“Apple wants to make their switch to Intel chips seem like a no-brainer, but the reality of it was a lot more complicated than just faster chips for Macs. Apple’s claims of their Intel systems being ‘4-5x faster’ than their PowerPC systems is a little much to swallow, especially with Intel Macs landing in users’ hands and failing to live up to the hype. So if these Intel chips aren’t really that much faster than the G5, why did Apple make the switch? The answer to this question is a lot more interesting than what Apple’s telling you.”
Please, as an owner of a new 2 GHZ 20″ iMac and 1.5 GHZ G4 (who uses a G5 iMac sometimes in lab)
This thing absolutely SCREAMS. UT2004 is actually playable on a Mac now, among other things. Failing to live up to the hype?
This machine has delivered everything and then some IMO.
If you RTFA, you’ll discover that he doesn’t back these claims up, or deal with the “real” reasons Apple made the switch to Intel (as opposed to, say, AMD). It’s mostly just a recent history of the PowerPC line of CPUs and a recap of things everyone already knew.
– chrish
First of all UT2004 throws the sound on the other core to get some game engine performance.
Second Mac OS X and other OS related stuff is on the other core as well even further freeing up the first core. Naturally if you don’t have another core the first one has to do all the work.
Third the hard drive has a bigger cache than before, which makes the map and game loading faster.
Fourth the video card in the new iMactels is much more powerful than the iMacG5
Fifth when UT2005 or 6 comes out it will run like a pig on your iMactel, but with my Dual 2 Ghz PowerMac G5 all I have to do is upgrade the video card. Because 3D games are mostly video card dependant.
I can actually run 2 3D games at once above 40fps each.
So yes you got a newer better iMac, it’s to be expected when you upgrade, the machine is 2x faster (in some tests) than a single core iMac G5, but that’s to be expected because DUH, it has another core.
However the Core Dual is 32 bit and is not faster than a most dual core PowerMac’s which are 64 bit and can beat the 4GB RAM limit.
So the Core Duo isn’t a great leap of performance power. It also scores not that much better than a iMac G5 on X-Bench scores. Around 60. My Dual 2 Ghz (3 years old) runs about 100 and the Quad is a monster.
What is got you excited is you got a taste of PowerMac performance in a consumer box, but when the new 3D games arrive the temptation to get a Quad will show itself.
So the Core Duo isn’t a great leap of performance power. It also scores not that much better than a iMac G5 on X-Bench scores.
There are so many real-world open source programs that could be used as benchmarks, so why the heck do people keep on writing and believing those pointless little synthetic closed-source “benchmarks” like XBench or geekbench?
>Fifth when UT2005 or 6 comes out it will run like a pig on your iMactel, but with my Dual 2 Ghz PowerMac G5 all I have to do is upgrade the video card. Because 3D games are mostly video card dependant.
I don’t think so. As your Powermac will be deprecated, you will not find any video card from the last generation and be owned.
Even today, PowerMac only have only three video cards, and not the better one…
Another thing, EPIC (yes those who make UT2006/7) have always said that they HATED altivec and prefered SSE. So the next UT will be more optimised for SSE (the poor altivec optimisation will be dropped as they don’t have to anymore) and you will be, one time again, owned.
Edited 2006-02-07 20:07
First of all UT2004 throws the sound on the other core to get some game engine performance.
Sound is like, what 3% of the total overhead?
Second Mac OS X and other OS related stuff is on the other core as well even further freeing up the first core.
That’s not correct. “Mac OS X” isn’t a program. It’s a set of libraries and the kernel, which behaves like a shared library in that it runs on whatever core the thread that called it was running on. If you have a single threaded app that calls OS services, those services will run on the same processor as the application.
Naturally if you don’t have another core the first one has to do all the work.
Take a look at your CPU utilization with nothing running. On a fast machine, its nearly zero. What tiny CPU is used is the result of daemons running in the background, which kick in once in a long while (from the POV of the processor). That’s all the CPU load a second processor will take off when running a single-threaded application.
Fifth when UT2005 or 6 comes out it will run like a pig on your iMactel, but with my Dual 2 Ghz PowerMac G5 all I have to do is upgrade the video card. Because 3D games are mostly video card dependant.
Actually, Apple’s OpenGL stack is so bad that games are quite often CPU dependent. The CPU cannot feed the GPU fast enough.
I can actually run 2 3D games at once above 40fps each.
Welcome to PC video gaming performance circa 2003.
However the Core Dual is 32 bit and is not faster than a most dual core PowerMac’s which are 64 bit and can beat the 4GB RAM limit.
First, I’d hope the iMac isn’t faster than the DC PowerMacs, since the latter cost quite a bit more. Second, its unfortunately not the whole story. On integer code, the new iMacs will beat anything short of a Quad.
So the Core Duo isn’t a great leap of performance power. It also scores not that much better than a iMac G5 on X-Bench scores. Around 60. My Dual 2 Ghz (3 years old) runs about 100 and the Quad is a monster.
XBench is bogus. That seems to be the consensus on the macnn forums, which entertainingly enough is full of people who seem to actually like Macs, as opposed to people who have a weird attachment to PowerPC.
What is got you excited is you got a taste of PowerMac performance in a consumer box, but when the new 3D games arrive the temptation to get a Quad will show itself.
And the point of the comparison is? The Quad costs twice as much (without monitor), weighs twice as much, sucks down three times the electricity, and makes much more noise. By your logic, the Quad is a pokey machine, because it doesn’t stand up well to an 8-way Opteron server…
Don’t get me wrong. The PowerMac G5 is still a good machine — for it’s era. Unfortunately, the G5 is more comfortable hanging out with circa-2003 processors instead of modern x86 chips. That’s okay — x86 will come to save the PowerMac soon enough. In the meantime, just be happy that you can run all your PowerPC software that doesn’t have a UB equivalent. At least, that’s what I told myself when I found out the new Mactel can run XCode builds way faster than my DC 2.3 PowerMac, although ironically enough, Apple broke my must-have PPC-only app (Matlab) in 10.4.4.
Edited 2006-02-07 22:26
Apple’s benchmark claims are for very specific tests. To say these are misleading may be because the media is not attributing what the numbers mean. The numbers are correct on the two tests Steve himself explained this at the unveiling.
I’m just getting tired of people claiming they are as fast Apple says. When in fact, again, Steve these WON’T make the computer 2x as fact but would make them perform better.
Well, the intel core duo is faster than G5, 2x faster in many cases. Not because the intel cpu is better, but because the code duo has two cores on it.
Of course ibm could have updated the imac with a dual core G5 like they did with the powermac, but that would have made the switch less convincing…
Of course ibm could have updated the imac with a dual core G5 like they did with the powermac
Yes, sticking a huge hot workstation processor inside a fricking SFF machine is entirely within the realm of possibility. Heck, Apple should’ve put an FX-60 in there — since they apparently aren’t beholden to the laws of thermodynamics…
Intel chips will get faster, it also opens the door to use the existence of AMD as an underlying threat, it makes XP easier to virtualise, easier for everyone to port their wares to Mac….
Pretty compelling move no ? especially with the long sighted planning that Apple is able of.
How do they compare a 3.2 GHz Xenon to a 2.5 GHz G5? The G5’s IPC for any code more complex than a game is going to be way more than Xenon’s. The Xenon doesn’t run faster because IBM put more effort into it, it runs faster because its a comically simple CPU design. A Mac site should be the last one to mistake clock-rate for overall processing capability. After all, a the P4 ran at 3.2 GHz a year ago, but its still a mediocore chip by today’s standards.
Yeah, I love how sites who had been preaching the megahertz myth are now equating frequency with chip performance…
There’s really not much interesting in this article that can’t be found in articles that were written months ago…
Maybe I didn’t read all those articles months ago, but I found it quite interesting how Motorola and IBM both failed Apple because the embedded market was more profitable to them than desktop systems.
These conspiracy theories were stupid when we first found out about the switch. One this late, that doesn’t propose anything we haven’t heard many times before, is way worse. I don’t normally complain about the stuff that gets linked from here, but did you actually think anyone would benefit from reading this?
Just face it, the new Intel Imac keeps up with the G5 Power Mac. That’s something to say about a laptop processor.
Microsoft can heat and the problems associated with it’s line of processors. I highly doubt MS will use IBM again.
Over time, I’ve started to believe that there were two primary reasons Apple ditched PowerPC:
1) Marketing. Constantly playing the “PowerPC really is fast… when you sacrifice virgin blood and make 110% use of AltiVec!” game is really kind of pointless for them. It just detracts from their real strengths.
2) There is little future in desktop PowerPC chips. This one is the big one. Apple cannot get IBM or Freescale to sustain development of a custom chip just for them. Therefore, they have to depend on desktop-varients of chips that Freescale and IBM already have. Apple was forced to move away from Freescale, because their focus on the embedded market meant that their chips, while low power, were also pretty slow for desktop tasks. IBM had something Apple could use, in the form of the POWER4/5 architecture, which was stripped down to make the G5, but that architecture is now at a dead end.
POWER6 will be a very different design, taking a cue from Cell and the PPE and moving to a narrower core that can be clocked very high (4-5GHz). The performance estimates IBM has posted are mixed: 2200-2600 SPECint and 4600-5200 SPECfp. The SPECfp numbers are great, but SPECint is very dissapointing for something that is running at 4 GHz. This is especially considering that Conroe’s SPECint should be well into the upper part of this territory when it ships this summer, a full year before POWER6, and at least 18 months before one could expect a POWER6-Lite. POWER6, being a high-clockspeed design, is going to be very dependent on massive memory bandwidth (32GB/sec is rumored) and enormous caches, and will dissipate quite a bit of heat. Now, this tradeoff is likely a fine one for IBM, which will put it into machines costing $10,000+, but its not such a good one for Apple. A stripped-down POWER6-Lite, with a consumer sized cache (2MB) and a consumer-sized memory bus (12.8GB/sec at best), is not going to be an impressive chip. Not to mention it will be impossible to shoehorn into a laptop.
It is thus easy to see that PowerPC was a dead-end for Apple. There was no modern G4 replacement in sight, and the potential replacement for the G5 was quite underwhelming.
There is a single-core PPC970MP. Could have been used in a laptop fo-sho.
I don’t see how. The 2.5 GHz 970MP is 100W (allegedly). Even in the most ideal of cases, then, you’re looking at a 40W 2.0 GHz single-core G5 (100 * 0.8 * 0.4). In reality, though, the power dissipation increase from single to dual core is a lot less than 2x, so you’re looking at ~50W realistically, and then only if you can get the voltage down a bit at 2 GHz. That’s a power dissipation about 2x as high as the G4, and about 2/3’s higher than Yonah’s. For your extra 20W, you get a chip that’ll perform like a 1.8 GHz single-core Yonah (near the lowest end of Intel’s linup) for most tasks.
I don’t see how. The 2.5 GHz 970MP is 100W (allegedly). Even in the most ideal of cases, then, you’re looking at a 40W 2.0 GHz single-core G5 (100 * 0.8 * 0.4). In reality, though, the power dissipation increase from single to dual core is a lot less than 2x, so you’re looking at ~50W realistically, and then only if you can get the voltage down a bit at 2 GHz.
Yonah and Turion are both low power chips, that is they use low power transistors specifically designed to consume lower power. The downside is they can’t be clocked to a high rate as they can’t take the higher voltages higher clocked parts use. The G5 is designed fr higher clock rates and uses high power transistors.
IBM are quite capable of building a low power G5 but they would only have one customer for it who are known for driving a hard bargain. IBM probably thought Apple would be more interested in Cell but evidently not.
Does IBM even have a high-performance/low-power 65nm process? I don’t doubt IBM could design a power-optimized single-core G5, but that’s a big leap from saying the single-core 970MP could be fit into a laptop.
“POWER6 will be a very different design, taking a cue from Cell and the PPE and moving to a narrower core that can be clocked very high (4-5GHz). The performance estimates IBM has posted are mixed: 2200-2600 SPECint and 4600-5200 SPECfp.”
It seems very unlikely that IBM would quote numbers this early. Could you provide a link? Btw, the real world tech article was speculation by David Kanter, not an IBM quote.
Hmm, good point. Should have read the article more carefully! In that case, POWER6’s integer performance might be quite better than expected, since the PA-Semi chip he gets the estimates from is quite a bit simpler than POWER6 is projected to be (3 issue versus 4 issue, etc). If the 604 reference is accurate, POWER6 will be wider than the PA Semi chip, though narrower than the G5.
I am certainly no Apple defender–far from it–but it annoys me when OSNews publishes articles that are factually incorrect for sake of drumming up readers.
Point of Fact:
NO PERSON, I mean no one, has in their hands the Mac which Apple claims is 4-5 times faster. That claim is for the new Mac Book which wil not ship until 15 February.
As for the iMac which is shipping now which Apple claimed was 2x times faster, we see the same old boring debate everytime a company claims any benchmark. (Doesn’t matter which company NVIDIA, Intel, ATI, you name it.) And some people find that the chip is not as fast as claimed based on different tests performed, while others find the chips to be as fast or faster than claimed.
All I know is this:
No one can test the 4-5x claim yet.
The new Intel Macs are as fast, and in many cases faster. Depends on usage. But for the same price, this is faster. Therefore, it is all good.
Edited 2006-02-06 23:30
The website used to say 2-3x faster, and this related to SpecInt and SpecFP scores shown by Jobs at the launch keynote. He even conceded at the keynote that users wouldn’t see such a significant performance boost in typical usage (and Jobs is not a man who likes to make concessions at keynotes).
The iMac and the Macbook Pro apparently use the same motherboard and the same CPU’s. The fastest Macbook will be 1.83 ghz; the fastest iMac is 2.0. I’d guess the speed will be pretty much the same.
Singer asked Mayer about the G5 and Apple’s move to Intel-based Macs:
Weren’t you there during the discussions when IBM convinced Apple to adopt the G5?
Mayer: In my previous job, I ran IBM’s semiconductor business. So I’ve seen both sides of the Apple story, because I sold the G5 to Steve (Jobs) the first time he wanted to move to Intel.
Five years ago?
Mayer: Yeah, that’s about right. So I sold the G5. First I told IBM that we needed to do it, and then I sold it to Apple that the G5 was good and it was going to be the follow-on of the PowerPC road map for the desktop. It worked pretty well. And then IBM decided not to take the G5 into the laptop and decided to really focus its chip business on the game consoles.
Ok can I get some score cookies now? Thank you.
http://news.com.com/Is+the+PowerPC+due+for+a+second+wind/2008-1006_…
Here is some realistic performance scores, the 32 bit Mactels are not all about that in actuality.
http://www.geekpatrol.ca/article/101/geekbench-comparison
Oh god. Geekbench is spreading! I’m having BYTEMARK flashbacks! Help meee!!!
Yeah, except that those scores and testing methodologies are so flawed that you might as well had monkeys pounding on number keypads to come up with them.
Maybe that’s what Steve did…
no, he cited and said that he uses spec.
Wow, Macslash at it’s best.
Short summary of the article:
Long, long time ago, the G4 was stuck at 500MHz. Then it finally climbed to 1.67GHz.
A few years ago, god promised 3GHz G5s. They didn’t come, the G5 maxes out at 2.7GHz.
The Intel Xeon runs at 3.2GHz! More GHz! Intel is so much better! It roxors!
That’s why Apple had to replace the 2.1GHz (G5) iMac with a 1.8GHz (Core Duo) iMac, it’s obvious.
They werent talking about Intel’s Xeon, they were talking about the Xenon designed by IBM for the XBOX.
They’re switching because Intel has better chipsets and CPUs for notebooks.
No need for tinfoil hats.
“They’re switching because Intel has better chipsets and CPUs for notebooks. ”
Thank you.
OS News is hitting the bottom of the barrel when it quotes MacSlash. Sheesh. But then Thom is not exaclty discerning about what he posts. Quantity over quality, keep those page views coming!!
People are missing on this, and it’s one of the most obvious reasons, is the power and heat to processing ratio.
Intel chips may not be the fastest per clock cycle, lets face it after the YEARS of the claims of the sub 1ghz power being the equal of 2 or three ghz x86 the sudden flip in position sounds like pure marketing bull.
… and intel chips certainly aren’t cheap for what you get …
But the P4M’s are DEFINATE contenders for cool running and low power consumption – both items EXTREMELY important on Laptops and small form factor designs; as such I figure a major attention grabber for Apple. Look at what they are releasing FIRST on the x86 line – iMacs and iBooks. The i line of computers have been and remain the budget entry models for Apples. Expecting a iMac or iBook to be a ‘screamer’ is just silly.
Don’t know how many of you have delved inside a iBook, but back in the G3 days things are rather… amusing. If you open up a ‘toilet seat’ first gen iBook, Apple spent the extra money on 433-600mhz cpu’s, then underclocked them to a mere 300-400mhz so they could run them without heat sinks. In fact, they put insulating foam around the cpu, chipset and graphics card trapping the heat IN. (I tell you, first time I opened one of those up…) You can ‘overclock’ them back up to the chips real speed (or at least close to it) stable by ripping out the dialup modem and three layers (3?!?) of RF shield to make room for a heat sink.
Ah that ‘apple’ quality people always praise.
So Apple showing interest in low heat/low power chips is the real no-brainer… Hell, I’d not be surprised if there isn’t even a CPU fan on the iBook version; it is Apple after all.
Edited 2006-02-06 23:47
Intel chips may not be the fastest per clock cycle, lets face it after the YEARS of the claims of the sub 1ghz power being the equal of 2 or three ghz x86 the sudden flip in position sounds like pure marketing bull.
Both the Opteron and the Pentium-M overtook the G4 and G5 on per-clock performance.
Both the Opteron and the Pentium-M overtook the G4 and G5 on per-clock performance.
Which is why tons of WWDC sessions were held on leveraging Altivec/Accelerate.framework – applications that used these got a performance out of the PPC chips that was at least equal if not better to the x86 SIMD equivalents (hint: Why do you think those Bioinformatics guys like XServe G5s? It’s not the blinky lights on aluminum bezels that caught them).
However, most developers didn’t take advantage of Altivec, even when they could have. They were too busy with the other things Apple asked them for already – adopting Carbon Events, moving to Xcode, replacing QuickDraw with Carbon – all that.
“People are missing on this, and it’s one of the most obvious reasons, is the power and heat to processing ratio.”
It is the most obvious reason, because it’s the one that Mr. Jobs mentioned when announcing the switch to Intel. Comparisons to the G5 are useless, since this isn’t about the iMac, the PowerMac or the XServe: It’s about portables – you know, those things that have surpassed desktops in sales and might become even more important, business-wise.
Apple had the crown in laptops long time ago, when they introduced the Titanium G4 and the dual-USB iBook. The tide turned when Intel came with the Pentium M and PC notebooks started including x-Black/Glare/whatever displays: Suddenly the PC notebooks were not only faster but they also had better displays and got more operating time out of a battery. Then, with the Core Duo, Intel was able to almost double the performance without much impact on battery life. From one day to the other (CES 2006), every PC maker would have a $1500 laptop with two cores that beat the shit out of a PBG4 – Apple had to do something. G4? Too slow. G5? Needed to much power. Turion64? Not too great either. Core Duo? Bingo!
Guys, stop saying that “G5 wasn’t fast enought” because it wasn’t able to deliver a 3 Ghz CPU. It’s FALSE
I’m sick of hearing that. Intel has been the ONLY CPU vendor that has needed to go beyond the 3 GHz limit to keep competitive. AMD has and is still beating Intel with cpus that work under 3 ghz.
In fact, look at the Intel core duo: It’s FAR from having 3 GHz. So, Apple swtiched to intel because IBm wasn’t able to deliver a 3 GHz CPU, and the new Intel CPUs are running a 2.2 GHz? Dude, 3GHz was not the reason to ditch IBM – it was a excuse which made easier to convince users and investors
The G5 wasn’t fast enough because it wasn’t able to deliver 3GHz. Denying it isn’t going to make it less true.
The G5 is a high-clockspeed design. It has a fairly long pipeline (16 stages) and lower IPC than a P-M or Opteron. It was designed to run at higher clockspeeds to make up for its longer pipeline. Design issues (likely thermal problems) prevented IBM from reaching its target clockspeeds at 90nm. At the lower clockspeeds that were reached, the G5’s performance was correspondingly less impressive. It’s like Prescott, to a lesser degree. Prescott had a 5GHz target. Had it reached that target, it would’ve been awesome. At 3.6 GHz, it’s fairly “bleh”. Same thing for the G5. At 3 GHz, it would’ve been awesome. At 2.5 GHz, it does a good job keeping up with lower-end Opterons.
Edited 2006-02-07 00:06
It’s going to die, the company, anyway.
MICROSOFT WILL TAKE THOSE BIG $$$ ACCOUNTS FROM IBM. Just like it did with the company with the obsolete network OS knows as Novell.
😀
You’re probably just trolling, but IBM is not going anywhere. Trust me. They may, and have, changed a lot over time, but they aren’t just going to “die”. Neither did Novell, by the way, so your reference makes no sense.
IBM haven’t change a bit for over 30 years. They’re still pushing their mainframe crap that no one needs. Trouble is that the world changes daily.
While Novell’s dead in the wood. All those Netware are being replaced by Windows Server 2003 servers. Pretty soon will be completely irrelevant except for history books.
…ig about Apple when Jobs shut down the cloners. Motorola and IBM had invested considerable sums to make the PowerPC a dominate platform, and it was heading towards that, when Jobs screwed them over by pulling the plug. The fact that they kept supplying him with anything is a testament to their class, and Jobs total lack thereof.
right… a guy saving the company he runs from going bankrupt is classless.
what ever
ha, and double ha. Apple survived in spite of Jobs, not becase.
…ig about Apple when Jobs shut down the cloners. Motorola and IBM had invested considerable sums to make the PowerPC a dominate platform, and it was heading towards that,
In your dreams. The clones were just eating into Apple’s own sales rather than convert Windows users. Not surprising since Apple’s operating system at the time was dated und unstable.
You’re giving away the secret. 🙂
“IBM probably thought Apple would be more interested in Cell but evidently not.”
Here we go….Cell is not a desktop or laptop processor.
remember around 1997 when Gates was keeping Intel from getting uppity by rumouring a “Windows computer”? Intel had to play ball or risk a version of Windows that ran on anything BUT Intel. It insured mutual cooperation, Intel made the new chips, Microsoft made the increasingly puffy OS’s that *required* new chips, everybody wins. Well think about it, Apple = proprietary hardware. Make that hardware x86 compatible, and you have your proprietary hardware platform, just as soon as Windows runs better on these so-called “Macs” than any Mac OS does. Let the hackers be the first to make Windows run on Mac, avoid any “monopolistic” legal charges. Devalue the Mac OS in increments, such as by announcing there will be no more versions of Internet Explorer for Mac. Break Office compatibility next. Make it so people will *prefer* or even *require* Windows on the so-called “Mac”.
Then all that’s left is to screw Intel for past greivances by switching to AMD or somebody, and game over. Anyone that thinks this is far fetched should study Gate’s tendency to go in for the slow kill on people he has a grudge against, such as the way the spent a small fortune to beat a $78 speeding ticket.
I remember reading somewhere that Steve Jobs once tricked IBM into paying
for NextStep 1.0 license, but if IBM wanted Nextstep 2.0 IBM had to pay again for the license.
He was known to hate IBM because of the
bureaucratic and inefficiency.
If Steve Jobs were still at Apple when IBM/Apple/Motorola
alliance was forged, he would most likely forbid it. It would more likely be an alliance between Apple/Motorola/Sun ( and Texas Instruments???? )…
to use the Sun’s Solaris/SunOS and SPARC.
Look at MAE ( or MAS?) on sun workstations.
Or maybe Apple would have stick with Motorola’s
88K RISC instead… Oh well.
Or maybe Apple would have stick with Motorola’s
88K RISC instead… Oh well.
Or they could simply have stayed with the 68k and avoid all the trouble of processor switches and emulations.
Motorola could have developed the 68k in much the same way that Intel & AMD developed the x86 line. The 68060, which Apple never used because it would have outshone the first PowerMacs, showed what would have been possible.
WHY??? IBM IS A LOSER THAT’S WHY!
IBM couldn’t keep up with the demand for Apple for one thing and if they were able, I’m sure Apple would of thought different in staying with em.
Only way IBM can keep up now is to outsource to other fabs/co.
Still to this day they deal with to much heat generation from even the XBox processor. They did a rush job for MS. IF IBM would of did more research on the G5 prior to the XBox they would of had a lower consumption and heat output chip. We might of even had a G5 laptop processor.
would of thought
would of did
would of had
You should invent a name for that new grammar rule of yours.
Edited 2006-02-07 15:10
would of thought
would of did
would of had
You should invent a name for that new grammar rule of yours.
Keep on the subject, Maybe you should be at http://www.grammernews.com since you “seem” so keen on writing, Nimble…lol
there’s a ( or two) reason(s) why IBM left the
desktop market and sold the desktop division to Lenovo.
There’s hardly any money ( or profit margins )
left in the PCs… And focused on
their bread-and-butter: business servers.
Same for Motorola/FreeScale. But for embedded markets
like cell phones and such.
Two pretty different markets for Apple unfortunately.
=========================================
Another reason is because IBM’s OS/2, Pink/Taligent
and/or Workplace OS lost the desktop
way back in middle 1990s. They were pushing for
the desktop with the PowerSeries ( I got one ;> )
Well, Apple choiche is good for those reasons:
– As Apple experienced in the past, a parthner for a custom solution may be not capable or not interested to deliver improvements on products made for Apple, on the other side chosing a mainstream CPU with a vital market and many cuncurrent vendors likely guarantee Apple that viable parthners will not loose interest in developing newer products; that, today, leave only x86 as a viable choiche.
– In the x86 (and compatible) market, VIA and Transmeta delivers not enough powerful CPUs for intended Apple product targhet, so remains AMD and Intel
AMD leads in desktop CPU power and tdp (and deploy 64 bit CPUs), but Intel leads in mobile market (for power and tdp) and has a roadmap for multicore mobile CPU some months ahead AMD one; since Apple is interested in mobile or small form factor and silent PCs, Intel mobile CPUs are the best solution at the moment.
Moreover, Intel produce good, stable chipsets for it’s own CPU, that *simplify* the task for Apple to assemblate a rock solid hardware (ok, that can be done also with AMD CPUs and good mobos, but a big producer would prefer to have to deal with as less as possible mobo models and subreleases, documented the best as possible, and that are Intel’s mobos).
Moreover (2), Intel is bigger than AMD and Apple would not risk product shortage if sales are awesome or if the producer experiences difficulties.
Moreover (3), Intel is a wider range producer than AMD and can offer good prices on many interesting products to Apple in excange of the parthnership.
That reasons counterbalance for Apple’s need the fact that AMD is in fact ahead of Intel in providing a state of art x86 CPU with integrated memory controller and mature x86-64 implementation.
John C. Dvorak claims in a column to be published soon in PC Magzine that Apple might be planning on ditching OS X for the new Intel-based Macs and moving to Windows.
One of the reasons for this is the recent Adobe announcement that they won’t have an intel version of their major apps until the next release cycle, which probably won’t be until 2007.
If you believe that then I have a bridge to sell you in Brooklyn.
He did predict the switch to Intel 18 months ago. You probably thought that idea was ludicrous at the time as well…
You take Dvorak with you to Vegas and really clean up, right?
The Mac is better with Intel, period!
I don’t know where “failing to live up to the hype” comes from, but checking out the intel CD 2.0GHZ iMac and DP 2.0GHZ G5, the iMac completely blows it out of the water.
The G5 couldn’t even play 1080p without dropping frames with nothing else running, while iMac was able to both play 1080p full speed, run iDVD in the background with motion preview on, launch safari in less than once bounce and still manage to resize all the windows fluidly. I would call that a huge leap in performance.
I’ve never wanted to buy a new mac just cause its new, but that iMac Core Duo makes me want to!