Intel at 4 GHz??? Come on Motorola...

1235789

Comments

  • Reply 81 of 170
    eric d.v.heric d.v.h Posts: 134member
    [quote]Originally posted by RazzFazz:

    <strong>So, in all honesty, you actually do believe that a "CRISC" CPU, which basically is a RISC core with additional decoding stuff slapped on, can inherently scale higher than a much less complex RISC CPU?</strong><hr></blockquote>



    No. I never said that.



    [quote]From my earlier post:

    <strong>It's called the truth. SGI's MIPS chips, DEC's Alpha, Hitachi's SH and IBM's POWER have all had much lower clock speeds than the vast bulk of CISC and CRISC chips.</strong><hr></blockquote>



    [quote]Also from my earlier post:

    <strong>The best-laid plans of mice and engineers?



    Seriously though. despite their best efforts. the RISC CPU turned out to be a fantastic "Low Frequency" chip. I wasn't talking about the original dreams of the first RISC engineers. I was talking about current reality. but the best things sometimes come from the least expected places.



    And RISC <a href="http://www.mackido.com/Hardware/WhatIsRISC.html"; target="_blank">isn't what it used to be</a> anyways?</strong><hr></blockquote>



    In other words. RISC chips don't inherently have lower clock speeds. they just happen to excel despite it anyways.



    [quote]Originally posted by RazzFazz:

    <strong>Has it ever occured to you that the clock speed advantage of current Intel and AMD CPUs is more related to the fact that both of them invest insane amounts of money in their designs, fabs and processes, rather than some inherent "low-frequencyness" of all RISC CPUs?</strong><hr></blockquote>



    Yes. I'm sure you've noted my references to the effects of the inverse: the Belluzzo/Itanium induced research droughts at SGI and HP.



    Eric,
  • Reply 82 of 170
    Well Eric,



    you really haven't addressed the economic motivation part of the dropping MIPS,



    [quote]On the contrary. what do you think has powered the N64, PS and PS2. as well as numerous printer, digital copier and network router embedded logic systems? the MIPS was(And is) one of the _most_ mainstream 64-bit CPUs on the market

    <hr></blockquote>



    Yes, this is true, but that was not the market I was talking about. Most of what you listed above are lower end applications that are more price sensitive and far less lucrative than the 64bit Server/Scientific applications. MIPS never sold well in the high end and that was even before the scare of IA64 become predominant.



    [quote]Yes. they used to have _quite_ a bit of resorces. SGI used to be first and foremost in the scientific/graphical computing world. and do you realize what an honored position third in scientific workstations behind IBM and Compaq is?<hr></blockquote>



    It means it was coming in third in a was was essentially a 5 horse race. It means it was coming in third in a market that may not have room for all five (Power, MIPS, Alpha, PA-RISC, and SPARC), and Power, PA, and Sparc all had significant marketshare in the server market to leverage development with.



    [quote]that means that the MIPS was competitive with. and surpassed _only_ by the POWER and Alpha CPUs. this means that SGI manufactured the _third fastest CPU on earth_.

    <hr></blockquote>



    I was talking marketshare and economics, you are talking about performance. One does not nessessarlily lead to the other.



    [quote]whereas the IA64 can not be said to hold such a title.<hr></blockquote>



    wrong.



    IIRC, when Merced debuted, it was second in SpecFP2000 behind the EV68 until Power4 numbers became available.



    [quote]if MIPS knew they still had SGI and Compaq as future customers. they could easily have roasted Intel's chip.<hr></blockquote>



    Probably in the short term, but not in the long term. This was what I think SGI was convinced of.



    [quote]Alpha dead after being handed on a silver plate to Intel by their thrall Compaq and even some rumors of the POWER being surrendered by IBM for some unknown reason. <hr></blockquote>



    Never heard the Power rumor. Probably some sort of joke as the Power4 is currently reigning at the high end of 64bit computing right now. It's marketshare in the very lucrative high end server market is all but assured for most of it's life cycle.



    [quote]And commodity pricing is right! with PA-RISC out of the picture from the start. and later MIPS. <hr></blockquote>



    Commodity meaning that it will price very aggressively to take marketshare away IBM, sun, and remaining R14k and PA customers.



    [quote]quote:

    --------------------------------------------------------------------------------

    Originally posted by johnsonfromwisconsin:

    No longer would SGI have to swim up a raging stream to develop it's own architecture, neither would it have to shell out a lot of money to buy competing 64-bit processors.

    --------------------------------------------------------------------------------



    Neither would much stronger companies like Dell. once their distinguishing advantages are gone. SGI's customers will scatter like a flock of pheasants after a shot. and I think that all the people at SGI obviously could have figured this out.

    <hr></blockquote>



    I don't think adoption of IA64 sounds SGI's death-knell. There's more that differentiates them from PC vendors than the fact that they use MIPS.
  • Reply 83 of 170
    blablablabla Posts: 185member
    [quote]<strong>

    Let me tell you a few things about SPEC.



    1) Did you know that SPEC has been proven to vary by more than 30% based on compiler alone?



    </strong><hr></blockquote>



    Even so, shouldnt a 1 Ghz G4 be able to beat a 1 Ghz cheapo 133 Mhz sdram low-end P3? ANY code could vary in speed, depending on the compiler used, even the very usefull app [email protected] Jupp.. its true.





    [quote]<strong>

    2) Are you aware that the gcc compiler is not what's used for production Mac apps? Are you also aware that gcc (though free) is notoriously slow?

    </strong><hr></blockquote>



    So? C't used a dog slow non-intel Compiler for the P3, and used different compilers for the Mac.. Still the Powermac wasnt able to beat it..



    <a href="http://www.heise.de/ct/english/02/05/182/"; target="_blank">http://www.heise.de/ct/english/02/05/182/</a>;



    [quote]<strong>

    3) How about the relevance of some of the tests? When was the last time you needed to do weather forcasting on your PC? </strong><hr></blockquote>



    And [email protected] is more usefull? Even if an algotithm or program is more usefull, it doesnt guarantee the mac will be faster... Im not a Photoshop guy, so why would I care about this app? Even the mac-market is not very uniform. My supervisor at the university bought a dual G4, to run Genetic algorithms on it. I think he turned back to X86. The FP-unit in the G4 is far from topnotch.



    [quote]<strong>

    4) You are correct about SPEC not allowing for code changes or optimizations. Of course, this is SPECs biggest downfall. What commercial product runs on multiple platforms that has no optimizations for a given platform? What about SIMD? This is the G4's biggest strenth, yet there is absolutely no way to measure this via SPEC. Intel has done some work in "auto vectorizing" in it's compilers, but it's nowhere near as effective as manual SIMD optimizations as seen in commercial software. To my knowledge, no auto vectorizing exists in current Mac compilers.

    </strong><hr></blockquote>



    C't didnt use auto-vectorizing compiler.... Still the G4 didnt beat it. 2 year old X86 technology, and the G4 is still behind!!! Lets hear some excuses from the Apple apologists!



    And there is at least one _code optimizer_ for the PPC:



    <a href="http://www.psrv.com/altivec.html"; target="_blank">http://www.psrv.com/altivec.html</a>;



    Its not Spec org or C'ts fault Apple/Motorola is lagging behind in the Compiler department.
  • Reply 84 of 170
    g-newsg-news Posts: 1,107member
    If they do that in CodeWarrior again, and the P3 still beats the G4, then I shall be convinced.



    G-news
  • Reply 85 of 170
    [quote]If they do that in CodeWarrior again, and the P3 still beats the G4, then I shall be convinced.<hr></blockquote>



    Supposedly, Motorola's has a compiler that is even more optimal, I think it's called "Mr. C". At least that's what it tends to show in the ArsTestBench.
  • Reply 86 of 170
    blablablabla Posts: 185member
    [quote]Originally posted by G-News:

    <strong>If they do that in CodeWarrior again, and the P3 still beats the G4, then I shall be convinced.



    G-news</strong><hr></blockquote>



    "For compiling the test programs, in the case of Mac OS X we used the version 2.95.2 of GNU C and the C++ compiler (GCC), supplied by Apple in January 2002 as parts of its developer tools. When compared briefly with the compilers by Metroworks the latter's values showed up as slightly worse. "
  • Reply 87 of 170
    g-newsg-news Posts: 1,107member
    Ein kurzer Vergleich mit den Compilern von Metroworks ergab für jene geringfügig schlechtere Werte.



    The sentence of the German original article could be interpreted either way, in favor of CodeWarrior or of the other one.



    It would also be interesting to know how they compiled it with codewarrior, as there are huge amounts of things you can do before compiling it, including Altivec opt on the fly etc.



    I still don't, and never will, believe that a P3 733 is supposedly faster than a 1 GHz G4 - and SPEC can repeatedly blow my *** . As long as real world app performance, and "Intel's Wonderland" SPEC performance don't even remotely match, SPEC an't be taken seriously.



    G-News
  • Reply 88 of 170
    [quote]SPEC an't be taken seriously.

    <hr></blockquote>



    I wouldn't dissmiss SPEC like you say.



    Try telling Engineers and CPU designers that and you won't likely be taken seriously yourself.



    Like any benchmark, you have to understand where it's value lies. SPEC scores are the measure of both the software and the hardware paradigm, meaning it shows the potential performance of an architecture with a decent (yet a more or less controlled) level of optimization. Each part of this suite is pertainent to code that is run in everyday programs who's contents are decided on by experts from industry to acadamia.



    But one thing you have to note is that this can be quite a bit different than what you'll find in commercial and home software. You may not see such optomization on thid sort of software. Intel (and about everyone else) uses a special optomizing compiler on it's scores, yet most Windows software is compiled on VC6 which isn't too great at scheduling code, and this is the difference that your seeing between SPEC and the real world.



    But your right, a 733P3 isn't as fast as a G41000 on code that has recieved a *comparably* optomized binary.
  • Reply 89 of 170
    [quote]Originally posted by johnsonfromwisconsin:

    <strong>Well Eric,



    you really haven't addressed the economic motivation part of the dropping MIPS,</strong><hr></blockquote>



    Sure I have. there isn't one(Except for Intel ).



    [quote]Originally posted by johnsonfromwisconsin:

    <strong>Yes, this is true, but that was not the market I was talking about. Most of what you listed above are lower end applications that are more price sensitive and far less lucrative than the 64bit Server/Scientific applications. MIPS never sold well in the high end and that was even before the scare of IA64 become predominant.</strong><hr></blockquote>



    Actually. the MIPS used in a lot of high-end routers, SAN boxes, telex switches, turnkey servers and such include the _very_ highest model ones MIPS offers(Did you know that Cisco uses DEC Alphas in some routers?). they're really quite lucrative. and lets not forget that SGI actually OWNED them before embarking on that foolish Itanic cruise.



    [quote]Originally posted by johnsonfromwisconsin:

    <strong>It means it was coming in third in a was was essentially a 5 horse race. It means it was coming in third in a market that may not have room for all five (Power, MIPS, Alpha, PA-RISC, and SPARC), and Power, PA, and Sparc all had significant marketshare in the server market to leverage development with.</strong><hr></blockquote>



    Yes. but the scientific and graphics space(ESPECIALLY the graphics space) were well known as SGI's home stomping grounds. IBM, DEC et al. were outsiders here. and SGI actually had a fairly viable server business. make no mistake. SGI was riding high.



    [quote]Originally posted by johnsonfromwisconsin:

    <strong>I was talking marketshare and economics, you are talking about performance. One does not nessessarlily lead to the other.</strong><hr></blockquote>



    Quite right. but like I said. SGI had it's(Quite healthy) market well defined.



    [quote]Originally posted by johnsonfromwisconsin:

    <strong>wrong.



    IIRC, when Merced debuted, it was second in SpecFP2000 behind the EV68 until Power4 numbers became available.</strong><hr></blockquote>



    So what? I'm talking about today. and today. the Alpha, POWER and SPARC all kick the snot out of the Itanium in both integer and floating point. while HP's own PA-RISC is practically equal to it.



    [quote]Originally posted by johnsonfromwisconsin:

    <strong>Probably in the short term, but not in the long term.</strong><hr></blockquote>



    I doubt it. The MIPS already stomps all over it in integer performance. and its only running at 600Mhz right now.



    [quote]Originally posted by johnsonfromwisconsin:

    <strong>This was what I think SGI was convinced of.</strong><hr></blockquote>



    For the same oft-discussed reasons that Apple has never done this. I can't imagine any sane(And loyal ) SGI employee deciding upon it. and SGI was much better off than Apple is now too.



    [quote]Originally posted by johnsonfromwisconsin:

    <strong>Never heard the Power rumor. Probably some sort of joke as the Power4 is currently reigning at the high end of 64bit computing right now. It's marketshare in the very lucrative high end server market is all but assured for most of it's life cycle.</strong><hr></blockquote>



    No. it's a bona fide rumor. but _is_ still just a rumor. heck. anything can happen. and DEC was flying high too?



    [quote]Originally posted by johnsonfromwisconsin:

    <strong>Commodity meaning that it will price very aggressively to take marketshare away IBM, sun, and remaining R14k and PA customers.</strong><hr></blockquote>



    Commodity pricing is also when you have practically no seriously threatening competition left. as well as _very_ deep pockets from your other business. so as to enter the market with prices that undercut others by being priced either at a loss or at an unsustainably low profit. and then raising prices up to whatever you feel like once dominance has been acheived(A common tactic for monopolies entering new markets or protecting old ones. MS is famous for it's use).



    [quote]Originally posted by johnsonfromwisconsin:

    <strong>I don't think adoption of IA64 sounds SGI's death-knell. There's more that differentiates them from PC vendors than the fact that they use MIPS.</strong><hr></blockquote>



    Really? like _what_? aside from IRIX and UNICOS. SGI will not have a _single_ thing that will protect them from the ravages of bigger, better established, more experienced Intel lackeys. and I don't think that SGI's OSs would hold out against the collective forces of every hacker on earth if they tried to make IRIX/UNICOS IA64 run only on their hardware.



    Speaking of Intel lackeys. remember those #&!*%[email protected] self serving IT managers that made sure no Macintosh networks would infiltrate their dippy. multimillion dollar. can-only-be-maintained-by-them-cuz-they're-such-hot-stuff 1950s era mainframe installations in the late 80s? their Intel/NT counterparts are still there. and they even exist in companies that have been SGI from the start. they've been just _itching_ for an excuse. ANY excuse to swoop in and get rid of the SGIs.



    The SGI loyalists have managed to defend themselves with comments like: "It would cost way too much to migrate." and "Re-training everyone would take forever.". but for since they'll HAVE to switch to an IA64 machine soon anyways. it wouldn't take much effort for the Intel/NT lackeys to justify swapping the SGI's the employees requested for Dells or IBMs to upper managment.



    It would only be a matter of time before the SGI loyalists would make one wrong move. and the Intel/NT lackeys would initiate a crusade to switch to MS Windows 64. that will spell the doom of SGI.



    \t\t\t\t\tEric,



    [ 03-09-2002: Message edited by: Eric D.V.H ]



    [ 03-09-2002: Message edited by: Eric D.V.H ]



    [ 03-09-2002: Message edited by: Eric D.V.H ]



    [ 03-09-2002: Message edited by: Eric D.V.H ]</p>
  • Reply 90 of 170
    HEHE, this is turning into a neat discussion



    [quote]Actually. the MIPS used in a lot of high-end routers, SAN boxes, telex switches, turnkey servers and such include the _very_ highest model ones MIPS offers<hr></blockquote>



    a lot of that consists of lower end MIPS



    [quote](Did you know that Cisco uses DEC Alphas in some routers?).<hr></blockquote>



    Yeah, they also use the old Motorola 68k as well.



    [quote] they're really quite lucrative. and lets not forget that SGI actually OWNED them before embarking on that foolish Itanic cruise<hr></blockquote>



    SGI owning DEC? I don't think so, I think DEC was it's own entity before being purchased by Compaq.



    [quote]I doubt it. The MIPS already stomps all over it in integer performance. and its only running at 600Mhz right now.

    <hr></blockquote>



    A couple of things:



    Itanium has been downplayed by even Intel (well, after the initial silly hype) as a development platform. McKinnley is currently shipping to vendors and I think will be available by sometime in Q2.



    The MIPS R14k is a very wide issue processor and usually these are a bit difficult to scale. Yes it does a lot at 600mhz, but that alone should impress you if it can't keep speeds up to be competitive.



    One thing Intel boasts that SGI doesn't is a world class process technology. Implementing better and smaller processes is expensive, and it's cost increases greatly with each new process generation. Manufacturing process and implementation, not excellence of ISA design (which is becoming less important) is where Intel will be able to push it's architecture forward. Regaurdless of whether SGI's development of MIPS had been somewhat interrupted, it would never have been able to compete in this way.



    [quote]Really? like _what_? aside from IRIX and UNICOS. SGI will not have a _single_ thing that will protect them from the ravages of bigger, better established, more experienced Intel lackeys. and I don't think that SGI's OSs would hold out against the collective forces of every hacker on earth if they tried to make IRIX/UNICOS IA64 run only on their hardware.<hr></blockquote>



    Well, as far as your fear of hackers grafting IRIX onto cheaper IA64 PC's:



    1)IA-64 won't replace cheap x86 units for years and SGI's units will probably be price competitive with Dell's and Hp's stuff.



    2)SGI has some proprietary Graphics systems I believe



    3)I don't think "Every Hacker" will be trying to crack IRIX to run on Dells. I don't see that kind of activity killing demand for machines as a lot of production houses need viable support.



  • Reply 91 of 170
    g-newsg-news Posts: 1,107member
    I should have said: SPEC can't be taken seriously outside of the developer's own labs.



    It's completely useless to the common user.



    G-News
  • Reply 92 of 170
    quaremquarem Posts: 254member
    [quote]Originally posted by G-News:

    <strong>I should have said: SPEC can't be taken seriously outside of the developer's own labs.



    It's completely useless to the common user.



    G-News</strong><hr></blockquote>



    I am not picking on you directly G-News but I thought it is funny that when the Register posted those supposed SPEC marks scores for the G5 no one on this board was claiming that they were useless to the common user. No one was saying, hey that SPEC Mark, totally inaccurate the P4 might still beat the G5. And now we have a SPEC mark for the G4 and everything is claiming that it's bogus. Its not!



    Those SPEC marks have an value in showing comparable FPU and Integer performance on the G4 and Pentium III. Sure you can vectorize for Altivec and optimize for the G4, but that isn't what they were testing. These scores are important in that the majority of the code that your CPU crunches fall into these two categories, FPU and INT. What is more disturbing to me is that the PIII 1 GHz has been out for a year and a half.



    SO in reality it is that Apple is using Altivec in their main stay applications, or segments were they can, BLAST, Aqua, Photoshop, etc. To deliver real-world performance that won't reflect those SPEC marks. Alti-vec is so good at doing what it does that it has allowed Apple to maintain a performance lead in certain areas that matter the most to its core user group.



    Apples still aren't dog slow in real world performance in many areas, but as these SPEC marks evidence, their are going to be certain tasks that the G4s seriously lag behind the x86 world: most notably 3D and apps in Science and Engineering.



    My advice would be not to reject these scores, but to accept that Apple has some hurtles to overcome, and will overcome, and continue to put pressure on Apple to get some faster processors into their machines.
  • Reply 93 of 170
    g-newsg-news Posts: 1,107member
    First of all it's not Apple's thing to see to it that the PPC gets better FPU scores, it's Motorola's and maybe IBMs. as long as Apple doesn't make their CPUs alone, they can only help, not dictate.



    As for those SPEC scores on the register for the G5: Are you telling me you believed that??



    I know that for my part I didn't...they were way above the Power 4 scores even...very unlikely.



    I'll let SPEC live for those who like it, developers who are interested in chip performance alone and don't give a damn about total system performance, real world app performance etc.

    As long as they can Penis compare their FPU units with their competition, that's fine with me.

    Maybe that's why Moto doesn't submit its scores anymore. Maybe they've grown up so much to know that it's no the size that matters, but how you use it.



    I for my part will prefer to look at real world apps and real world feel.



    Last but not least: on int the P3 and G4 scored almost identical, within the range of error at the same speed. That's pretty nice given the 604 wasn't the best chip at INT back then.



    G-News
  • Reply 94 of 170
    [quote]Originally posted by johnsonfromwisconsin:

    <strong>a lot of that consists of lower end MIPS</strong><hr></blockquote>



    [quote]Originally posted by johnsonfromwisconsin:

    <strong>Yeah, they also use the old Motorola 68k as well.</strong><hr></blockquote>



    I already said that. I was just noting the fact of that MIPS' higher end chips are in use too.



    [quote]Originally posted by johnsonfromwisconsin:

    <strong>SGI owning DEC? I don't think so, I think DEC was it's own entity before being purchased by Compaq.</strong><hr></blockquote>



    :rollyeyes:Ugh. I was referring to MIPS Technologies Inc.



    [quote]Originally posted by johnsonfromwisconsin:

    <strong>A couple of things:



    Itanium has been downplayed by even Intel (well, after the initial silly hype) as a development platform. McKinnley is currently shipping to vendors and I think will be available by sometime in Q2.</strong><hr></blockquote>



    And I hear the POWER5 and Sun's new chip among numerous others will be showing up by then too. all of which should easily put McKinley in it's place easily.



    [quote]Originally posted by johnsonfromwisconsin:

    <strong>The MIPS R14k is a very wide issue processor and usually these are a bit difficult to scale. Yes it does a lot at 600mhz, but that alone should impress you if it can't keep speeds up to be competitive.



    One thing Intel boasts that SGI doesn't is a world class process technology. Implementing better and smaller processes is expensive, and it's cost increases greatly with each new process generation. Manufacturing process and implementation, not excellence of ISA design (which is becoming less important) is where Intel will be able to push it's architecture forward. Regaurdless of whether SGI's development of MIPS had been somewhat interrupted, it would never have been able to compete in this way.</strong><hr></blockquote>



    That's what I said in the first place.



    [quote]From my older post:

    <strong>As for MIPS. the research money for high end MIPS has dried up since that slime Belluzzo(Whom promptly moved to COO at MicrosoftÉ) decided that SGI would start migrating to Intel's sucky IA64. and when Compaq subsequently killed it's himalaya line for the same reason. if they(MIPS) still had those two income sources. the MIPS would still crush all CISC and CRISC comers.</strong><hr></blockquote>



    If MIPS had the money and backing it used to. their technologies would easily exceed of Intel's(I hope you notice that MIPS and PA-RISC, the two architectures owned by companies tainted by Belluzzo's foul hand. are the only ones Intel currently outclasses).



    quote:

    Really? like _what_? aside from IRIX and UNICOS. SGI will not have a _single_ thing that will protect them from the ravages of bigger, better established, more experienced Intel lackeys. and I don't think that SGI's OSs would hold out against the collective forces of every hacker on earth if they tried to make IRIX/UNICOS IA64 run only on their hardware.



    [quote]Originally posted by johnsonfromwisconsin:

    <strong>1)IA-64 won't replace cheap x86 units for years</strong><hr></blockquote>



    I didn't say they would.



    [quote]Originally posted by johnsonfromwisconsin:

    <strong>and SGI's units will probably be price competitive with Dell's and Hp's stuff.</strong><hr></blockquote>



    I doubt it. Dell, HP etc. like you said. have IA64 markets in many other places than the ones that SGI does. the higher volume at which they buy their parts would give them lower prices.



    [quote]Originally posted by johnsonfromwisconsin:

    <strong>2)SGI has some proprietary Graphics systems I believe</strong><hr></blockquote>



    I don't think that's enough to insulate them from the ground level assaults of the other IA64 companies.



    [quote]Originally posted by johnsonfromwisconsin:

    <strong>3)I don't think "Every Hacker" will be trying to crack IRIX to run on Dells. I don't see that kind of activity killing demand for machines as a lot of production houses need viable support.</strong><hr></blockquote>



    I guess you're right on the support count.





    Eric,
  • Reply 95 of 170
    quaremquarem Posts: 254member
    [quote]Originally posted by G-News:

    <strong>First of all it's not Apple's thing to see to it that the PPC gets better FPU scores, it's Motorola's and maybe IBMs. as long as Apple doesn't make their CPUs alone, they can only help, not dictate.



    As for those SPEC scores on the register for the G5: Are you telling me you believed that??



    I know that for my part I didn't...they were way above the Power 4 scores even...very unlikely.



    I'll let SPEC live for those who like it, developers who are interested in chip performance alone and don't give a damn about total system performance, real world app performance etc.

    As long as they can Penis compare their FPU units with their competition, that's fine with me.

    Maybe that's why Moto doesn't submit its scores anymore. Maybe they've grown up so much to know that it's no the size that matters, but how you use it.



    I for my part will prefer to look at real world apps and real world feel.



    Last but not least: on int the P3 and G4 scored almost identical, within the range of error at the same speed. That's pretty nice given the 604 wasn't the best chip at INT back then.



    G-News</strong><hr></blockquote>



    I am not saying that I believed the SPEC marks for the rumored G5, just pointing on that no one was claiming that they could be false and that in fact the P4 could potentially be faster. (Someone on these boards might have, but no one mentioned that from what I read, or can recall )



    You're right, a lot of real world performance is great on the Mac right now, but in a lot of areas its not. For example, Quake3. Quake3 scores are typically slower on the Mac then on the PC. I have also heard that 3D performance is lacking on the Mac compared to the PC. FPU is also critical for science and engineering apps, and since SPECfp uses real world code, I think it is plausible to infer that a lot of science and engineering apps are slower on the Mac then the PC.



    So lets just be honest with our selves and admit that Macs need more speed in some areas, while in other they hold a strong lead over the PC competition.



    I don't think Motorola has matured in some way as a company and that is why they haven't released SPEC scores in a while. Most likely it is because they know those scores are lacking.



    I think its funny how Motorola is causing all these problems for Apple in the CPU department but at the same time provided the one technology that allowed Apple to get through a CPU lag and maintain real world performance in a lot of areas with PCs, Alti-vec. I am so thankful for that one, very well designed vector-processing unit.
  • Reply 96 of 170
    [quote]the Alpha, POWER and SPARC all kick the snot out of the Itanium...floating point<hr></blockquote>



    Newest Alphas? Most likely.

    POWER4? Yes.



    SPARC? &lt;snicker&gt; Got uncontested proof? Because I call bullshit. SPARC sucks in FP compared to IA64 processors.



    At any rate, "McKinley" is expected to have competitive SpecINT 2K performance, and possibly best in SpecFP 2K, at least until EV7 systems show up. It's unknown which will roll out first.



    One thing's for sure...barring outliers and cheating, IA64 systems are probably always going to crush anything Sun can turn out in FP with SPARC, comparable systems wise. It's simply a vastly superior architecture for FP.



    [ 03-10-2002: Message edited by: TheAlmightyBabaramm ]</p>
  • Reply 97 of 170
    g-newsg-news Posts: 1,107member
    Well, considering the G4 has but one doubel repc FPU, while the P3 has what, 2, 3?? it's still quite impressive.



    G-news
  • Reply 98 of 170
    mattyjmattyj Posts: 898member
    Well, lets face it, if it wasn't for altivec, Apple would be a lot worse off than they are now.



    This is why people are hoping that the G5 will go up to 4Ghz, have a 333Mhz DDR memory bus, ata 133, etc.



    To be blunt, the G4 series is a bit of a flop, and motorola should be dumped, apple should take the G5 into their own hands (if they haven't already), and make a better job of it.
  • Reply 99 of 170
    g-newsg-news Posts: 1,107member
    If only it was so easy as you say.



    G-News
  • Reply 100 of 170
    [quote]And I hear the POWER5 and Sun's new chip among numerous others will be showing up by then too. all of which should easily put McKinley in it's place easily.



    <hr></blockquote>



    I don't know where your hearing the Power5 stuff from, but the Power4 isn't even out in force yet. And personally, I have little faith in Sun's ability to put out world class individual MPU's.
Sign In or Register to comment.