Steve Jobs on P.A. Semi, love for Intel; 3G Blackberry delayed

135

Comments

  • Reply 41 of 94
    mr. hmr. h Posts: 4,870member
    Quote:
    Originally Posted by echosonic View Post


    MY BIG-TIME PREDICTION #1 ON RECORD HENCEFORTH:



    Within 5-8 years, Apple will be developing their own CPUs, and will leave intel much they way they left Motorola.



    This will put them into a position to fully control every aspect of Mac design and production from Chip to Chassis, and remove their need to rely on any other company less well-run or secure than they are.



    It will also remove for another 10-15 years the possibility of Apple suffering the same stability and hacker fates that Windows has suffered, as well as prevent companies like Psystar from knocking off cheap clones.



    If you can't get a proprietary Apple CPU, how can you possibly build a clone?



    In fact, I would not be surprised if Apple was already developing a new OS to run on their own proprietary CPU platform as we speak.



    Nope. They already tried that. You do know that PPC was co-designed by Apple? PPC failed. The reasons it failed are twofold:



    1.) Manufacturing. Motorola and IBM couldn't keep up with Intel on the manufacturing side of things. Intel were always at least one feature-size node ahead, and that's very significant. It takes billions and billions of dollars of investment to stay on the bleeding edge of die manufacturing, and that is beyond Apple's financial capability.



    2.) Code optimisation. The PPC architecture was superior to x86. But that means nothing if people write shit code for it. Most major pieces of third-party code were ported from x86-based codebases for Windows. Ported badly. Resulting in poorly-optimised code that ultimately resulted in the Mac platform being "slower" than the Windows platform despite the Mac having superior hardware*.



    It's clear that x86 has conquered mainstream computing for good.





    * Toward the end of PPC, equally speedy hardware despite it being one feature-size node behind Intel.
  • Reply 42 of 94
    Hey, does anyone know how important hardware is for real-time machines?



    I mean - a real time system responds instantly (in real time). Very useful for nuclear power plants to be able to shut down a problem instantly, rather than after they finish calculating whatever else they were working on.



    Similarly, Apple's iPhone responds very quickly and naturally when you touch the screen (not instant, but fast)



    I'm wondering, with PA Semi's military experience, to what degree Apple would look to hardware to continue to make an incredibly responsive iPhone or other mobile wifi device.
  • Reply 43 of 94
    murphywebmurphyweb Posts: 295member
    I am sorry to disappoint everyone BUT blackberry have had a 3G handset for two years now. The 8707g supports 3G, though from reading this story am guessing that maybe it just never suported 3G on the US version. The Blackberry Curve that was released last year also has 3G. I used the 3G 8707g myself last year in Australia, though now have the 8120 Pearl which is not 3G but is edge & wifi.



    I have got no idea why someone would write this article to make it seem that there is an issue with Blackberry releasing a 3G handset when they seemed to have done fine with it outside of the US.
  • Reply 44 of 94
    iq78iq78 Posts: 256member
    Could it be possible that Apple is looking for a way to distinguish it's hardware from typical PC hardware by adding co-processor technology that would be fully supported by OSX, making it scream on Apple hardware?



    Once the distinction between running OSX on Apple hardware (with co-processors) and running it on non-Apple hardware is made, it could open the doors to license a non-coprocessor supported version of OSX on PCs? A Trojan horse method of getting a huge spike in OSX users while still trying to protect their hardware sales. They wouldn't have to be seen as "crippling" OSX on PCs because it would be merely that PCs don't have the coprocessors in place that OSX supports.



    Nah, probably not. But certainly, if there would be a time for Apple to try to take control of the OS war, the time would be soon. Apple's growth has been great lately, and they have a potential to be huge in the mobile market. If there was anytime for people to jump ship away from Microsoft and towards Apple, now would be the time. Getting people to install OSX on their current PC hardware (without a new purchase) would help Apple as long they could grab a huge percent of these users next time they purchased hardware. Is there a way that Apple could distinguish their hardware to grab enough of the market to justify licensing their OS to other vendors?



    Probably not... They will probably want to distinguish themselves in hardware to lure people WITHOUT having to license their OS. But I'm just throwing the thought out there.



    I AM IQ78
  • Reply 45 of 94
    mr. hmr. h Posts: 4,870member
    Quote:
    Originally Posted by GregAlexander View Post


    Hey, does anyone know how important hardware is for real-time machines?



    I mean - a real time system responds instantly (in real time). Very useful for nuclear power plants to be able to shut down a problem instantly, rather than after they finish calculating whatever else they were working on.



    Similarly, Apple's iPhone responds very quickly and naturally when you touch the screen (not instant, but fast)



    I'm wondering, with PA Semi's military experience, to what degree Apple would look to hardware to continue to make an incredibly responsive iPhone or other mobile wifi device.



    Real-time stuff generally doesn't hinge on specialist hardware. But I guess it depends what you mean by "specialist".



    If you are designing a real-time system, you generally use a real-time operating system, unless you can get away with no OS at all. The software design needs to go hand-in-hand with the hardware design, including choice of CPU. Obviously there are limits on what software can achieve, you do need to make sure your CPU is fast enough to do what you want to do. Depending on what you want to do, a mainstream CPU like an x86 or PPC might be the right choice, alternatively it might be a specialist DSP such as a Texas Instruments TMS320xxxx series, or Analog Devices Blackfin, or a specialist chip with an ARM core, etc. etc.



    OS X's kernel does have a few clever tricks up its sleeve to deliver low latencies for time-sensitive stuff such as audio, but it's not a real-time OS or a real-time kernel.



    The iPhone is responsive simply due to a decent kernel and decent code optimisation.
  • Reply 46 of 94
    solipsismsolipsism Posts: 25,726member
    Quote:
    Originally Posted by IQ78 View Post


    Could it be possible that Apple is looking for a way to distinguish it's hardware from typical PC hardware by adding co-processor technology that would be fully supported by OSX, making it scream on Apple hardware?



    Once the distinction between running OSX on Apple hardware (with co-processors) and running it on non-Apple hardware is made, it could open the doors to license a non-coprocessor supported version of OSX on PCs? A Trojan horse method of getting a huge spike in OSX users while still trying to protect their hardware sales. They wouldn't have to be seen as "crippling" OSX on PCs because it would be merely that PCs don't have the coprocessors in place that OSX supports.



    Nah, probably not. But certainly, if there would be a time for Apple to try to take control of the OS war, the time would be soon. Apple's growth has been great lately, and they have a potential to be huge in the mobile market. If there was anytime for people to jump ship away from Microsoft and towards Apple, now would be the time. Getting people to install OSX on their current PC hardware (without a new purchase) would help Apple as long they could grab a huge percent of these users next time they purchased hardware. Is there a way that Apple could distinguish their hardware to grab enough of the market to justify licensing their OS to other vendors?



    Probably not... They will probably want to distinguish themselves in hardware to lure people WITHOUT having to license their OS. But I'm just throwing the thought out there.



    I AM IQ78



    That is a pretty interesting idea.
  • Reply 47 of 94
    mdriftmeyermdriftmeyer Posts: 7,503member
    Raise your hand if you thought PA was going to be used as the CPU for existing Macs?



    I better see no hands raised.



    Steve Jobs just reinforced that relationship.



    Raise your hands if you think Steve Jobs just reinforced this relationship for products other than Macs?
  • Reply 48 of 94
    solipsismsolipsism Posts: 25,726member
    Quote:
    Originally Posted by mdriftmeyer View Post


    Raise your hand if you thought PA was going to be used as the CPU for existing Macs?



    I better see no hands raised.



    Steve Jobs just reinforced that relationship.



    Raise your hands if you think Steve Jobs just reinforced this relationship for products other than Macs?



    1) How will you know if our hands are raised?



    2) How do you know question we are answering with our raised hands?
  • Reply 49 of 94
    mr. hmr. h Posts: 4,870member
    Quote:
    Originally Posted by IQ78 View Post


    Could it be possible that Apple is looking for a way to distinguish it's hardware from typical PC hardware by adding co-processor technology that would be fully supported by OSX, making it scream on Apple hardware?



    Once the distinction between running OSX on Apple hardware (with co-processors) and running it on non-Apple hardware is made, it could open the doors to license a non-coprocessor supported version of OSX on PCs? A Trojan horse method of getting a huge spike in OSX users while still trying to protect their hardware sales. They wouldn't have to be seen as "crippling" OSX on PCs because it would be merely that PCs don't have the coprocessors in place that OSX supports.



    That's a nice idea and I'd like to see Apple do it, but I don't think it's going to happen.



    I remember in 1995 or so, before Steve returned to Apple, Apple were somewhat more open about what they were working on for the future, and I read in an Apple-published magazine that they were working on PowerMacs with co-processors for media acceleration. It never saw the light. I'll see if I can dig anything up online...
  • Reply 50 of 94
    solipsismsolipsism Posts: 25,726member
    Quote:
    Originally Posted by Mr. H View Post


    That's a nice idea and I'd like to see Apple do it, but I don't think it's going to happen.



    I remember in 1995 or so, before Steve returned to Apple, Apple were somewhat more open about what they were working on for the future, and I read in an Apple-published magazine that they were working on PowerMacs with co-processors for media acceleration. It never saw the light. I'll see if I can dig anything up online...



    Cringely has been saying it's going to happen for years now.



    Also in Cringely's prediction portfolio: Complete WiMax coverage, portable and numerous Google data centers built into shipping containers, and free super-highspeed internet from Google from all the dark fiber they've bought up.
  • Reply 51 of 94
    mr. hmr. h Posts: 4,870member
    Quote:
    Originally Posted by Mr. H View Post


    I remember in 1995 or so, before Steve returned to Apple, Apple were somewhat more open about what they were working on for the future, and I read in an Apple-published magazine that they were working on PowerMacs with co-processors for media acceleration. It never saw the light. I'll see if I can dig anything up online...



    Ah ha! It was a couple of years later than I thought, but the co-processors were called "TriMedia". Here's a few pages for those that are interested:



    CNET news article



    Page with lots of details of a prototype TriMedia card



    See bottom of this page
  • Reply 52 of 94
    solipsismsolipsism Posts: 25,726member


    That link is worth a look.
  • Reply 53 of 94
    noirdesirnoirdesir Posts: 1,027member
    Quote:
    Originally Posted by NasserAE View Post


    Maybe. But the attitude of most people here and other sites, Europeans in particular, is this "iPhone sucks because it is not using 3G". No one said the same thing about RIM.



    Quote:
    Originally Posted by Solar View Post


    That's because RIM makes 3G products, this report is specifically talking about the new Blackberry 9000, for AT&T's new HSDPA 3G network, on which the iPhone will also be running. The phrasing in the AI article kind of makes it sound like it's RIM's first 3G product, when it's really their first HSDPA 3G product. (I think)



    Maybe missing GSM 3G support (if that is correct) or missing HSDPA support was not as important to RIMs main customers: corporations, ie, customers who care about actually useful feature like great e-mail service.



    Companies targeting the consumer market might have had to please more the 'armchair' analysts who analyse products on specs alone and not on the actual usefulness of features.
  • Reply 54 of 94
    bsenkabsenka Posts: 801member
    Quote:
    Originally Posted by federmoose View Post


    I'm missing something here. Mac sales BOOMED partially as a result of moving to intel, or moreover their ability to run Windows OS. That ability is a key feature for switchers, and Apple would be throwing that away if they were to move to their own chips.



    I've used Macs for as long as I can remember, and the move to intel was a great day for me; I would be pissed if apple were to move to their own chips. As a developer, the ability to run multiple OS's (I have Windows XP Pro SP2 and Ubuntu Client 64-bit installed) allows me to use a Mac as my work computer. I'd hate to see that feature go.



    We do not know what any new chip will or will not be able to do, OS support included.



    I don't believe for a second that Apple's recent successes have anything whatsoever to do with whether Macs can run Windows.
  • Reply 55 of 94
    mr. hmr. h Posts: 4,870member
    Quote:
    Originally Posted by bsenka View Post


    I don't believe for a second that Apple's recent successes have anything whatsoever to do with whether Macs can run Windows.



    (there needs to be a smilie with eyes popping out of the head).



    You think Apple's huge uptick in sales has nothing to do with Macs now being able to natively run Windows? That's bananas.



    Do you agree that it's got something to do with x86? Do you agree that if they had stuck with PPC, the Mac would be doing much worse than it is now?
  • Reply 56 of 94
    mdriftmeyermdriftmeyer Posts: 7,503member
    Quote:
    Originally Posted by solipsism View Post


    1) How will you know if our hands are raised?



    2) How do you know question we are answering with our raised hands?



    Just use your textual imagination.



    Did you hear me ask the question?



    :Raising hand high over here:



    That was tough.
  • Reply 57 of 94
    SpamSandwichSpamSandwich Posts: 33,407member
    Quote:
    Originally Posted by echosonic View Post


    MY BIG-TIME PREDICTION #1 ON RECORD HENCEFORTH:



    Within 5-8 years, Apple will be developing their own CPUs, and will leave intel much they way they left Motorola.



    This will put them into a position to fully control every aspect of Mac design and production from Chip to Chassis, and remove their need to rely on any other company less well-run or secure than they are.



    It will also remove for another 10-15 years the possibility of Apple suffering the same stability and hacker fates that Windows has suffered, as well as prevent companies like Psystar from knocking off cheap clones.



    If you can't get a proprietary Apple CPU, how can you possibly build a clone?



    In fact, I would not be surprised if Apple was already developing a new OS to run on their own proprietary CPU platform as we speak.



    It's possible that for Apple's future plans involving nano-scale engineering this is the quickest route for Apple to get an "in"... In another 5 to 10 years, nano-tech will be far more advanced and there will be more MEMS integration in handheld devices.
  • Reply 58 of 94
    Quote:
    Originally Posted by Mr. H View Post


    You think Apple's huge uptick in sales has nothing to do with Macs now being able to natively run Windows? That's bananas.



    I know that's not directed at me - but let me add my voice to yours in saying the Windows compatibility concept has been very important in Mac mindshare.



    I do think that Apple could easily technically go down a dual architecture track if they felt like.... have some PPC models in the mix - MacBook Air PPC or a 5Ghz Power server if they felt like it. I don't think they will, I don't think it's worth it.



    Quote:
    Originally Posted by SpamSandwich View Post


    It's possible that for Apple's future plans involving nano-scale engineering this is the quickest route for Apple to get an "in"... In another 5 to 10 years, nano-tech will be far more advanced and there will be more MEMS integration in handheld devices.



    I have to say that's the most intelligent speculation on this purchase that I've read on this or any blog. Thank you.
  • Reply 59 of 94
    backtomacbacktomac Posts: 4,579member
    Quote:
    Originally Posted by solipsism View Post


    I wholeheartedly disagree. I think a large part of the switch to Macs is because of the potential for Boot Camp and visualization of Windows. I doubt Intel is scared of this deal in relation to Apple's Mac line.



    I think you misunderstand me.



    I don't think Apple will abandon Intel cpus fro laptops/desktops/server products.



    It does appear to me that they may not adopt Atom and its line of cpus for the iPhone/iTouch product lines. Many felt that it was only a matter of time, and a die shrink to 32 nm, before the iPhone would begin using Atom.



    That seems to be seriously in question now.
  • Reply 60 of 94
    solipsismsolipsism Posts: 25,726member
    Quote:
    Originally Posted by backtomac View Post


    It does appear to me that they may not adopt Atom and its line of cpus for the iPhone/iTouch product lines.



    Then I wholeheartedly agree with you. Atom power output can not compete with ARM at this point. I believe the slowest Atom chip in the works is 800MHz, which is almost double the current under-clocked speed of the iPhone at 412MHz.



    Even if the iPhone was clocked to it's 620MHz maximum it would still be slower. But more importantly, it would be much more power efficient that the Atom chip which still killing the battery. I have to agree with AnandTech that it's 5-10 years before Intel would be able to have anything that can compete with ARM.
Sign In or Register to comment.