Submit Hint Search The Forums LinksStatsPollsHeadlinesRSS
14,000 hints and counting!

How do you feel about Apple's announced switch to Intel CPUs?

1/1: How do you feel about Apple's announced switch to Intel CPUs?

10 - Excellent, perfect thing to do! 597 (13.42%)
9 171 (3.85%)
8 381 (8.57%)
7 - Good decision in the long run 1,262 (28.38%)
6 309 (6.95%)
5 - As of now, I'm not sure 1,105 (24.85%)
4 132 (2.97%)
3 - Pretty poor decision 188 (4.23%)
2 44 (0.99%)
1 26 (0.58%)
0 - The worst Apple move ever! 232 (5.22%)
Other polls | 4,448 votes | 33 comments

How do you feel about Apple's announced switch to Intel CPUs? | 33 comments | Create New Account
Click here to return to the 'How do you feel about Apple's announced switch to Intel CPUs?' hint
The following comments are owned by whoever posted them. This site is not responsible for what they say.
Buy now or wait?
Authored by: pete on Jun 07, '05 10:44:40AM

I was looking at buying a 20-inch iMac G5 or fully decked out mini-Mac this summer.

Now that Steve wants to change to Intel, I'm not so sure I want to buy now. My old iMac has done me well for the past 6 years and I want another that will last as long, while being supported in some way.

Maybe I'll save my money, put up with this slow beast and see what happens next year - or maybe buy it now so that all my software will work OK.

Or maybe... I stick with a PDA and the benevolence of friends!

I spent a small fortune upgrading software from System 9 to OS X - I don't want to have to do that all over again next year with Intel. Being a home user, I can't afford BIG changes every year or so like governments & companies with money to burn!

[ Reply to This | # ]
Buy now or wait?
Authored by: Wayne_S on Jun 09, '05 11:21:31AM

The standard hardware maxims still hold true even with Apple equipment. The best way to see a performance increase is usually to increase the amount of RAM available. If you aren't maxed out, I would do that before buying a new machine. RAM is relatively inexpensive and hopefully the performance increase will give you enough time to further evaluate things when Apple decides to release some more detailed information - like, for example, exactly WHICH Intel processor(s) will be used.

Every rule has an exception. Especially this one.

[ Reply to This | # ]
Buy now or wait?
Authored by: dasil003 on Jun 15, '05 12:48:35PM

Personally I think you'll run into more compatibility problems if you buy one of the first Intel Macs (which will likely be laptops anyway).

The way I see it is, it will take some time for apps to be recompiled, and for developers working outside of XCode or on highly optimized apps, the wait could be even longer. The PPC stuff works now and I imagine everything new will work on on PPC AT LEAST until Apple ships the last G5, which I see being at least 2.5 years from now. After that new versions will likely continue to work on PPC for most software for several more years (if the promises of XCode fat binaries hold true).

I'm actually planning on buying one of the last G5s (just ordered a new PowerBook G4 yesterday) because I think it's such an amazing architecture that I want to own one while I still can. Down the road in the distant future when it's no longer supported at all by Apple's OS, I can see myself putting Linux on it to get even more mileage. Then 50 years from now my kids will inherit it and sell it on e-bay for $50,000 (not adjusted for inflation).

Of course, I could be totally wrong... but when have computers ever lasted forever anyway?

[ Reply to This | # ]
Buy now or wait?
Authored by: ebow on Jun 09, '05 12:07:31PM
I'm in almost exactly the same boat. My G3 iMac is nearly 5 years old, and while it's mostly good enough for what I do with it, I've been promising myself a new computer for a while now. My take on the situation is:
  • If I got a new iMac and the Intel transition wasn't happening, it would last me 4-5 years.
  • Even with the transition, I'll probably feel compelled to make it last as long.
  • The apps I want to use (mainly iApps, Office, mail, and web) might stop being updated for PPC in 2.5 years or so, which could leave me with "stale" apps for 1.5 - 2.5 years.
  • But even if that's the case I probably won't mind, judging by my current habits (e.g. I find Office v.X and iPhoto 2 good enough to put up with until I get a new computer).
So if you're in a situation similar to mine, a new iMac might just be the thing to coast you well past the transition without sacraficing too much. Hmm, I could even treat myself to one of the inevitably sweet notebooks in 3 years or so if the iMac is really showing limitations.

And yet, part of me thinks getting a Mac mini would be the smarter move right now, to last me maybe 3 years until the 2nd revisions of Intel-based Macs are available. Tricky call.

[ Reply to This | # ]

Buy now or wait?
Authored by: lemkebeth on Jun 11, '05 03:06:46AM

To be honest, I'll be upgrading at some point. I have an old Grape iMac DV/400 running 10.3.9.

In any case the transition will be like the 68K to PowerPC transition in every respect. It was several years before the OS for the 68K machines was stopped being updated after the release of the first PowerPC Mac. In fact the last version of the classic OS to run on 68K machines was 8.1.

Me, I'd buy a G5 at a nice discount thank you very much.

[ Reply to This | # ]
This transition is no big deal for users
Authored by: cfoster on Jun 07, '05 11:32:21AM

There are very few people who need to be concerned about this transition. (e.g., scientists with limited budgets and a large investment in custom software AND a need for new hardware in 2007.) 99.9% of users will never know there's an Intel chip instead of a PowerPC chip inside their box. After all, Steve fooled 3800 software engineers at the WWDC -- how would the average user know the difference?

"But what if my G5 is made obsolete by a P6 instead of being made obsolete by a G6?!"

What if it is? All computer hardware is 'made obsolete' by the next generation of hardware. It will still run. It will still have been a good investment for you at the time you bought it, and continue to be a good investment long after Apple has switched processors.

"Maybe I should wait a year or two to buy an Intel Mac instead of buying the computer I need today?!"

But then it will be first-generation hardware and you won't want to risk hard earned dollars on new, untested technology -- better wait another year until the hardware is proven. THREE YEARS is a long time to lose productivity for a couple of extra processors cycles.

Those that want all the reliability they have come to know and love with Macs and are concerned about the transition will actually want to buy SOONER, not later. More conservative consumers will want to to stick with "tried and true" PowerPC Macs for as long as they can. Of course, the Intel Macs have been in testing for the last FIVE YEARS, so they aren't exactly experimental. If Steve Jobs is doing his WWDC Keynote on one, you can bet it's pretty solid.

The only people that really need to be concerned are the developers (I'm one), but the road map for us looks pretty straightforward as well. This is a ridiculously well planned transition. In fact, we don't need to change our code at all if it's not a processor intensive app. 'Rosetta' takes care of all existing apps without any changes, but they'll get a small speed bump when created as a universal binary.

So, I think (almost) everyone can relax -- unless you are a developer with a processor-intensive app, or have a situation similar to the scientist scenario above, you won't even know it's happening.

[ Reply to This | # ]
Only 32 bit?
Authored by: qubex on Jun 07, '05 12:39:39PM

What worries me is that Xcode 2.1 (the new version that supports Intel) appears to only support 32-bit IA-32. It doesn't support new 64-bit chips that use the AMD64/EM64T architecture. The Intel Developer machine Apple is now selling at $999 is a Pentium 4 box.

So, after the jump from 32-bit G4 to 64-bit G5, they're jumping back again?! That really doesn't make any sense!

[ Reply to This | # ]
Only 32 bit?
Authored by: yellow on Jun 07, '05 01:17:03PM

I suspect when they actually come out with a REAL MacIntels, some will have an Intel 64-bit chips. In the meantime, I suspect the laptop/mini/tablet markets will be the first to see 32-bit Intels with a "player to be named later" option.

[ Reply to This | # ]
Only 32 bit?
Authored by: qubex on Jun 08, '05 04:46:11AM

I'm sure Adobe will just love this.

Port Photoshop from Motorola 68K to PPC.
Port Photoshop from OS9 to OSX.
Add support for VelocityEngine.
Add support for 64-bit G5.
Take two steps back, removing dependencies on AltiVec/VelocityEngine and 64-bit.
Port Photoshop to 32-bit Intel x86, recode AltiVec functions as SSE2.
Port Photoshop to 64-bit Intel EM64T.

Then what? They may as well adapt it to run on the proverbial kitchen sink.

As a developer that finds himself often in the mires of PPC32/PPC64 assembler optimising things to use VelocityEngine etc. (mathematical models, huge sparse matrices, floating-point arithmetic to make your eyes bleed), I am mighty pissed. I'm used to having a big-endinan CPU and word-alignment. From now on, the stack will look upside-down and inside-out to me. I'll waste ages calculating fixed offsets. Intel CPUs have far fewer registers than PPC makes available and that means performance penalties aplenty.

Apart from anything else, the G5 still beats the socks off any processor Intel has to offer in terms of floating point arithmetic. I wouldn't be quite so pissed if they were switching to AMD64. Going to Intel x86 without 64-bit extensions is a huge step backwards.

If Apple is going to make x86-64 available later, after we have all slaved over a hot stove to get our stuff running under x86-32, I'll be furious. They could at least show us a modicum of respect and make us do all this work only once.

Quite honestly, I'm actually thinking of switching my supported platform to Linux/PPC64, since that way I could keep all my low-level algorithmic optimisations and I'd "only" have to port the high-level code that interacts with the system through all the relevant APIs. And the GUI. And the filesystem support. And the networking code. And come to think of it, everything else.

[ Reply to This | # ]
Only 32 bit?
Authored by: lemkebeth on Jun 11, '05 03:17:57AM

Thats the problem with using assembly and why languages like C, C++, and Objective-C were invented.

My advice is to keep things as processor independent as possible. Thats not to say that optimizations can't be added just that they should be self contained.

[ Reply to This | # ]
Only 32 bit?
Authored by: qubex on Jun 14, '05 12:36:28AM

For years they have been telling us that PowerPC is superior, and that it makes sense to spend ages optimising our applications to use AltiVec.

Now they turn around and toss the baby out with the bathwater.

[ Reply to This | # ]
Only 32 bit?
Authored by: boredzo on Jun 14, '05 05:41:22AM
  • Port Photoshop to 32-bit Intel x86, recode AltiVec functions as SSE2.

since Photoshop already runs on x86 (in Windows), I think it's safe to assume that this work has already been done.

[ Reply to This | # ]
Intel Inside/IBM out of picture
Authored by: gsgm on Jun 07, '05 03:27:36PM

I think desicon is more a "IBM wouldn't/couldn't" keep upgrading the PPC chip and produce enough quantity to keep Apple happy. We all know the benefits of RISC vs. CISC, and I think if IBM would have gotten a 3+Ghz PPC this wouldn't have happened.

But now that it has, what do I expect?
I want the same look and feel that I get now. I want consistantly easy applications that "just work" without hassles. I want to say "that's cool!" when I turn my Mac on. Since all of this is software, I truly believe Apple will pull this one off.

But what are my fears? For one, the Pentium chip architecture's uses of IRQ. (Mostly conflicts and a limit of the number of devices) How much more "extras" can Intel cram on the chip? (Is there really an improvement besides size if it's the same pentium, but with dual cores?)

But I also wonder what else is up Steve's sleeve. Since all of the recent Apple applications have been developed to work on a Pentium chip, does that mean I can by iLife and install on my Windows machine? (Probably not... why buy a Mac if I can do that on any machine) But what if OS X could run on a Dell? (Not yet for the same reason as above) If X-Code can handle cross compiling for 2 processors, why not 3?

[ Reply to This | # ]
Intel Inside/IBM out of picture
Authored by: Shawn Parr on Jun 07, '05 04:22:31PM

Just because you don't have to deal with IRQ's doesn't mean they don't exist on the Mac.

IRQ's are used for different parts of hardware (and software) to talk to each other.

On the Mac Open Firmware and the Mac OS work together to make this all transparent. I would expect that under the Mac with Intel design they would still be able to handle this. Even though OF will not be used according to the documents so far released. Even XP does a pretty good job of you not having to ever set an IRQ, some people may run across it, but most do not.

Also a lot of people don't seem to understand the timeframe we are looking at. Within 12 months even portable Intel chips could all be 64 bit. We are also probably not really looking at P4 for actual released machines, but something based on Pentium M and Pentium D designs. PD is 64 bit, and Pentium M is amazingly efficient for its power use and low heat output (a 2Ghz easily outperforms a 3.6Ghz P4 in Intel's own demos).

Yes I know that the current developer kit is a P4, but that is the _current_ kit, not what will ship in 12-18 months. That is a couple revisions of machines in computer lifetime.

I am interested to see what will happen with the PM and PD lines now, especially in the way of vector optimization, as that is the one area where the PPC truly rules. Other areas (memory bandwidth, etc) could easily catch up in that time frame.

[ Reply to This | # ]
Intel Inside/IBM out of picture
Authored by: chris_on_hints on Jun 08, '05 06:51:23AM

Steve even said in the description of the P4 boxes that he was selling to the developers that they were only to aid in the transition, and would be nothing like what will be the first IntelMac...

... I take that to mean that the IntelMacs will NOT be running pentiums, which are far hotter and run at higher clock speed than the newer chips Intel makes, and the AMD ones.

...I am hoping to see a nice 64-bit Intel processor in the Mac mini and then in the powerbooks.

[ Reply to This | # ]
Intel Inside/IBM out of picture
Authored by: on Jun 07, '05 05:08:14PM

I think it has to do with 3 things:

1. Portables. There is no G5 laptop on the horizon - ever. Yet laptops are a huge part of Apple's business. The currnent G4 laptops are looking pretty sad beside a Wintel box

2. Windows, un*x and OS X GUI all on the same box - nuff said.

3. Future products. The iPod is just the beginning. Remember Steve J.'s vision is for Apple to be the next Sony (not to replace but to be the next). Imagine, if you will, consumer electronics products built on a stipped down OS X running on embedded CPUs.

----- Totally wild speculation on -----

The iPod revenue has also given Apple something it has never had before -- freedom from total dependence on Macintosh hardware revenue. This may tempt Apple to once again try OS licensing (remember the last disaster occurred as a result of hardware revenue lost to OS 7 licencees) secure in the knowledge that the iPod revenue (and even its Application software like iLife being sold as an add on for third party OS licensees) is safe and could cushion things during the licensing transition period. Licensing OS X is a MUCH easier sell if it runs on Intel.

[ Reply to This | # ]
Intel Inside/IBM out of picture
Authored by: chris_on_hints on Jun 08, '05 07:02:41AM

Nice thinking.

1) - absolutely agree. Intel have a range of low power chips for laptops that run cooler and are more efficient than the Pentiums (which are as hot as a G5)

2) - yes, but i wouldnt put XP (or longhorn) on my IntelMac. EVER! But I would like to use Wine to run Windows apps, without the need for windows itself... Maybe developers will stop porting the PPC version to x86 and just make sure their windows x86 binaries will run under Wine... Might be less effort for some of them.

3) - yeah - bring on those future products. Jobs is a smart cookie, and from the "every version of OSX has been compiled for Intel since 10.0" bombshell, it is clear that he is thinking ahead and has a good stretegy.

My speculation would be to suggest that maybe the on-chip DRM that Intel have was the deciding factor. Maybe that would be how Apple could lock OSX down to their machines, and give further security to the iTunes music store components... especially if they are wanting to move into film.

[ Reply to This | # ]
It all depends...
Authored by: gshenaut on Jun 07, '05 06:23:05PM

If the supporting libraries & APIs are as identical as they can be given the different endian tribes of the CPUs, and if the next few distributions use "fat" Mach-O's to be multi-lingual, and applications developers do the same, then the transition should be pretty painless--probably most people won't even have to know which CPU they have.

Greg Shenaut

[ Reply to This | # ]
Authored by: qubex on Jun 08, '05 04:49:38AM

This kind of makes Xgrid irrelevant, unless Xgrid has support for heterogeneous networks and fat binaries (and last time I checked, it didn't). Now you won't be able to offload compute-load onto the LAN unless all the workstations have the same architecture processor.

This is awful.

[ Reply to This | # ]
Authored by: chris_on_hints on Jun 08, '05 07:06:40AM

But if a PPC can build binaries to run on x86, why couldnt you have a network of a mixture of chips doing a Xgrid compile?

Surely if the OSX is the same, and the Xcode is the same (ie the same code, just compiled for different chips) then all we need is an update to Xgrid to handle the x86 'option'?

[ Reply to This | # ]
Authored by: Shawn Parr on Jun 08, '05 04:35:28PM

Xgrid doesn't need to support fat binaries. Fat binaries need to have support for Xgrid.

Future upgrades and new Xgrid capable binaries will either be fat, or will be stuck on one platform or the other, just like any other app.

Xgrid could be updated to check the platform the app is built for (or if it is fat) and make sure it only runs on one platform if it is "thin" however you should segment your grid if you are going to be doing this.

[ Reply to This | # ]
What if...
Authored by: trex67 on Jun 08, '05 10:42:38PM

What if IBM comes to Apple in a year or two and has something that blows Intel out of the water? Will Apple stick with the dual platform or say no thanks? If options are good, as Jobs says, why not use PPC as an option for the high end users that prefer it?

I will not "upgrade" if a Pentium can't handle the workload that my almost two-year-old 2X2.0 GHz first gen G5 PowerMac handles with ease, namely, nearly unlimited VST and MAS plug-ins in MOTU Digital Performer.

[ Reply to This | # ]
What if...
Authored by: FenrisUlf on Jun 23, '05 09:11:20PM

Trust me, it will. Pentium D (dual core) chips can handle it now. When the transition starts, Intel chips will handle the load just as well as PPC. We don't have to worry about that, at least.

I look forward to buying a new Mac mini with an Intel chip inside, and I also think that I will buy myself a nice portable mac too.. to compliment my Dual 2.0 and my G4-733.

I think we'll be surprised how well they will work. ;)

Who are you that walk across the graves of giants at this late hour?

[ Reply to This | # ]
No Intel!
Authored by: KenaiTheMacFan on Jun 11, '05 07:17:21PM

anyone who voted 6 or higher is a windoze user


[ Reply to This | # ]
No Intel!
Authored by: gaudior on Jun 13, '05 10:07:11AM

You really should get out in the world. Intel <> Windows.

[ Reply to This | # ]
No Intel!
Authored by: Githon on Jun 22, '05 08:42:25PM

That means as of now, 2,080 "Windoze" users have voted in the poll, but only 1,346 Mac users.

Isn't your Wi-Fi reception screwed up by your tinfoil hat?

[ Reply to This | # ]
Problem with upgrading soon
Authored by: Panjandrum on Jun 16, '05 04:47:21PM

I think that those of us who are "worried" about upgrading our machines before the new Intel-based units become available probably remember the '040 to PPC switch. It was only a very short time, much less than a year, before applications started being written for PPC only, even though in many cases the 68k machines would have had plenty of power to run the program. This didn't happen for major products like Photoshop for quite some time, but it did happen with a lot of programs, including a substantial portion of the shareware scene. Was this Apple's fault? Not really, it was the fault of the people writing the software, but the result was the same; lots of hosed 68k owners.

I, for one, have decided to put off any purchase of Apple hardware until I can buy and Intel-based box. I require my machines to have a very long usability arc, and just can't afford to be out a major investment. I'll put one last G4 CPU upgrade in my sawtooth to make it through the next couple years, and see how things turn out with the whole Intel switch.

On another note. Aren't the chips Steve just talked about 32-bit and not even multiple CPU-capable? That's would really suck. I do a lot of heavy-multitasking on my system and am not looking forward to a machine with a single-cpu. I'm hoping I'm wrong on this point...

David Butler

[ Reply to This | # ]
only matter is Intel monopoly...
Authored by: Herve5 on Jun 19, '05 04:07:31AM

I believe the switch is good for Apple, the only sour taste being that while OSX *will* stay a good alternative, on the CPU side OTOH there will be a monopoly. Intel.

[ Reply to This | # ]
only matter is Intel monopoly...
Authored by: sveinbjornt on Jun 28, '05 04:23:33PM

That's nonsense. AMD competes quite aggressively with Intel -- in fact, they have the fastest x86 chips out there at the moment. I suspect that Apple went to Intel because AMD has had supply problems. But I'm certain that Apple will play these two off against each other, offer successively better deals...

[ Reply to This | # ]
pool structure biases to 'commented' level?
Authored by: Herve5 on Jun 19, '05 04:11:58AM

While this is quite off-topic, it really strikes me that those poll's levels that are associated to textual comments (whether they are positive or negative) attract *much* more votes than the other ones. I wonder whether one could deduce some "uncertainty bars" on the poll levels...


[ Reply to This | # ]
pool structure biases to 'commented' level?
Authored by: Han Solo on Jun 20, '05 08:59:49PM
It is certainly the case that the labeled responses attract more attention -- but also no surprise: this type of "framing" often affects surveys and other social experiments. However, these comments are, IMO, necessary in a survey like this: what does "6" mean exactly? Without some "guidance" or framing, one person's 3 or 4 may be another person's 7 or 8... and thus the aggregate results would be completely uninterpretable.

Error bars don't seem to be appropriate here: in no way can this survey be considered a random sample, so there is no statistically valid way to construct such error bands. Best one can do is look at the histogram and draw one's own conclusions. The median and modal responses both are 7 (53.2% of the mass, as of right now, was at 7 or higher), so most of the participants don't seem overly concerned. Notice that this is not how most the comments read, so it says something about people's willingness to click a button versus type in a text box. (Yes, one should always expect more extreme responses in a text box! ;) )

[ Reply to This | # ]

This is 2005, people.
Authored by: maximxygo on Jun 24, '05 09:40:37PM

Seriously. What do you see when you turn on your Mac? The hardware? No, you see software. It's Apple's software that makes the Mac experience so wonderful. And when was the last time any major platform underwent a hardware architecture change? Not an OS change, but architecture. That was 1994, with the transition to PowerPC on the Mac. That transition went smoothly, VERY smoothly. And that was 1994. This is 2005. I really don't think the processor is going to affect the end user, hardly if at all.

[ Reply to This | # ]
monoculture vs. multiculture
Authored by: dc_rees on Jun 27, '05 11:30:39PM

Finally announcing the x86 port was good. Announcing hardware for Mac OS X86 was good.

Deciding to drop PowerPC was evil.

We all know the monoculture argument. Nothing has repealed that effect. Mac on x86 will be a larger target for malware of all ilk.

There are valid reasons for a customer choosing PowerPC, even if it might not be quite as fast as iNTEL (which won't be as fast as AMD, anyway). Current PowerPCs are sufficient for most desktop and notebook purposes, and, unless IBM is planning to complete cut all PowerPC except embedded and game (which is clearly not the case), performance will improve at a reasonable rate, possibly leapfrogging iNTEL from time to time.

Apple now has sufficient financial resources and a large enough customer base, they can and should give the customer a choice on CPUs.

This should not be an either-or proposition.

Say yes to CPU multiculture

[ Reply to This | # ]