New IBM chip could serve a PowerBook G5, but is it too little too late?

124

Comments

  • Reply 61 of 90
    boogabooga Posts: 1,082member
    Quote:

    Originally posted by melgross

    Don't knock this new FX either. A 1.6 GHz model can compete with a 2.2GHz x86. A 2GHz G5 has been shown to compete with a 3GHz P4 quite well.



    This is a common misconception. You can't lump all "x86" chips together, or you're falling prey to the same "MHz myth" that plagued Apple all those years. A Pentium M at 2.1GHz competes reasonably well against a 3.6GHz P4 in typical tasks. GHz for GHz, the Pentium M seems to be similar or slightly faster than a G5. The area where today's Pentium M falls a little short and the G5 excels is floating point and SIMD, though, so for a few, this could have been a nice chip.



    But basically, this chip only confirms Jobs' analysis about how much a G5 has to be hobbled to be put in a laptop. Comparing the new G5, which isn't even available for purchase yet, with Pentium M's that are in widespread use, and have it still come up a little short isn't inspiring.
  • Reply 62 of 90
    boogabooga Posts: 1,082member
    Quote:

    Originally posted by aegisdesign

    For comparison the Pentium M at 1.4 is 18W typical.



    I'm curious where you got this data point. Intel only releases their Wattage for maximum, not typical, so it must be from some other testing that has produced this number. URL?
  • Reply 63 of 90
    a j steva j stev Posts: 79member
    I'm surprised that people have already forgotten the great Powerbook G5 debates over the past 12 months and the reasons why a G5 would be a fraught choice to go in the Apple laptop line. Use the search function and trawl through all the multitudinous Powerbook threads over the past 18 months



    Anybody remember the discussion of the heat/power issues with the system controller for the G5? On a Power Mac, its got a wacking great heatpipe on it. OK, you may say, but thats probably due to the size of the pipeline (1.35Ghz). Granted, but the fact that the system controller sucks on latency and power disipation tests doesn't fill me with any hope that the lower power G5 will be a good choice in a laptop. In other words, the engine may be OK, but the transmission bites.
  • Reply 64 of 90
    mynameheremynamehere Posts: 560member
    Quote:

    Originally posted by melgross

    When the word "consumer" is used, it doesn't mean "pro". We have four Powermacs at home, and in my business we had eighteen.



    Mom and dad going to the Apple store aren't looking to compare a P4 (or M) to the G5 or G4. They just want what seems to be snappy, does what they want, looks good, and isn't subject to internet garbage.



    At least that's what the surveys of people buying these machines say.



    And again i'll state, from someone who has bought a lot of Powermacs over the years; we buy our Macs based on the premise that we WANT Macs as opposed to Windows machines. We compare a new Mac to the older ones we have.



    If x86 machines are MUCH faster, maybe we would consider them instead. But they have to be MUCH faster.



    I can tell you from inside the media industry that the reason that Powermac sales have fallen is mostly because people are sitting on their current models, not because they are rushing out and buying XP. I'm sure that this has happened in a few cases, but not many. And those are usually those who were not comfortable with Macs anyway.



    In the future Apple sees the x86 line getting MUCH better than the PPC. That's a problem So they are switching.




    Apple switching to Intel might have something to do with the fact that they realized that eventually their underpowered hardware was going to come back and bite them in the ass when it could no longer run the latest and greatest apps. Eventually, those computers that Mom and Pop want WON'T feel snappy. Right now it's happening, but we're still ahead, but if hardware development doesn't pick up, then that's what will happen; there's no denying it.



    Maybe the reason people aren't letting go of their old PM's is that they realize that there's not enough of a difference to justify new powermacs.



    As for your "we haven't switched to windows yet" comment: neither have I and I don't intend to, but I grow steadily more unenthused with Apple hardware (not software, which rocks) when I see the offerings on the PC side. Like you, I wouldn't consider switching unless the PC hardware was a lot further ahead, but I think we're slowly getting to that point, and no amount of Apple PR's twisting the numbers can stop that.



    EDIT: This thread is about PB's so bringing up PM's here is rather pointless.
  • Reply 65 of 90
    mynameheremynamehere Posts: 560member
    Quote:

    Originally posted by a j stev

    I'm surprised that people have already forgotten the great Powerbook G5 debates over the past 12 months and the reasons why a G5 would be a fraught choice to go in the Apple laptop line. Use the search function and trawl through all the multitudinous Powerbook threads over the past 18 months



    Sort of a disclaimer I guess: All those threads were from before we realized what a piece of sh*t that G5 would be. I'm not trying to speak for everybody here, but I think that most people thought that it was going to be a speedy processor, comparable to those in the PM, but without the heat issues.



    Now that the specs are out in the open, we can finally have a meaningful debate that isn't entirely based on guessing.
  • Reply 66 of 90
    I see things with IBM of late as a wasted opportunity! I think if IBM got off their "butt" and got on the ball, we would have seen the G5 Powerbook with some power and 3.0 Ghz Power Macs last Summer! Is this "save face" for IBM? Who knows...
  • Reply 67 of 90
    Although there's plenty of logic in the IBM>Intel switch, and I've read much good reasoning throughout this thread, I think people are underestimating the role that ego (corporate and personal) played in this decision. Tech CEOs tend to be high-profile and intertwined with their companies (Michael Dell, Bill Gates, Andy Gore, Scott McNealy, etc.). No one is more so than Steve Jobs and his baby, Apple. Steve Jobs' ego and ire are legendary, even in this industry, and you've got to believe his ire was up, and ego bruised, by IBM's shift of focus to gaming systems. Afterall, this is supposed to be Steve's, Apple's (and the iPod's) 15 minutes of fame. How dare IBM not play the groupie to Apple's rock star like everyone else is doing these days? That had to P.O. Steve Jobs, and I bet it played a role in his decision.



    Funny thing is, IBM is one of the few tech companies (even more so than MSFT and Intel) that doesn't play the giddy school girl. IBM operates more like a Citigroup, GE, or Pfizer....we don't care, we don't have to. The even funnier thing is that this list used to included GM, AT&T, PhillipMorris...so be careful IBM, nothing lasts forever.



    Here's to Apple putting 64bit AMD chips in Macs by 2007 (I still can't bring myself to feel good about this Intel(only) Inside thing). Jobs' should know that Intel is no kinder or gentler a partner than IBM and he'd better not put all his eggs in one basket again.
  • Reply 68 of 90
    aegisdesignaegisdesign Posts: 2,914member
    Quote:

    Originally posted by Booga

    I'm curious where you got this data point. Intel only releases their Wattage for maximum, not typical, so it must be from some other testing that has produced this number. URL?



    They don't even do that. The TDP figure they quote is a design goal that the vendor has to be able to dissipate. It explicitly states it's not a maximum. I imagine it's as near as dammit though.



    For Dothan it's 27W which compares favourably with 30W for 7447A at 1.42. Since they quote 21W typical for the 7447A and less than 15W for a 1.5Ghz 7448, we can assume max on the 7448 is going to be around the 23-25W or they can just run them faster of course.



    The current laptops have seen processors consuming much more than the 7447A though. Not G5 temperatures though, or at least not till the low power G5 but those aren't any faster than the G4 really so a bit pointless.



    The 18W figure was from a test of putting dothan into a desktop I read. Toms or Anandtech IIRC.
  • Reply 69 of 90
    aegisdesignaegisdesign Posts: 2,914member
    Quote:

    Originally posted by melgross



    This isn't a new architecture for Apple after all. It would be a new mobo, but they have two years of experience with G5's, and have no doubt been working on this for a while now. It just might need to go to production.



    I really doubt very much that a 7448 machine would be faster than the 1.6 G5 unless they clocked it. something Apple has done a number of times in the past.




    I'll just take these two points since we've been over the others over and over. The Barefeats tests are the only ones out there.



    There's more to an architecture than the R&D and production. There's support, system chips (the desktop G5 system chips suck), spare parts, warranty backup, casing... It's a whole new laptop. It's major expense for Apple to come up with a new design.



    Speed wise, the G5 wins over the G4 because of it's higher clock speed and high speed FSB. On a laptop you don't have the room for either of those and apart from that the G5 support chips are frankly terrible - hot and slow. When an eMac beats a PowerMac at drive tests, you know there's something seriously wrong. That's why I think it will even up when you have a 1MB L2 on the G4 v 512K on the G5.
  • Reply 70 of 90
    melgrossmelgross Posts: 33,510member
    Quote:

    Originally posted by Booga

    This is a common misconception. You can't lump all "x86" chips together, or you're falling prey to the same "MHz myth" that plagued Apple all those years. A Pentium M at 2.1GHz competes reasonably well against a 3.6GHz P4 in typical tasks. GHz for GHz, the Pentium M seems to be similar or slightly faster than a G5. The area where today's Pentium M falls a little short and the G5 excels is floating point and SIMD, though, so for a few, this could have been a nice chip.



    But basically, this chip only confirms Jobs' analysis about how much a G5 has to be hobbled to be put in a laptop. Comparing the new G5, which isn't even available for purchase yet, with Pentium M's that are in widespread use, and have it still come up a little short isn't inspiring.




    By the time Apple moves to the Intel line there will be, I expect, an insurmountable lead from Intel's chips. for now though, video editing on a Powerbook with a proposed G5 1.6 would be pretty comparable to an M running at 2.1. I know that for integer tasks they aren't equal. But the truth is that for most programs right now that need speed, Altivec still contributes enough oomph, even at the lower speed so that for the next 9 months to a year, when Apple converts, it might pay. That's all I'm saying. It certainly isn't an ideal chip, and if Apple were staying with IBM, I'd be rather disappointed with it. As it is, I'm not. It's about what I expected. If IBM had come out with a 2GHz chip with 1MB L2 cache and integrated controller, I would be wondering what Jobs was smoking. As it is, I think that it has more power than the new 7448. Of course we will have to wait until all of the specs of this and the 7448 specs are released to make a real, rather than guessed at comparison.



    I do think that the Freescale 600 series looks interesting, but it has several incompatibilities with Apple's designs. Whether that would matter either is anyone's guess, but none of us actually know at this point anyway. We're just taking what we do know and extrapolating from there.



    One problem we have in comparing G4's with G5's is that the programming models diverge in several areas. Where the G4 gains power, the G5 loses it. The programming must be changed. If a test suite was optimised for the G4 then it might not work well for the G5.



    I've been beta testing Photoshop since ver 1. It's a mess of contradictions in its programming models. Some filters were optimised for the G3, some for the G4, and some for the G5. Some were updated when newer chips came out, and some weren't. That's why some tests show a big improvement, some just a small one, and the occasional few even slow down. I'm sceptical about testing therefore, as Adobe won't always tell us what they've done.



    I feel the same about the test suites out there. mostly they are trash. It's tough to get a handle on this.



    It's the same when comparing a Mac to a PC. Apple uses GCC for both. The complaints pour in. "They didn't use the Intel compiler". Great! Just how much software was written using the Intel compiler? Very little. We might as well write a compiler in Visual Basic.



    With all the tests people cite, I haven't seen, over the years such a great disparity. Not in practice. A difference, yes.



    The problem isn't over the next year, it's after that. Even the 7448 will make for a noticeably faster machine.
  • Reply 71 of 90
    melgrossmelgross Posts: 33,510member
    Quote:

    Originally posted by aegisdesign

    I'll just take these two points since we've been over the others over and over. The Barefeats tests are the only ones out there.



    There's more to an architecture than the R&D and production. There's support, system chips (the desktop G5 system chips suck), spare parts, warranty backup, casing... It's a whole new laptop. It's major expense for Apple to come up with a new design.



    Speed wise, the G5 wins over the G4 because of it's higher clock speed and high speed FSB. On a laptop you don't have the room for either of those and apart from that the G5 support chips are frankly terrible - hot and slow. When an eMac beats a PowerMac at drive tests, you know there's something seriously wrong. That's why I think it will even up when you have a 1MB L2 on the G4 v 512K on the G5.




    The thing is; you don't know. I don't know if IBM has improved the support chips. Do you? These evolve over time just as the cpu's themselves do.



    Spare parts mean nothing. Neither does warranty, casing, etc. You are telling us that Apple wouldn't have spare parts? That they couldn't warrant it? That they haven't been working on a new case since the day they released the current ones?



    None of that matters because if they are like any other manufacturer, they are continously working on their new generation, as well as the one after that, and are thinking about the one after as well. I know I did.



    No machine is by itself. It's always an intermediate model in a continual evolutionary process. Apple knew about this chip for quite a while. They thought about it and went through a preliminary design stage. After analysis they may have dropped it as being unwieldy. Or not.



    IBM's chips, at least these new ones, have power down features. That may help. It may not.



    I'm simply saying that Apple has to decide whether the glass is half empty, or half full. If either of us knew what they know then we could say more, but we are just speculating with very little knowledge.
  • Reply 72 of 90
    sybariticsybaritic Posts: 340member
    OnceUGoMacYoullNeverGoBack's initial post is a great one. Way to jump in with a splash. I think you're right about the ego factor, and your reference to GM and company is apropos. The difference, though, between IBM and many of the companies that you mention is that Big Blue has a serious game plan in the works (pun intended, but not just in gaming) whereas the other companies often rested securely on their laurels. It can happen to IBM, too, as you say.
  • Reply 73 of 90
    kreshkresh Posts: 379member
    Quote:

    Originally posted by Sybaritic

    OnceUGoMacYoullNeverGoBack's initial post is a great one. Way to jump in with a splash. I think you're right about the ego factor, and your reference to GM and company is apropos. The difference, though, between IBM and many of the companies that you mention is that Big Blue has a serious game plan in the works (pun intended, but not just in gaming) whereas the other companies often rested securely on their laurels. It can happen to IBM, too, as you say.



    IBM does not rest on their laurels, that is very true. Their problem is that they are like my 3 year old. They always start out with good intentions but somewhere along the way a shiny object attracts their attention and they sit down to play with it and forget about what they were doing.



    Let's see.



    IBM PC and IBM PC XT - They took the world by storm. They owned so much of the market that the catch phrase of the day was "IBM Compatible". They lost focus and let clone makers start dictating the market. They totally failed to compete in a hardware market that they created.



    OS/2 - They actually owned rights to Windows code. OS/2 was truly a work of art. Won't get into the whole sad story of OS/2, but in typical IBM fashion, it just seemed to slip away.



    PPC - We all know what a great architecure PPC is. We all know the story here. For what ever reasons, IBM again lost interest. Hey a new shiny object, a game machine.



    Cell - No doubts the new game boxes will be astounding, IBM's chip will shine. Things will look extremely bright, right up to the point IBM sees a new shining object leaves the whole gaming community left scratching it's head wondering what IBM was thinking of.



    Good bye IBM. Thank goodness Apple is leaving them behind. With IBM's track record in the digital age I sleep better at night knowing that Apple is moving to Intel.
  • Reply 74 of 90
    All I know is that this back-and-forth makes my head hurt. My family (and my wife's business) desperately needs a new computer, and we really want a Mac. Specifically, we want a Powerbook, but I don't want to put money into technology that will be swept aside in the next two years. I think the best compromise might be an iBook. It'll fit well within our budget and get us on the happy road of OS X.
  • Reply 75 of 90
    kenaustuskenaustus Posts: 924member
    I think that there is going to be a difference between when Apple is capable of shipping their first Mactel and when they actually ship. The difference is going to be the developers that are getting ready for the new Macs. Because Apple has these developers well identified with their leased development Mactel computers it is rather easy to keep track of how they are progressing. I don't see the first Mactel before the big guns (Adobe, MS/MBU, etc.) ready to deliver. If there are a lot of developers (including the big boys) ready in January then Steve J will let the Mactels hit the market, otherwise he'll wait.



    For now there is a bit of a breather called Back-To-School selling. Buy a Mac and get an iPod mini - which means nothing new until this promotion is over. At that time I believe we will see some updates. In the notebook area I wold bet it will be the best choice between the new Freescale and the IBM options. Since Freescale is close to delivering a dual core G4 I believe we might see a bump in the G4 iBook, followed by a dual core G4 in the PB. Because this will probably be the last G4 notebook in both ranges I think Apple will go for the fastest choice if other factors (cost. heat, etc.) are equal.



    As for notebook sales (excluding education sales, which will probably be very hot) I believe that most people will buy when they need the notebook. I had a 667 15" PB and was "waiting for the G5". When the 1.5 PB came out I went with it simply because I needed the performance improvement for business and decided it was not worth the wait.



    There are a lot of Mac users with older computers that will be in the same position. They'll decide to go for the improved performance now because they need it and the Mactels will be in rev b well before they need another computer. Others will get the G5 PMs and iMacs before they're gone because of specific software they are running. Remember that the G4 PM was available long after the G5 PM was announced. Lots of people, especially small studios using ProTools, needed the G4s to avoid upgrading their ProTools investment.
  • Reply 76 of 90
    melgrossmelgross Posts: 33,510member
    Quote:

    Originally posted by Squirrel_Monkey

    All I know is that this back-and-forth makes my head hurt. My family (and my wife's business) desperately needs a new computer, and we really want a Mac. Specifically, we want a Powerbook, but I don't want to put money into technology that will be swept aside in the next two years. I think the best compromise might be an iBook. It'll fit well within our budget and get us on the happy road of OS X.



    Take an aspirin, because this won't go away for at least a year. But...



    If Apple has a Powerbook that meets your needs then buy it. It doesn't matter what anyone here says about how fast or slow it might be. We have been using Powerbooks for high end tasks for years. For a business that isn't doing hi end 3D rendering, or pro level video editing, it just doesn't matter.



    If you are planning on buying a new Powerbook again in a couple of years or so, then the new Intel models will be a great replacement.



    Your machine won't be any more obsolete with x86 machines coming out a year or so then it would be with new PPC models coming out during that time period. But the Intel machines might be a pleasant surprise for you when you do upgrade again.



    If you use business software, then it should run quite well on any Intel replacement that you might get down the road, using Rosetta. If you plan on upgrading your software at some point a year or so away, it should be a Universal Binary, permitting it to work optimally on either machine.



    Don't worry, be happy.
  • Reply 77 of 90
    aegisdesignaegisdesign Posts: 2,914member
    Quote:

    Originally posted by melgross

    The thing is; you don't know. I don't know if IBM has improved the support chips. Do you? These evolve over time just as the cpu's themselves do.





    Those are Apple's design. They just aren't on a par with even the old Moto MPC106.



    Quote:

    Originally posted by melgross



    Spare parts mean nothing. Neither does warranty, casing, etc. You are telling us that Apple wouldn't have spare parts? That they couldn't warrant it? That they haven't been working on a new case since the day they released the current ones?



    None of that matters because if they are like any other manufacturer, they are continously working on their new generation, as well as the one after that, and are thinking about the one after as well. I know I did.





    Sorry, I've not made it clear. You're still thinking it's all about R&D. It's not. Tooling up production for a laptop is expensive. Building up inventory is expensive. Mould costs, machining, training support engineers, software development, yada, yada, yada...



    If your new laptop design only has max 18 months to run before you do it all again, but you've a get-out-of-jail-free card in the 7448, I'd bet they'd take the 7448 and make do. We all know how buying Rev A is sometimes not a good idea, well, there can't be anything safer than a Rev D or is it E now G4 laptop.
  • Reply 78 of 90
    aegisdesignaegisdesign Posts: 2,914member
    Quote:

    Originally posted by melgross



    If you use business software, then it should run quite well on any Intel replacement that you might get down the road, using Rosetta. If you plan on upgrading your software at some point a year or so away, it should be a Universal Binary, permitting it to work optimally on either machine.









    Rosetta is going to be interesting.



    On one hand, it's got to be able to run your existing PPC software as fast as a PPC machine would.



    On the other, if the disparity between native Intel and Rosetta emulated code is large, running PPC code on an Intel Mac is going to stand out like a sore thumb. I get annoyed at carbon apps not running as quickly as cocoa ones, and that's on the same CPU!



    I wonder how many applications don't make the switch to cocoa and universal binaries. I've a couple of carbon apps which barely squeaked through the OS9 to OSX change that have no alternatives. eg. MYOB. I wonder if they'll just give up on the Mac and stick to Windows rather than invest in moving to cocoa and intel.
  • Reply 79 of 90
    melgrossmelgross Posts: 33,510member
    Quote:

    Originally posted by aegisdesign

    Those are Apple's design. They just aren't on a par with even the old Moto MPC106.







    Sorry, I've not made it clear. You're still thinking it's all about R&D. It's not. Tooling up production for a laptop is expensive. Building up inventory is expensive. Mould costs, machining, training support engineers, software development, yada, yada, yada...



    If your new laptop design only has max 18 months to run before you do it all again, but you've a get-out-of-jail-free card in the 7448, I'd bet they'd take the 7448 and make do. We all know how buying Rev A is sometimes not a good idea, well, there can't be anything safer than a Rev D or is it E now G4 laptop.




    Well, here we go again. You still don't know.



    I know all about production. My company produced electronic products and speakers for quite some time (until we sold it). Some of those were products that I designed. I know the process. I never said that it wouldn't cost. I'm saying that it can be done, and that it might be further along than you think. I told you that's it's a continous process. It's not manufacturus inturruptus. It's an evolution from one design and chip in the same family to another. A bigger shift from G4 to G5 than from 7447 to 7448, true. But Apple might find reasons to do it. As I've said, part of it might be publicity. People waiting for a G5 might not buy a 7448 even if the performance might be as good, simply because they KNOW that the G5 is better.



    It has to do with sales. If Apple can do this, and I haven't said that they could, just that it might be possible, then it might be worthwile to them. Even if it's only for 9 months. There is a perception that Apple will lose a fair amount of business over the next year or two. They have to assure people that they are working on the best, most advanced solutions over that time. Even if Apple didn't make a penny of profit on these machines, if they stem a slide in sales, it could be worth it.



    I happen ti think that the chip COULD be better than the 7448, but until we see the full specs for both we won't know. The same is true of the support chips. We don't know. At least, you should realise that.
  • Reply 80 of 90
    melgrossmelgross Posts: 33,510member
    Quote:

    Originally posted by aegisdesign

    Rosetta is going to be interesting.



    On one hand, it's got to be able to run your existing PPC software as fast as a PPC machine would.



    On the other, if the disparity between native Intel and Rosetta emulated code is large, running PPC code on an Intel Mac is going to stand out like a sore thumb. I get annoyed at carbon apps not running as quickly as cocoa ones, and that's on the same CPU!



    I wonder how many applications don't make the switch to cocoa and universal binaries. I've a couple of carbon apps which barely squeaked through the OS9 to OSX change that have no alternatives. eg. MYOB. I wonder if they'll just give up on the Mac and stick to Windows rather than invest in moving to cocoa and intel.




    Considering that the word is that Apple won't be using the P M chip, but rather something newer, the performance might be good enough to be equal or better right off the bat.



    Of course, that's another rumor, though widespread.



    I think that most programs will make it if they are selling well enough now. But most don't access Altivec, and most aren't optimized past the G3. For them it shouldn't matter. Many companies didn't do much to their code because they didn't have to. MYOB is a good program. I don't use finantial programs so I'm not familiar with sales in that sector. But I would imagine that if they want to continue it they will move it over at some point. MYOB should run under Rosetta. So they have time.
Sign In or Register to comment.