Macbidouille: 970's on sale at the end of May

1910111214

Comments

  • Reply 261 of 300
    Quote:

    Originally posted by ConcoursAL

    I doubt Mercury has a 970 product ready. From the ImpactRT 3100 product page:



    "Harnessing PowerPC 7445 microprocessors to reach 600 GFLOPS or 2.4 TeraOPS, the ImpactRT 3100 multicomputer has the muscle to plow through the toughest computing tasks."



    Companies like Mercury are rarely at the forefront of processor technology. It took them a year or so after the introduction of the G4 to get a product to market. I don't expect things to change with the 970.



    While I don't know about hardware, Mercury does have a relationship with Apple in the software arena. They wrote vDSP. Mercury knows how to squeeze a proc for every last drop of performence when it comes to vector math.




    So true! But look at the picture for the ImpactRT 3100 http://www.mc.com/press_room/image_l...tegory=Systems and you'll see a motherboard of sorts leaning against the machine. That MB looks mighty familiar: wasn't something like that shopped around as a futute Apple PM MB or the IBM 970 blade MB a few month back? Anybody who has these old spy pictures might want to take a look?
  • Reply 262 of 300
    lemon bon bonlemon bon bon Posts: 2,383member
    I don't buy Programmer's logic on the above IBM insider rumour.



    Apple have updated machines when it has suited them. The fact that they turned over almost their entire line-up earlier this year shows what Apple can do.



    No reason that a similar sweeping move couldn't be on the cards following a May-July release of the 970 in the towers.



    And it's not as if Apple haven't got an urgent motivating factor. Their market share is taking a hammering. At 2%...and sliding slowly towards the nexus of irrelevant oblivion I can only hope that the calm, steely reserve of the share holder meeting suggests that the planets have almost moved into alignment re: the cpu situation. ie some master plan is in the offing.



    Apple is heading in the enterprise direction. It is THE online music store. It has THE music player marketshare.



    It has Unix on Mac on Unix.



    The impossible has already been done. None of this would surprise. It's hardly out of the ordinary. If Apple wanted to kick Motorola's butt, it would make sense that they would have a plan in place from the towers to laptops to afford that kind of assault on their primary chip supplier.



    Apple have shown with the music store, after years of putting all the pieces in place, that they can execute formidable plans.



    Given the current management team at Apple and a very sobre Steve Jobs, something about the whole 970 and IBM link means something more than 'just' the 970. I smell a plan born of a long time calculated.



    I think 2003/early 2004 is going to me more than 'the year of the laptop'.



    Lemon Bon Bon
  • Reply 263 of 300
    sc_marktsc_markt Posts: 1,402member
    Quote:

    Originally posted by barbarella

    So true! But look at the picture for the ImpactRT 3100 http://www.mc.com/press_room/image_l...tegory=Systems and you'll see a motherboard of sorts leaning against the machine. That MB looks mighty familiar: wasn't something like that shopped around as a futute Apple PM MB or the IBM 970 blade MB a few month back? Anybody who has these old spy pictures might want to take a look?



    From a page from Mercury's web site: Link



    From the top of the page:



    "March 31, 2003 - DALLAS, TX - March 31, 2003 - Mercury Computer Systems, Inc. (NASDAQ:MRCY), announced today at the Global Signal Processing Expo (GSPx) its ImpactRTTM 3100, the first scalable, high-performance, signal and image processing system based on the open-standard RapidIO® interconnect architecture. Combining next-generation PowerPC® microprocessors with AltiVecTM technology and the RapidIO communications switch fabric,"



    So they are using a chip with RapidIO and Altivec. Where is it coming from?



    From the bottom of the page, it says...



    "About RapidIO

    RapidIO is an open-standard, high-performance, packet-switched interconnect architecture that addresses the high-performance, embedded computing industry?s need for reliability, increased bandwidth, and faster bus speeds in an intra-system interconnect. The RapidIO interconnect allows chip-to-chip and board-to-board communications at performance levels scaling to tens of Gigabits/s and beyond. The initial RapidIO concept was the result of collaboration between Mercury Computer Systems and Motorola Semiconductor Products Sector (SPS). "



    In short, this sounds like (at least to me) that Mercury is using a powerpc chip from Motorola that has both Altivec and RapidIO.



    Interesting...
  • Reply 264 of 300
    sc_marktsc_markt Posts: 1,402member
    Quote:

    Originally posted by ConcoursAL

    I doubt Mercury has a 970 product ready. From the ImpactRT 3100 product page:



    "Harnessing PowerPC 7445 microprocessors to reach 600 GFLOPS or 2.4 TeraOPS, the ImpactRT 3100 multicomputer has the muscle to plow through the toughest computing tasks."







    Interesting that on one page, it kind of sounds like the 3100 has a chip with both RapidIO and Altivec. However, when you click to the page that has the text you posted above, it says they are using the 7445

    chip. However, there is no mention of RapidIO on the 7445 webpage.



    They must be using a lot of G4's (assuming they are using them) to get 600GFLOPS.
  • Reply 265 of 300
    kurtkurt Posts: 225member
    I was starting to think that Apple may update the entire line as soon as possible to the 970. With the wide speed range rumored, that may be possible. It would remove their dependence on Motorola and give them fast machines across the range. My big concern though is even if the chips are reasonably priced, won't the support chips and memory to support the 970 be expensive? I would think it would be difficult to have iBooks with a slow 970 and keep the price low. The other choice would be to cripple the machines with low speed ram but I am not sure that makes sense. The IBM rumor above makes sense and gives Apple a couple of different chips with different buses to support: high end/expensive interfaces and low end/cheap interfaces.



    Any one else have any thoughts?
  • Reply 266 of 300
    kupan787kupan787 Posts: 586member
    Quote:

    Originally posted by Kurt

    I was starting to think that Apple may update the entire line as soon as possible to the 970. With the wide speed range rumored, that may be possible. It would remove their dependence on Motorola and give them fast machines across the range. My big concern though is even if the chips are reasonably priced, won't the support chips and memory to support the 970 be expensive? I would think it would be difficult to have iBooks with a slow 970 and keep the price low. The other choice would be to cripple the machines with low speed ram but I am not sure that makes sense. The IBM rumor above makes sense and gives Apple a couple of different chips with different buses to support: high end/expensive interfaces and low end/cheap interfaces.



    Any one else have any thoughts?




    I made a few posts on pages 5, along with others. Check out that page for more on this thought, but here is basicly what I said:



    The 970 will go in the towers, and powerbooks (low power 970's) in there next revisions.

    The iBooks will get GOBI or 850s.

    The iMacs will either get faster G4s (they have 1.2, 1.3, and 1.4 GHz parts they could use right now), and then move to the 970. Or the iMacs might just get a low power 970 in their next update.

    The eMac was just updated, so the next update will probably be a low power 970, or a faster G4 part (depending on what teh imac goes to).
  • Reply 267 of 300
    programmerprogrammer Posts: 3,461member
    Quote:

    Originally posted by Lemon Bon Bon

    I don't buy Programmer's logic on the above IBM insider rumour.



    Apple have updated machines when it has suited them. The fact that they turned over almost their entire line-up earlier this year shows what Apple can do.





    Yeah, but they just updated the eMacs. The rest of the lineup will be 6-7 months old at WWDC and an update seems reasonable, but why would they have updated the eMac just to do it again 1-2 months later? Of course there is always the possibility that they'll just use up their stock of G4 chips in the eMac and update it to an IBM G3 w/ SIMD in 6 months...
  • Reply 268 of 300
    keyboardf12keyboardf12 Posts: 1,379member
    Quote:

    chips in the eMac and update it to an IBM G3 w/ SIMD in 6 months...



    That's where i would place my bet.
  • Reply 269 of 300
    nevynnevyn Posts: 360member
    If I was pissed off enough at Motorola to _really_ sue them, and I knew I would have replacement chips at most 8mo in the future, I might just buy a 8mo supply _now_.



    There isn't going to be much in the 'speed bump' area. Sales of iMacs isn't going to double without major changes -> you sort of know just how many chips you're going to need.
  • Reply 270 of 300
    shaktaishaktai Posts: 157member
    Quote:

    Originally posted by Programmer

    Yeah, but they just updated the eMacs. The rest of the lineup will be 6-7 months old at WWDC and an update seems reasonable, but why would they have updated the eMac just to do it again 1-2 months later? Of course there is always the possibility that they'll just use up their stock of G4 chips in the eMac and update it to an IBM G3 w/ SIMD in 6 months...



    I could agree with that.



    Purely speculative, but I am guessing this is what we will see based upon all the rumors I could put together and discarding those that just didn't seem feasible to me. Mind you I reserve the right to change my mind in the next thread.



    WWDC - an acknowledgement of the 970 and 64 bit Panther without detailed hardware specs. Running demos, but no "open" details, but probably some closed door briefings. Focus will be on Panther and 64 bit software (maybe a surprise announcement) with just routine passing references to other Apple successes and hardware.



    Shortly After WWDC, a dedicated Apple Event, just for the hardware announcements with details. PowerMac and X-Serve announced with 970, probably in 1.4 single to 1.8 dual ranges. Available immediatly or within a week. Still don't see a 970 Powerbook, just because I don't think the mobo's to support them are there yet.



    Speedbumps for PowerBooks, iMacs and maybe iBooks.



    G3's continue in iBooks for sometime, no G4's, but the G3's will be dramatically improved (Gobi, etc., as per rumors) over the current models. Maybe eMacs too, but who knows.



    Early 2004: PowerBooks and iMacs to 970. iBooks and eMacs to advanced G3's. Power Macs get lots faster.



    OR - I am am proven completely wrong again.
  • Reply 271 of 300
    kurtkurt Posts: 225member
    Quote:

    Originally posted by kupan787

    I made a few posts on pages 5, along with others. Check out that page for more on this thought, but here is basicly what I said:



    .




    I knew it was too good to be my own thought. Must have subconciously copied it.



    Thanks.
  • Reply 272 of 300
    macjedaimacjedai Posts: 263member
    I don't buy the L3 cache ideas from the inside info . . . if you have a fast enough FSB, why have an L3 . . . The RAM is the L3.



    As for the the G3 variants, I believe it. They exist and I'm pretty sure Apple has "samples".
  • Reply 273 of 300
    bungebunge Posts: 7,329member
    Quote:

    Originally posted by MacJedai

    I don't buy the L3 cache ideas from the inside info . . . if you have a fast enough FSB, why have an L3 . . . The RAM is the L3.



    RAM isn't that fast yet. A cache is still really valuable considering how many things are normally running.
  • Reply 274 of 300
    telomartelomar Posts: 1,804member
    The advantage of L3 cache isn't so much the speed as the latency improvements. It costs far too much to be worthwhile on a desktop system with a decent FSB though.
  • Reply 275 of 300
    guarthoguartho Posts: 1,208member
    MacMall has a special promotion on PowerBooks and PowerMacs that ends May 25th, a Sunday. Hmmmmmmm.....



    \
  • Reply 276 of 300
    lemon bon bonlemon bon bon Posts: 2,383member
    I kinda agree with Shaktai. No later than 2004. It may happen sooner though. But no later than what he said. Moto's on their way out. Give or take the odd 8 months or so.



    I'll be glad to see the back of this incompetant FAB cripple. And as for those worried about the 2nd cpu supplier...I've got a feeling it may come from someone other than Motorola.



    Lemon Bon Bon
  • Reply 277 of 300
    stoostoo Posts: 1,490member
    Quote:

    I've got a feeling it may come from someone other than Motorola.



    Not that topic again.



    Seriously, are you thinking about someone else designing or fabricating PowerPC CPUs? And who?
  • Reply 278 of 300
    programmerprogrammer Posts: 3,461member
    Quote:

    Originally posted by Stoo

    Not that topic again.



    Seriously, are you thinking about someone else designing or fabricating PowerPC CPUs? And who?




    A second source could simply be an alternative fab which uses IBM's technology.
  • Reply 279 of 300
    programmerprogrammer Posts: 3,461member
    Quote:

    Originally posted by MacJedai

    I don't buy the L3 cache ideas from the inside info . . . if you have a fast enough FSB, why have an L3 . . . The RAM is the L3.





    IBM's embedded DRAM technology could provide a cost effective way of shielding the rest of the system from the latency and read/write multiplexing issues of the memory subsystem. It would also improve how multiple clients of the memory system deal with contention issues.
  • Reply 280 of 300
    macjedaimacjedai Posts: 263member
    Quote:

    Originally posted by Programmer

    IBM's embedded DRAM technology could provide a cost effective way of shielding the rest of the system from the latency and read/write multiplexing issues of the memory subsystem. It would also improve how multiple clients of the memory system deal with contention issues.



    Now the "multiple clients of the memory system", I buy. I wish I knew if that was part of "Apple PI" or not. Unfortunately, my sources don't give details, but instead refer to secondary sources that talk about "PI". Sure wish that they would talk to me.



    Thanks for making me think about it more!
Sign In or Register to comment.