un-Happy 3rd Birthday G4 (Sep 2002)

in Future Apple Hardware edited January 2014
While I'm as big an Apple (and X) fan as anyone I gotta tell you the G4 has been NOTHING but a POS processor. While many here already know that here are some cold hard facts. (thanks in part to Mactracker).


Sept 99 - Apple intro's the G4 at 400/450/500Mhz

Nov 99 - Apple cranks things back to 350/400/450

Jan/Feb 2k - Apple cranks back to 400/450/500Mhz

Jul 2k - Apple pulls out the DP boxes at 2x450 2x500

Jan 01 - 466/533/2x533/667/733

Jul 01 - 733/867/2x800

Jan 02 - 800/933/2x1Ghz


G4 stuck at 500Mhz for SIXTEEN MONTHS without a single speed bump and over the next NINTEEN MONTHS we were able to add 500Mhz to top out at 1Ghz (where we are today).

I'm sorry but I have to ask... Was MOT's AltiVec worth it? Would Steve (knowing what he does now) have still gone with MOT over IBM? Quotes in the press from MOT (I think) state that the G4 'still has legs' and after looking over the past (close to) THIRTY SIX MONTHS with a Mhz POP of only 500Mhz... I'm sorry but how could people say that with out busting out laughing? :confused:

Now to keep this post in future hardware... After looking at the cold hard facts as to what MOT's been able to do with the G4 up to today... Does anyone REALLY think we are gonna see 1.5Ghz machines in August? Don't get me wrong I'd really love to eat some crow on this but given the fact that it's taken MOT TWENTY NINE MONTHS to go from 500Mhz to 1Ghz I'd have to say the 'smart money' is betting against MOT giving Apple a 1.5Ghz G4.

Just my two cents but 3 years is a long time with any single CPU (in the case of the G4 it's WAY TOO LONG) and it's time to write the G4 off as a bad idea.

Depressed... Dave

[ 07-30-2002: Message edited by: DaveGee ]</p>


  • Reply 1 of 65
    vvmpvvmp Posts: 63member
    After all the high-end software packages Apple has invested in, signaling a move into the high-perfomance market, I think they would be commiting suicide if they didn't have an ace up their sleeve. They are well aware a mere speed bump won't cut it. I think we are in for a real surprise.
  • Reply 2 of 65
    pscatespscates Posts: 5,847member
    Man, I've been saying this for a while now...mostly in response to all the jokers who think we're going to wake up one day and suddenly have dual 1.6GHz G5s or whatever.

    I didn't have the cold, hard dates/specs (like your post), but I knew roughly that the G4 had only grown about 500-600MHz in three years. And EVERY Macworld that rolls around, some people here get torqued-up and COMPLETELY ignore (or forget) the previous 2-3 year's worth of update amounts/patterns and suddenly think there's going to be this HUGE leap.

    Not that I bank on MHz and all (could care less, to be completely honest), but the G4's - had everyone had their act together - should be, right now, WELL into the mid-GHz range.

    We should be talking about 2.0GHz G4s (or G5s?) making their debut in 5 months at MWSF 2003. Instead, we'll be lucky to get 1.4GHz.

    But who knows...maybe next month, these "new" G4s really make up some ground?

    I'm not holding my breath, however...
  • Reply 3 of 65
    maskermasker Posts: 451member
    I too think we are in for a suprise, but what I can't figure out if Apple has made some quantum leap in working around the processor hurdle, or if the processor is a accelerated....

    I'd love to hear Steve Jobs say..

    Introducing the new 100 Mhz G6 processor, low wattage allows for 10 hours of battery life in a Ti Book and does 154 Gigaflops... (dramatic pause) per processor.

    We 've got these babies in dual, quad and sextuplet groupings... yadda yadda yadda..

  • Reply 4 of 65
    davegeedavegee Posts: 2,765member
    Oh I totally agree with you guys... Mhz isn't **EVERYTHING** but Mhz-Myth aside the fact remains that MOTOROLA's advancement (gotta give 'credit' where credit is due) of the PPC G4 has been close to STAGNANT! (inactive sluggish inert - pick your own favorite)

    Mhz gap doesn't really bother me but an almost total LACK OF PROCESSOR ACTIVITY/ADVANCEMENT does.

    I too keep the faith and as always it's the OS that's doing it... I always said I loved Mac OS 9 and now that I'm on X the only issue I have is now I have to come up with an even better term to express my feelings about the OS.


    [ 07-30-2002: Message edited by: DaveGee ]</p>
  • Reply 5 of 65
    500 MHz in almost three years?

    Ouch, I knew it was bad but... ouch.

    And the amazing thing is -- as slow as this chip has come along -- Motorola still manages to deny it even its full capability by choking it with the 133MHz bus.

    It's a sad, sad time for the Power Macs. (Everything else is looking right spiffy, although it's too bad the Power Macs have the iMac "pinned" at 800 MHz...)

    [ 07-31-2002: Message edited by: iconmaster ]</p>
  • Reply 6 of 65
    MHz might not be **everything**, but unfortunately the "MHz Myth" has turned into the "GHz Gap"

    <img src="graemlins/oyvey.gif" border="0" alt="[No]" /> <img src="graemlins/oyvey.gif" border="0" alt="[No]" /> <img src="graemlins/oyvey.gif" border="0" alt="[No]" />
  • Reply 7 of 65
    groveratgroverat Posts: 10,872member
    The Mhz Myth... isn't.
  • Reply 8 of 65
    algolalgol Posts: 833member
    Yes the old G4 had huge problems scaling but since then motorola has changed it. We will start to see higher speeds than we have been. The August announcement will include some major enhancements not only to the bus speed ect. but also to the chip speed it self.

    Also with the advent of 10.2 we will notice even greater speed jumps and our current setups will be much snapier. I can vouch for this because I have been testing build 6c106. On older systems you will notice a huge difference and on newer ones you will notice a difference in games ect. For example Warcraft 3 now plays better than it does in 9.

    You will see speed jumps of up to 400Mhz (maybe higher it depends on motorola) and completely new architecture come august. Apple does not want to bump speeds to much because they need to have growth room toupgrade untill the 7500 comes out which they will call the G5 or similar.

    Someone posted earlier about the new X product naming at apple. Then some power hungry moderator locked it claiming it had no evidence ( or some other bs) The 7500 may very well be called the X1 or X5. Apple wants to depart from the G naming structure because of the problems people link to it. Also since the G3 is about done for and the G4 has been around for awhile.

    The 7500 is a G4 but it is also very different. It will climb to much higher speeds. The chip, however, puts out huge amounts of heat. Don't expect powerbooks anytime soon. I have been studying the case of that leaked photo ( I'm not sure if it is actually legit but since apple legal went crazy over it I am beginning to think it is) It seems to have huge venting and fans ect. This makes me think that the 7500 may be in the august powermacs.

    From what I have heard it is not ready though so I don't know about this. If the 7500 is indeed ready that would explain the new case design ect. If the new PM s have the 7500 I am willing to bet that it will have speeds up to 1.6Ghz and will be called the PowerMac X1 with duals on top end model.

    This post is based partly on inside info from a developer I know who has connections. But it is mainly based on existing knowledge and common sense. The X naming structure I have known about for awhile and can attest that apple has considered it. I do not know anything for certain for it is always changing. Steve can change his mind the last week if he wants. Everafter the X serve though I have been more certain of the coming X brand. Plus The PowerMac X1 sounds good. Better than the PowerMac G5 which won't really have a G5 but a G4 7500. Also the real G5 will not be callled a G5 I know this for certain as well.



    "utilizing drugs to pay for secret wars around the world"

    [ 07-31-2002: Message edited by: Algol ]</p>
  • Reply 9 of 65
    It seems that Algol, at least, seems to know what he's talking about. I'll take his words with a grain of salt (of course).

  • Reply 10 of 65
    algolalgol Posts: 833member
    yea I thought about starting a new thread for my post but then I figured they (the mods) would lock it. I can see it now: "You don't have enough evidence so therefore lock"!!!!!

    Ok ok I'll stop it about the X mac thread.

    One more thing about the new powerMacs I forgot to include in my original post. As far as I can tell from what I have heard they will ship with 10.2 Wether or not they come out before August 24. I am not sure if they will require 10.2 to run or not I would bet they don't need it but apple will not release new machines without 10.2.
  • Reply 11 of 65
    eugeneeugene Posts: 8,254member
    [quote]Originally posted by groverat:

    <strong>The Mhz Myth... isn't. </strong><hr></blockquote>

    Don't lie.
  • Reply 12 of 65
    bartobarto Posts: 2,246member
    The G4 may have been stuck at 500MHz for a while, but when it got going, it kept pace with the rest of the industry (DP 500MHz -&gt; DP 1GHz = Doubling in ~18 Months)

    And despite the G4's relativly high IPC (instructions per clock cycle), the DP 1GHz Power Mac is not as fast as DP Xeons/Athlon MPs, even in Photoshop (one of the Mac's "good" apps, and a platform cannot survive on a few good apps alone). A Power Mac update in August might change that in Photoshop and the other "good" apps, but just once I'd like to see a Power Mac to beat a PC in the majority of benchmarks thrown at it.

    However, as we have seen with AMD, just because a chip maker doesn't have the best chip right now doesn't mean it won't in the future.

    AMD developed the K7 (athlon) from scratch, building a deeper pipeline, loads of execution units, a 200MHz front side bus... Just because one company in the industry is lagging doesn't mean that innovation in that industry stagnates.

    The other companies' innovation serves as an example for the lagging company (Motorola), and the lagger's next product (CPU), combined with big investment in a new chip, and the fact (as Eskimo succinctly put it once) that chip designers all go to the same schools, the next Motorola CPU should be way better than their current ones.

    And lo and behold, Motorola has released an (apparently) praiseworthy chip for embedded computing (the 85xx range), it's just that their next desktop chip isn't here yet, or isn't planned (whether that is a good or bad thing is up to you).


    Motorola/IBM turned away from Apple with the G3/G4, only producing chips for the high-end embedded market (the 7xx/7xxx range), and not the desktop-computing market (the 6xx range). This was probably for solid financial reasons, but not good for Mac users. I am in doubt that ANY 7xx/7xxx PPCs will EVER compete with Intel/AMD, due to their embedded focus.

    But I believe that the PPC line holds promise, with the move to more modular cores boding well for possibly a return to the 6xx series by Motorola/IBM, as a desktop-computing chip would cost less to develop. The Power5 (and rumor has it a stripped down powermac suitable Power4) is on the horizon, and if Apple becomes doubtful about the PowerPC, then that's why they have US$4.3 billion for a rainy day.


    Edited for clarity, and to add that Power Macs are not always used to wipe the floor.

    [ 07-31-2002: Message edited by: Barto ]</p>
  • Reply 13 of 65
    algolalgol Posts: 833member
    ok good point but you don't make complete sense. The G4 series has always stacked more power per clock cycle than any other chip be it P4 or AMD. Apple will eventually leave motorola in favor of IBM that is for the future to tell. But remember apple is currently using both companies. The G3 from IBM and the G4 from Motorola. It would not be hard to switch to IBM for the Power4 or 5 or whatever. This I do not know. In time.
  • Reply 14 of 65
    bartobarto Posts: 2,246member
    [quote]Originally posted by Algol:

    <strong>ok good point but you don't make complete sense. The G4 series has always stacked more power per clock cycle than any other chip be it P4 or AMD. Apple will eventually leave motorola in favor of IBM that is for the future to tell. But remember apple is currently using both companies. The G3 from IBM and the G4 from Motorola. It would not be hard to switch to IBM for the Power4 or 5 or whatever. This I do not know. In time.</strong><hr></blockquote>

    Yes, the G4 has a higher IPC than Intel/AMD (afaik), but in most benchmarks around the 7455 is used to wipe the floor by AMD and Intel's latest and greatest. Some photoshop actions combined with small images to fit in the L3 might favour the G4, but can anyone link to benchmarks showing the G4 comeing close in IA-32? (Glad to be proved wrong).

  • Reply 15 of 65
    kecksykecksy Posts: 1,002member
    Apple doesn't need 1.6GHz G5s!

    Provided a RapidIO G4 does exist, Apple could pair one with a dual channel PC266 memory controller, and suddenly they'd have a chip with access to over 4GB/s of memory bandwidth.

    This new chip could be called the PPC7470. Clock speed would be 1GHz (so CPU and bus are clocked 1:1).

    Front-side bus would be 1GHz RapidIO. A single RapidIO channel can be 8-bits or 16-bits wide, clocked anywhere from 100MHz to 1000Mhz, and is double-pumped. At 16-bits and 1GHz (effectively 2GHz DDR), RapidIO delivers up to 4GB/s of throughput.

    Combine this with a dual channel PC2100 memory controller and the G4 has 4.3GB/s of memory bandwidth to match 4GB/s of I/O bandwidth.

    The result? According to Arstechnica, a 1GHz G4 could outperform faster x86 chips if it wasn?t so data-starved. Clock speed is a less important factor.

    But is 4GB/s enough to take away Intel's performance crown? Let's hope so.

    Of course, if Apple really wanted to spit in Intel?s face, they wouldn?t do it with the single CPU machine I just described. They?d go one step further and announce this:

    All-new PowerMac G4


    Dual 1GHz PowerPC G4 processors

    Velocity Engine vector processing unit

    Full 128-bit internal memory data paths

    Powerful floating-point units supporting single-cycle, double-precision calculations

    Data stream prefetching operations supporting four simultaneous 32-bit data streams

    512K on-chip L2 cache running at processor speed

    8MB DDR SRAM L3 shared cache, with up to 16GB/s throughput

    Dual 1GHz processor busses providing each CPU with up to 4GB/s throughput


    1GB or 2GB of 4-channel 266MHz PC2100 DDR SDRAM with up to 8.5GB/s throughput

    Four DIMM slots supporting up to 4GB of DDR SDRAM using the following:

    256MB DIMMs (64-bit-wide, 128Mb technology)

    512MB DIMMs or 1GB DIMMs (64-bit-wide, 256Mb technology)

    This equals:

    15 gigaflops of computational power

    16GB/s L3 cache throughput (250MHz 256-bit DDR SRAM)

    4GB/s of throughput available to each CPU

    8.5GB/s of memory throughput (enough to provide each CPU with 4.3GB/s)

    And of course, "Pentium-crushing performance, but now in every benchmark."

    I could care less about clock speed. Going from 1GHz to 1.5GHz doesn?t do enough on its own. What we really and what I really want is a ?Bandwidth Behemoth,? and I hope Apple delivers.

    [ 07-31-2002: Message edited by: Kecksy ]</p>
  • Reply 16 of 65
    bartobarto Posts: 2,246member
    Back to the good old days, where memory, cpu and fsb were all at the same speeds :-p

    I'd sure like to have someone's professional opinion on whether a DP 1GHz G4 with an ungodly fsb (such as the full-proc-speed system described) and memory (triple-channel DDR333 for full-proc-speed again) would outperform a P4/XP, and in what situations.

    Of course, you *might* be able to fit quad cpus on the current motherboard, but over 4 (and probably even 4) and you'd need to have a new case/mobo/cooling design to compensate for it.

    The max number of CPUs in a system is finite, so you'd still have to get clock-speed higher eventually.

    <img src="graemlins/smokin.gif" border="0" alt="[Chilling]" /> Barto <img src="graemlins/smokin.gif" border="0" alt="[Chilling]" />

    [ 07-31-2002: Message edited by: Barto ]</p>
  • Reply 17 of 65
    eugeneeugene Posts: 8,254member
    [quote]Originally posted by Barto:


    Yes, the G4 has a higher IPC than Intel/AMD (afaik), but in most benchmarks around the 7455 is used to wipe the floor by AMD and Intel's latest and greatest. Some photoshop actions combined with small images to fit in the L3 might favour the G4, but can anyone link to benchmarks showing the G4 comeing close in IA-32? (Glad to be proved wrong).


    Been discussed to death already at Ars Technica. My Dual 1 GHz holds court against the fastest single CPU Athlon XPs and Pentium 4s out there right now, even though many of the filters in Photoshop do not take much advantage of the two CPUs. And no, the 50 MB image does not fit in any cache.

    <a href="http://www.geocities.com/sw_perf/PS5Bench.html"; target="_blank">http://www.geocities.com/sw_perf/PS5Bench.html</a>;

    <a href="http://arstechnica.infopop.net/OpenTopic/page?q=Y&a=tpc&s=50009562&f=48409524&m=7760969205&; p=1" target="_blank">http://arstechnica.infopop.net/OpenTopic/page?q=Y&a=tpc&s=50009562&f=48409524&m=7760969205&; p=1</a>

    In fact, if you look, the dual 2 GHz Xeons and dual 1.533 GHz Athlon MPs aren't that much faster either.

    PS7Bench is also not AltiVec biased because the same filters/actions have been used since before the G4 was even around.

    These results makes me wish there were more AltiVec maestros like Dan Oetting and Chris Cox though.
  • Reply 18 of 65
    bartobarto Posts: 2,246member
    Thanks for proving me wrong

    So we've established that Photoshop is good on Macs vs PCs, what else?

  • Reply 19 of 65
    arisaris Posts: 65member
    id like to make a comparison/point

    I, along with prolly 90% of the computer society never thought ATI would ever be able to beat Nvidia no matter what benchmark it ran. and even if it did it would be only slightly, and only to again be beaten by Nvidia in the next month.

    and now we have ATI with a graphics card that is in the low scale 40% better and on the large scale 300% better than Nvidia's best..

    and as far as anyone can tell Nvidia wont have a reply to this for at least 6 more months. by which time ATI will be getting ready to release their next generation vid card in just a couple months.

    if ATI can do it, i think apple can do it.
  • Reply 20 of 65
    bartobarto Posts: 2,246member
    300% - Don't exaggurate! It's a measly 251% in 3DMark at 1600x1200, 4x FSAA. nVidia shouldn't worry (not).

    And don't forget the 3rd party Quad-GPU Radeon 8500, soon to be Radeon 9700

Sign In or Register to comment.