 looks to be around the same performance as a Pentium-M or something, i.e. faster per core/MHz than Atom but a long way off even the low power 17W Core chips and A15 is still 32bit (with some memory addressing enhancements).
Upscaling ARM architectures to Intel/AMD performance would require a lot more juice, they wouldn't be the low power chips we're familiar with anymore. Intel have shrunk Atom down to match ARM chips (see Motorola Razr i) but ARM licensees haven't shown much interest in making desktop chips to compete in the mid-high end, only in "basics" segment and some flirtation with low power or highly parallel servers. There must be a reason why it's not really being done on a large scale... and Haswell is probably going to make the job of competing with Intel in non-handheld markets a lot harder next year, nevermind 2017.
Apple could in theory use some superclocked A15 quad in the Air in next few years and just about get away with it, but for anything better they would need to do years of R&D, and it's not like Intel or AMD will be just sitting around twiddling their thumbs. I don't fancy their chances of upscaling an ARM design to match or better what Intel or even AMD are churning out by 2017.
This rumour is one of those bullsh&t ones made up by clueless analysts who caught some ARM buzz or read some frothy fanboy's rants on a forum.
Last edited by kingpotnoodle; 06-11-2012 at 02:14 PM.
Would you not then have the issue that software is developed for either ARM or x86 so you would then have a software divergence?
Apple would need to find someone other than samsung to make there soc's for this. Only global foundries (amd) might have the capacity and ability to do this.
Personally I can see where this is going. In the news recently: ARM announce 64-bit chips; AMD announce ARM processors; Everyone is buying mobile devices (tablets up/desktops down); now Apple mull ARM desktop replacement. and iOS/Android is already on ARM. I call that a "trend".
http://www.extremetech.com/computing...r-room-in-2014). And remember we're talking about laptops which, unless you're in the niche of the "barely luggable desktop replacement", are usually highly focussed on weight and power/performance - especially the latter. So I'm not so sure that we could chalk this up as an automatic fail for the ARM gear - who knows what we'll be up to with our portable kit in five years time?
As far as I see it, Apple's got three major advantages, one - they've done an architecture switch before, so presumably know some/most of the pitfalls; two - ARM architectures are a known tech for them - so not scary; three - they've got utter control on the environment, it's their OS on their hardware.
Of course with my cynical hat on, perhaps this is just a trial balloon to see if they could get away with ARM, or do MBP owners need/want that Intel poweeerrrrr (a la Jeremy Clarkson), or mayhaps this is just a focussed rumour to get some concessions from Intel in the next round of supplier negotiations?
Its one of programming choices, current main stream applications are generally huge and very bloaty, not helped by the excessive use of frameworks and several layers of API.
That said, for some tasks, there will always be the need for something vastly more powerful, its why companies still use super and large mainframeesque computers.
Most likely, should Apple go down the ARM route, all the home use machines would be ARM, whilst the high end/graphic design machines would be Intel CPUs still.
Cortex A57 is still only a 20% boost over A15, presumably at equal clocks/cores, it won't be displacing Intel's main range anytime soon but may find a niche in many core servers, but the UltraSPARC T2 indicates that approach is hardly mass market. Boston's Viridis range use Calxeda A9 based CPUs and have very good compute density with modest power requirements but the many wimpy cores model doesn't suit an awful lot of workloads and almost none that relate to desktop use.
Apple would have to be a lot more ambitious than ARM themselves are to be able to use ARM based chips in their Mac line-up inside 5 years. We'd have to know because the would be hoovering up chip design talent and their cash pile would be falling... surely?
Then there's the fact that if they go ARM then there's less 3rd party upgrade options.
If there's no option to chuck in a discrete gfx card or upgrade the memory via standard DDR3 (or 4 by then) Dimms then you're into the i(Device pick a number) no upgrade other then buy a new one system.
"Look at the the iMac 3 it's got everything you loved about the iMac 2 but with the extra 4 Gig of ram and a 20% clockrate increase for only $1399.95" followed 12 months later by "Look at the the iMac 4 it's got everything you loved about the iMac 3 but with the extra 8 Gig of ram and a 10% clockrate increase for only $1429.99"
There are currently 1 users browsing this thread. (0 members and 1 guests)