[FireDancer]

1. Apple made a mistake in choosing Intel because everyone knows they could've and should've: 1) used the abundant IBM PPCs in the pipeline and/or 2) just used the Cell processor in it's Macintosh line.

I think it's safe to say IBM has become unwilling to design and fab a chip that is essentially a custom part for Apple. IBM for whatever reason has not embraced the PPC970 in it's own products. I whole heartedly think that both Apple and IBM thought the G5 was going to ramp in speed without a problem and that the move to 90nm would go with out a problem. Now faced with "power density" issues requiring additional cooling on Apple's part, poor yields, and IBM's general move away from anything desktop - it probably didn't make any business sense to continue to develop the additional CPUs Apple would require for their product line. Unless IBM was going to aggressively use it's own CPUs, Apple's 2% market-share is not going to cover the cost of developing and fabbing the PPC970. I also think it's safe to say that in it's current form the G5 is dead @ 2.7GHz.

As for the Cell Processor - Everything I've read about the cell processor has said that due to it very poor out-of-order execution it is not at all suitable for a desktop CPU with current software. Without completely rewriting significant amount of code, this is not possible. Furthermore, the problems with rewriting and optimizing code for the Cell processor would have been an order of magnitude more difficult for everyone - much more difficult than moving to x86.

2. Apple will orchestrate and finish the largest shipment/upgrade of Intel based Macs in the history of computers by replacing the 20-40 million actively used PPC Macs in exactly 2 years. This will cause all PPC Macs to become obsolete in exactly 2 years.

With 20-40 million active mac users and Apple selling about 3.5 million macs per year, at the current rate it will take 11 years to completely refresh the current mac community with Intel products. Considering Apple doesn't plan to ship an all Intel product line until 2007, it would take more than 11 years. Even after Apple begins shipping nothing but Intel-based computers, it will be years before the Intel based Macs become a majority of the actively used macs (at least 4-6 years from now).

3. Because of Apple's incredible replacement program, my PPC Mac will be useless in two years because all software developers will start shipping Intel only versions in 2007 to the 40 million new Intel based Macs sold on 1/1/2007.

Conservatively, we are looking at PPC being the predominant Mac CPU in actively used Macs for the next 4-6 years. In other words if software developers want to sell software to the MAJORITY of Mac customers, they will have to continue PPC software development for at least 4-6 years and it will probably be profitable to develop software for longer than that.

My 4-6 year time period before the majority of PPCs are phased out is very conservative in my opinion.

So basically, everyone needs to chill! Am I dead on, probably not but there is such a large group of OMFG this is the end of Apple because Intel is evil and so is Steve Jobs because he very clearly didn't think this through and they should've just used a cell processor because my new Mac is worthless and so I'll show Apple by never buying another Mac again - long live the switching back to Wintel movement people

[Luca]

You are the winner!

[FireDancer]

Better than being the whiner...or the weiner for that matter.

[Franz Josef]

Whereas being a Wiener .. that's a differnet matter

[FireDancer]

yea yea, that too Dam the "i" before "e" except after "c" rule

[ezkcdude]

I especially agree with your second point, FireDancer. Most Mac owners have no clue what processor they have. In my lab we have 8 Macs. The day of the announcement I went around telling everyone (like Chicken Little!), and the responses were mainly like, "Oh, really? Huh." I don't even think it registered. So, I'm guessing that maybe 1 in 5 Mac owners actually give a crap about the type of processor used, but most could care less. Those folks won't be buying new Macs because the processor changes, they will buy new Macs regardless of the processor, because they like using Mac OS X.

[Banana]

Everyone, a toast for Firedancer!

[Luca]

Yeah I'm going to need to link everyone to this thread. It's amazing that people think what they do because the three points FireDancer raised are all very simple if you watch the WWDC presentation. I think a lot of people simply didn't watch it, and then hear "Complete transition by June 2007" and think "OH NOES my Mac won't work then!"

[FireDancer]

Did Apple make an expensive mistake by announcing a switch to Intel, and a year-long wait for the first hardware? According to a survey of Register readers, Apple faces a sales chill that could cost it billions of dollars. Hundreds of emails have poured in - one every minute over the past thirty hours - representing an impressive cross section of our more than 3 million monthly readers across the globe.

http://www.theregister.co.uk/2005/06...e_sales_chill/

3 million people who aren't going to get the satisfaction of owning a PPC Mac, with support for years to come, because they're too stupid to think the situation through.

Trying to resist the RDF, I think Steve Jobs was right when he said something like, "it's a big deal but it's not that big" on CNBC after the keynote.

[Banana]

This is a bit alarming.

After all, Perception is reality.

Will this "perceived nonsupport" hurt Apple in long run?

I do hope the marketing dept at One Infinity Loop has some good eye candy....

[Kickaha]

Originally Posted by Banana I do hope the marketing dept at One Infinity Loop has some good eye candy....

iBabes!

[Banana]

Originally Posted by Kickaha iBabes!

This?

[FireDancer]

I do hope the marketing dept at One Infinity Loop has some good eye candy....

Forgot where I said it yesterday but anyway.......I think Apple needs to break its tradition of silence on this topic and keep a steady stream of positive information flowing out to the masses....there certainly are enough "Dooms Day" people out there that Apple needs to regulate this situation carefully.

Apple has a habit of making great announcements and then disappearing for 3-6mo depending depending on the next Expo

Apple may be heading towards more openness with hardware when it switches to Intel because how the hell are they going to realistically hide big Intel hardware announcements? It has been easy to hide new hardware in the past with IBM and Moto but we'll all know what Intel has released and I think hardware updates will become more predictable too.

[Banana]

Yeah- that's one reassuring thing about Intel- It will mean more consistent update cycle and less wild speculation on where Apple will go next.

Is my prejudice against trade secret showing?

[naren]

Great post FireDancer, I'm wondering if I could post it on my blog? With accreditation of course...

[FireDancer]

Sure....not responsible for spelling mistakes, gross inaccuracies, subtle inaccuracies, or bad humor Let me know which blog.

[LudwigVan]

Originally Posted by FireDancer http://www.theregister.co.uk/2005/06...e_sales_chill/ 3 million people who aren't going to get the satisfaction of owning a PPC Mac, with support for years to come, because they're too stupid to think the situation through.

I finally got around to watching the WWDC keynote earlier today. And looking over a few of the emails sent into the Register and other posts elsewhere in the web, I'm surprised by the number of people who think they must wait a full year before getting any Mactel (or "Macintel," a new word I came across today) hardware or supported software.

Jobs stated that by this time next year, Apple will have new Intel-based computers shipping and some software vendors will have their products in customers' hands. (OK, so this is another prediction by Steve Jobs -- and we know how well some of those have played out in recent years. Still, I prefer to remain optimistic.)

It appears to me that Jobs is shooting for an earlier release for both hardware and software, whether at MWSF or some other event.

[morningstarrising]

Well without a "hot chip" or "vaporware chip" this prediction has a better chance to happen then 3ghz

But of course we'll keep an eye out on the "curse of the Apple processor"

[BlueRabbit]

Originally Posted by LudwigVan I finally got around to watching the WWDC keynote earlier today. And looking over a few of the emails sent into the Register and other posts elsewhere in the web, I'm surprised by the number of people who think they must wait a full year before getting any Mactel (or "Macintel," a new word I came across today) hardware or supported software. Jobs stated that by this time next year, Apple will have new Intel-based computers shipping and some software vendors will have their products in customers' hands. (OK, so this is another prediction by Steve Jobs -- and we know how well some of those have played out in recent years. Still, I prefer to remain optimistic.) It appears to me that Jobs is shooting for an earlier release for both hardware and software, whether at MWSF or some other event.

Well, some companies are already announcing Intel support, and the week isn't even over yet. So, most software should be ready by the end of the year, if not within the next few months.

Likewise, Steve said that part of Apple lineup would be partially Intel-ized by WWDC next year, i.e. the transition will be partially complete. So, I'm expecting new Intel-based hardware at MWSF, for sure.

[scratt]

Very succinct FireDancer.

Your analysis of the amount of PPC machines out there and the pick-up rate for new machines is very interesting and spot on, IMHO.

Worthy of a Daring Fireball article in fact.

The only caveat I would add is that as we progress through the changeover we may find that some features of new software are only available on the new platforms.
Example - Motion.

I know that was a graphics card issue. But the same may arise with architechture changes and forward looking software companies. But then that happens all the time so it's not a biggie.