new G4 is a 7455. will there ever be Apollo?

krassykrassy
Posted:
in Future Apple Hardware edited January 2014
as motorolas site <a href="http://e-www.motorola.com/webapp/sps/site/taxonomy.jsp?nodeId=03M943030450467M98653"; target="_blank">http://e-www.motorola.com/webapp/sps/site/taxonomy.jsp?nodeId=03M943030450467M98653</a>; is saying, the new G4 in apple's tower-line is a 7455 G4.



my question: is this apollo or will there be the apollo between the 7455 and the G5? i think an apollo could come with up to 1.6 Ghz, but the G5 is rumored to start at 1.2 Ghz... that leads me to believe that there may never be an apollo 7460 in the power mac line... the next big thing will come at MWNY and will be the G5....



apollo - well, it could be for iBook and PowerBook as well as for the iMac...



what do you think?
«13

Comments

  • Reply 1 of 43
    heinzelheinzel Posts: 120member
    Looks like they also have enhanced the last version's layout, manufacture with SOI and added some other twists and got an added 25% extra MIPS out of it (2280 as opposed to the theoretical 1806 when just scaling up from 733 MHz to 1 GHz). I don't know what the 7460 is supposed to be like re-engineering-wise but it seems like this puppy DOES have some life left in it. I wonder if they wouldn't just add larger L2 cache and support for more L3 cache (say, 8MB?) to the Apollo's redesigned core to cope with the starved memory bus and of we'd go? Plus, it sounds like the could just go multiprocessor now with the 7550; will we see Apple multiprocessor rack server farms replace the G4 servers soon?



    PS: Could anybody please explain the purpose and benefits of larger block address translation registers?
  • Reply 2 of 43
    marcukmarcuk Posts: 4,442member
    Is that Apollo 7455 (im confused). If it is a 7540 with SOI then it technically is what we assume to be Apollo, if not then what is or isn't it?
  • Reply 3 of 43
    *l++*l++ Posts: 129member
    This computes to the new 7455 1GHz G4 corresponding to 1.262GHz previous generation(7450/1) G4 and a 1.834GHz original (7400) G4.



    I guess the G4 has come further than we all tought.
  • Reply 4 of 43
    *l++*l++ Posts: 129member
    [quote]Originally posted by heinzel:

    <strong>Looks like they also have enhanced the last version's layout, manufacture with SOI and added some other twists and got an added 25% extra MIPS out of it (2280 as opposed to the theoretical 1806 when just scaling up from 733 MHz to 1 GHz). I don't know what the 7460 is supposed to be like re-engineering-wise but it seems like this puppy DOES have some life left in it. I wonder if they wouldn't just add larger L2 cache and support for more L3 cache (say, 8MB?) to the Apollo's redesigned core to cope with the starved memory bus and of we'd go? Plus, it sounds like the could just go multiprocessor now with the 7550; will we see Apple multiprocessor rack server farms replace the G4 servers soon?



    PS: Could anybody please explain the purpose and benefits of larger block address translation registers?</strong><hr></blockquote>



    The block address translation registers helps map out memory to cache.
  • Reply 5 of 43
    marcukmarcuk Posts: 4,442member
    I have a feeling that a 7455 is Apollo on a 0.18 SOI process. I think Apollo 7460 will be on 0.13 SOI, but basically they're the same chip. Yes we have Apollo!
  • Reply 6 of 43
    heinzelheinzel Posts: 120member
    [quote]Originally posted by *l++:

    <strong>



    The block address translation registers helps map out memory to cache.</strong><hr></blockquote>



    Thanks. I didn't look at it the 7400 way, 1.834 GHz doesn't sound bad at all, does it?

    Seems like instead of looking into new faster memory/busses they are just improving the cache management and circumvent a redesign of the old boards while getting better performance. Nice idea, keeps investments for updating the old platform low and still makes it faster.

    Now, will the 7460 be a redesign or was it just supposed to be 7450 SOI'd? I thought there was going to more to it than just that.



    [ 01-28-2002: Message edited by: heinzel ]</p>
  • Reply 7 of 43
    *l++*l++ Posts: 129member
    The other interesting thing to note is that their is not dearth of 1GHz chips. Apple would never have offered the fastest chip in a dual chip configuration otherwise.



    This means that they must be getting quite a number of chips that clock faster than that.



    It looks to me that Apple is accumulating faster chips for a later introduction.
  • Reply 8 of 43
    krassykrassy Posts: 595member
    [quote]Originally posted by heinzel:

    <strong>



    Thanks. I didn't look at it the 7400 way, 1.834 GHz doesn't sound bad at all, does it?

    Seems like instead of looking into new faster memory/busses they are just improving the cache management and circumvent a redesign of the old boards while getting better performance. Nice idea, keeps investments for updating the old platform low and still makes it faster.

    Now, will the 7460 be a redesign or was it just supposed to be 7450 SOI'd? I thought there was going to more to it than just that.



    [ 01-28-2002: Message edited by: heinzel ]</strong><hr></blockquote>



    yes. 500Mhz DDR L3 RAM is quite fast. i wonder how these new machines compete to my 400Mhz/100Mhz bus AGP G4...



    sad that the 800Mhz G4 has got no L3 cache.

    but this is for getting most buyers as possible to the iMac i think.



    anyway - the new improvements in performance would help apple to give new stuff into their mhz myth story...



    could that be? a 26% advantage over a P4 for the G4/800 ????? and how is a G4/800 now faster than the DP 800 (quake) seems strange ...
  • Reply 8 of 43
    agent302agent302 Posts: 974member
    Yeah, my bet is that the 7460 is on .13 micron, seeing as how they'll need to shrink the die to reduce power consumption. 21.3 watts? Don't expect seeing a 1 GHZ 7455 in a powerbook. Hopefully shrinking to .13 will reduce heat a bunch.
  • Reply 10 of 43
    moogsmoogs Posts: 4,296member
    [quote]This computes to the new 7455 1GHz G4 [being equivlanet to a] 1.262GHz...7450/1 [and a 7400 at] 1.834GHz.<hr></blockquote>



    What are you suggesting by this? Is it even correct (the way I've interpretted it above)? Are you basically saying the new GHz chip is not twice as fast as my 500 MHz 7400/7410 CPU, but rather closer to 3.5 times as fast? Perhaps even better than 6x as fast when coupled with another GHz processor, in terms of how quickly OS X operates (being a system optimized for dp)?



    How about the 2MB L3 cache - what kind of effect does this have on OS responsiveness or something like app launch times?



    [ 01-28-2002: Message edited by: Moogs ? ]</p>
  • Reply 11 of 43
    krassykrassy Posts: 595member
    [quote]Originally posted by agent302:

    <strong>Yeah, my bet is that the 7460 is on .13 micron, seeing as how they'll need to shrink the die to reduce power consumption. 21.3 watts? Don't expect seeing a 1 GHZ 7455 in a powerbook. Hopefully shrinking to .13 will reduce heat a bunch.</strong><hr></blockquote>



    what about the 10 Watts for the low power version of the 7455 @ 733 or 800 Mhz ?



    is this too much?
  • Reply 12 of 43
    krassykrassy Posts: 595member
    [quote]Originally posted by Moogs ?:

    <strong>



    What are you suggesting by this? Is it even correct (the way I've interpretted it above)? Are you basically saying the new GHz chip is not twice as fast as my 500 MHz 7400/7410 CPU, but rather closer to 3.5 times as fast? Perhaps even better than 6x as fast when coupled with another GHz processor, in terms of how quickly OS X operates (being a system optimized for dp)?



    How about the 2MB L3 cache - what kind of effect does this have on OS responsiveness or something like app launch times?



    [ 01-28-2002: Message edited by: Moogs ? ]</strong><hr></blockquote>



    i think that these numbers for MIPS (Million Instructions Per Second) could not be taken as an indicator of the real world performance of the processor. but, however - if the 7455 does the same things per instruction as the 7400 does that would mean nearly a 2.5 times speed increase. (917 @ 7400 G4 500Mhz vs. 2280 @ 7455 1000 Mhz G4). as usual you don't get the doubled performance if running a 2-processor-system. only if you run many different tasks... so 5 times the performance of a 7400 @ 500 Mhz for the new DP 1Ghz is just a theoretical number....



    greets,

    krassy



    [ 01-28-2002: Message edited by: Krassy ]</p>
  • Reply 13 of 43
    postheidepostheide Posts: 13member
    This is not the (original) Apollo. The technically specifications officially announced by Moto. (some times ago) were:



    - Name: 7460

    - L2 cache size 512 KB

    - low-power/heat-dissipation



    So this is a tweaked 7451 on a new process (SOI). Maybe they simply lowered their goals in order to deliver. Or G5 is faster progressing than expected while the Apollo slipped behind schedule? But then: What about new PowerBooks? 550 and 667 MHz is not competative.
  • Reply 14 of 43
    krassykrassy Posts: 595member
    [quote]Originally posted by postheide:

    <strong>This is not the (original) Apollo. The technically specifications officially announced by Moto. (some times ago) were:



    - Name: 7460

    - L2 cache size 512 KB

    - low-power/heat-dissipation



    So this is a tweaked 7451 on a new process (SOI). Maybe they simply lowered their goals in order to deliver. Or G5 is faster progressing than expected while the Apollo slipped behind schedule? But then: What about new PowerBooks? 550 and 667 MHz is not competative.</strong><hr></blockquote>



    again - look over at <a href="http://e-www.motorola.com/brdata/PDFDB/docs/MPC7455FACT.pdf"; target="_blank">http://e-www.motorola.com/brdata/PDFDB/docs/MPC7455FACT.pdf</a>;



    there is a low power version of the 7455 running at 10,3 Watts @ 733 Mhz and is available up to 800 Mhz



    greets,

    krassy
  • Reply 15 of 43
    brussellbrussell Posts: 9,812member
    Wow, good catch, Krassy. So this actually is the Apollo.



    Some possible implications/questions:



    1. If the 7451 tops out at 867, I wouldn't expect this 7455 to get much higher than 1Ghz. Maybe 1.1 Ghz or thereabouts, if we assume SOI can add about 25% to the clock speed.



    2. I also wonder if this is the chip in the iMac, or if that has the non-SOI chip. Anybody know? In any case, it should be easy to scale the iMac up in the near future.



    3. New PowerBooks should get this chip next.



    4. Why are they calling this the 7455? Is this more evidence that the G5 will just be 7455 + .13µ?
  • Reply 16 of 43
    krassykrassy Posts: 595member
    [quote]Originally posted by BRussell:

    <strong>Wow, good catch, Krassy. So this actually is the Apollo.

    </strong><hr></blockquote>



    thanks



    [quote]<strong>

    Some possible implications/questions:



    1. If the 7451 tops out at 867, I wouldn't expect this 7455 to get much higher than 1Ghz. Maybe 1.1 Ghz or thereabouts, if we assume SOI can add about 25% to the clock speed.

    </strong><hr></blockquote>



    seems right. maybe that's why they have enough of the 1 Ghz procs to go DP



    [quote]<strong>

    2. I also wonder if this is the chip in the iMac, or if that has the non-SOI chip. Anybody know? In any case, it should be easy to scale the iMac up in the near future.

    </strong><hr></blockquote>



    as far as i know the chip in the iMac is the 7441 or s.th. - if it were the 7455 the motorola specs for the 7455 must have been out for weeks. i could be wrong on that though -&gt; the iMacs starts shiping now and perhaps they didn't show their specs for a new proc to keep tight-lipped on new G4-processors (PMG4 rumors)



    EDIT: just thought it over. the iMac chip has to be the 7445 - the new low power version without L3 cache... 800 Mhz - so, scaling is possible - but only with 0.13µ



    [quote]<strong>

    3. New PowerBooks should get this chip next.

    4. Why are they calling this the 7455? Is this more evidence that the G5 will just be 7455 + .13µ?</strong><hr></blockquote>



    3. i think so too

    4. i think they call it 7455 because the 7460 should be the 7455 at 0.13µ

    the G5 will have a completely new architechture (up to 500Mhz system bus, 333Mhz DDR Ram controller, 64bit etc...)



    [ 01-28-2002: Message edited by: Krassy ]



    [ 01-28-2002: Message edited by: Krassy ]</p>
  • Reply 17 of 43
    moogsmoogs Posts: 4,296member
    [quote]Originally posted by Krassy:

    <strong>



    i think that these numbers for MIPS (Million Instructions Per Second) could not be taken as an indicator of the real world performance of the processor...however - if the 7455 does the same things per instruction as the 7400 does that would mean nearly a 2.5 times speed increase. (917 @ 7400 G4 500Mhz vs. 2280 @ 7455 1000 Mhz G4). as usual you don't get the doubled performance if running a 2-processor-system. only if you run many different tasks... so 5 times the performance of a 7400 @ 500 Mhz for the new DP 1Ghz is just a theoretical number....

    </strong><hr></blockquote>





    So how could we verify if the 7455 does, in fact, do as much per cycle as earlier models? Are there any benchmarks out there that would indicate something like this? I wonder if the chip is sufficiently different from previous models that Ars will do a report on it.
  • Reply 18 of 43
    krassykrassy Posts: 595member
    [quote]Originally posted by Moogs ?:

    <strong>





    So how could we verify if the 7455 does, in fact, do as much per cycle as earlier models? Are there any benchmarks out there that would indicate something like this? I wonder if the chip is sufficiently different from previous models that Ars will do a report on it.</strong><hr></blockquote>



    i think an instruction can last as long as 4 cycles or any other number - so it has nothing to do with the tasks per cycle.



    but it's pretty sure that the G4 is that faster than the older G4 cause the G4 PPC's all have the same instructions set which applications can use. if motorola states that the new G4 on same Mhz could do more of those Instructions per second, this leads us to the performance-gain mentioned above...



    though i'm no specialist in that things....
  • Reply 19 of 43
    jlljll Posts: 2,713member
    [quote]Originally posted by *l++:

    <strong>This computes to the new 7455 1GHz G4 corresponding to 1.262GHz previous generation(7450/1) G4 and a 1.834GHz original (7400) G4.</strong><hr></blockquote>



    No, no. The original G4 does 917 MIPS @ 500MHz which theoretically is 1834 MIPS @ 1GHZ thus making the MPC7455 24.3% faster.
  • Reply 20 of 43
    *l++*l++ Posts: 129member
    [quote]Originally posted by Moogs ?:

    <strong>



    What are you suggesting by this? Is it even correct (the way I've interpretted it above)? Are you basically saying the new GHz chip is not twice as fast as my 500 MHz 7400/7410 CPU, but rather closer to 3.5 times as fast? Perhaps even better than 6x as fast when coupled with another GHz processor, in terms of how quickly OS X operates (being a system optimized for dp)?



    How about the 2MB L3 cache - what kind of effect does this have on OS responsiveness or something like app launch times?



    [ 01-28-2002: Message edited by: Moogs ? ]</strong><hr></blockquote>



    This is not a system performance computation, simply a CPU performance computation from the Drystone MIPS as stated in Motorola's documentation.
Sign In or Register to comment.