Should Apple Postpone the Power Mac Replacement?
May 26th, 2006This may seem like a radical idea, but the conventional wisdom, such as it is, states that Apple will announce an Intel-based professional desktop Mac during the WWDC in August. That is, along with Leopard and perhaps a second generation Intel-based model or two. This would complete the transition to Intel, assuming the Xserve is part of the program.
But hear me out: While this may be a good idea from the politically-correct standpoint, there are issues from the practical standpoint that may, to some, argue against such a speedy product release. In a case of the chicken and egg syndrome, not all software creative people need will be ready. While I can’t speak to 3D software, such as Autodesk’s Maya, it is certain you won’t see Adobe Creative Suite 3 until the second quarter of 2007. That news comes straight from Adobe, and it doesn’t seem likely they’ll suddenly change their tune and have it available six months earlier. Only Apple manages to release products ahead of schedule.
Now if your main software requirements begin and end with Apple’s Final Cut Studio or other Universal applications, this won’t be a serious matter. I’m certain that the Power Mac replacement, which we all expect to bear the Mac Pro moniker, will be an outstanding performer. No doubt it’ll have processor power sufficient for Apple to boast that it will be two to four times faster than the model it replaces.
What this means is that emulated PowerPC applications may run no faster than they do, today, on a Power Mac G5 Quad. So is that any incentive to pay a bundle for an upgrade that won’t enhance your productivity, or only provides the potential for that enhancement? Would you pay three grand for one, and maybe a thousand more for gobs of memory, knowing full well the promised speed enhancements won’t affect you for another nine months, if then?
What if Adobe doesn’t get is updated Creative Suite ready before the end of that second quarter, say June of 2007? This isn’t unlikely, because software development is a highly uncertain process. Adobe has a huge job ahead of it, and things can go wrong. What if the date slips to the summer of that year?
How would that affect Apple’s sales?
Now it’s quite true that the rest of the creative applications may have made the transition to Universal by then, even Microsoft Office for the Mac. Maybe Adobe will be the only significant holdout. More to the point, although Rosetta gets a bad rap for killing roughly 50% of your Mac’s performance potential, it’s quite possible things will get more efficient. Don’t forget that MacIntel users are working with a version of emulation software that’s basically at the 1.0 stage, or slightly beyond.
It isn’t out of the realm of possibility for Apple to announce a souped up version, Rosetta 2.0, which will be included in Mac OS 10.5 Leopard. Maybe it’ll deliver 60% or 70% of native performance. If true, performance of your older applications may actually be faster on the Mac Pro, even though still somewhat short of what a Universal version can do.
All right, I’m speculating here, but I do not think for a moment that Apple isn’t looking for ways to make its emulation environment perform faster, take better advantage of the advanced features of the new Intel chips it’ll be using in its next generation hardware.
So, even if you don’t get the full speed boost, you’ll get enough to justify the investment in a new professional Mac desktop. Another possibility, which I’m not affording a high probability rating, is that Adobe will release its Creative Suite 3 products in stages. From a marketing standpoint, Photoshop is mission critical, and it’s always possible it’ll get priority above the others, and ship several weeks earlier. It may present some marketing complications, because Adobe would be forced to sell it separately, with a coupon to get the rest of the updates at a later time. But that’s not something that hasn’t been done before in the software industry.
In the end what approach should Apple take? Frankly, they have no choice, in my opinion. Power Mac sales have stalled for years, and getting a super-powerful Intel-based version out by late summer could goose sales substantially, even if all the key creative software isn’t quite ready. That, along with an Intel-based Xserve, will mean the transition to a new processor will be done months and months ahead of the original schedule.
It will also provide a tough incentive for Adobe to push harder to get its work done. Maybe a larger base of waiting customers will help them jump through some extra hoops. Besides, even if some key productivity software must run in emulation for a while longer, things may still be pretty good. Don’t dismiss Rosetta until you’ve tried it, as I have. It’s really not as bad as some will have you believe.
Postpone the Mac Pro? Not on your life!
Print This Post
There’s another possibility – that Apple could keep some or all of the current G5 PowerMacs in production while producing Intel boxes. Remember what they did with the G4 tower – the last OS9 machine. they kept it around – quietly – for quite a while before dropping it completely. Maybe Apple will keep one or two of the PPC G5 configurations around for awhile.
If Apple really cared about their customers, they should keep one PowerPC Powermac and XServe alongside the Intel models at least until the end of 2007 ( in the same way it was made with the last G4 Powermac ) so that the work of those businesses dependent on non Universal apps and plugins is not disturbed. Even better, make a small speed bump on those PowerPC machines.
But Apple is so entertained with their own marketing that many of us doubt that they care about their corporate ( small and large ) and edu clients.
What I might hope, even expect, to see is a situation like the OS 9/OS X transition, where Apple switched their entire lineup to OS X-only booting — with the exception of one last-generation PowerMac kept around for people who had to use OS 9 software that wouldn’t run in Classic.
Indeed, “Rosetta 2.0” may be coming soon. Evidence pointing in this direction is “Boot Camp” and the general emulation history Apple has known, including Universal Binaries.
The earlier the better. FCP for Intel should speed things up considerably for pros. And it just gives a bigger push for stragglers to make programs. Look what happened to Quark when they were tardy, Damn near destroyed them as a company. Obviously they have learned their lesson and are releasing Intel Quark probably on stage at WWDC. I wouldn’t be surprised if Adobe is there showing a demo of CS3 as well. And I believe the pent up demand should sell quite a few of these boxes.
I think that the comment on Adobe being pressured to jump through some hoops to get CS3 out early may be very astute. It is amazing how quickly that a Mac only beta of Lightroom appeared after Aperture came out. Adobe is very competitive and is watching Apple closely. If Apple did another update to Aperture at the release of the Intel PowerMac they would not only sell lots of copies of Aperture, but Adobe would pull a groin muscle meeting the challenge. I can’t imagine they would only see Lightroom as the answer. They really don’t want to lose Photoshop sales to either Lightroom or Aperture.
I think it makes sense for Apple to wait for Ethel Merman, Conroy, Woodstock, whatever the new Intel chips will be called, before they start replacing the G5 towers and to wait for 64-bits for the X-Serve if not the pro towers.
One more round of faster G5 Quads might not make them a ton of money, but would make a few folks happy and give them something to keep in the line for a while at least until Adobe gets CS3 done.
Heck, I wish Apple would make a powerbook with the Freescale dual core 8641D that has an onboard memory controller.
I can’t wait to see the specs of this chip.
– Mark
Agreed. The Power Mac G5s should be sold along with the new Mac Pros.
Well, it’s a standard “chicken-or-the-egg” question.
I don’t think Apple will ditch the PowerMacs at WWDC like they did with the iBooks. But they might revamp the website and stick the PowerMacs in a corner someplace (so the website will prominently show all Intel machines, but there’ll be a link down at the bottom saying, “Still want a PowerMac? Click here”).
One interesting question will be the speed of the new Intel “Macintosh Pros” (I like that better than “Mac Pro”) compared with the PowerMac G5 Quad. It might be a good idea to cache Apple’s benchmark pages before WWDC just so we can do a before and after comparison… :^)
I see a few beat me to my comment.
At least one G5 should remain in production until Photoshop is Universal. Seems like common sense, I hope Apple agrees.
Prob. keep the Quad in production and replace the other G5s with the new Intel Mac Pros.
just my $0.02
Oh, Peter, it looks like the iBook G4 is still available in EDU channels.
I think you’re right on the money. Apple needs to jump start their pro sales, but they’ve done the damage to themselves.
They always price the towers well above other models even when their performance starts to lag. In a repeat of the 604 to G3 transition in 1997 their mid-price machines are faster than their top end ones. The situation would be embarrasing for anyone other than arrogant Apple. To make matters even worse, the iMac supports Aperture whereas even the G5 Quad doesn’t because of the crap video card it ships with. When Aperture 1.1 came out Apple should have admitted their “Pro” machines aren’t and improved the stock video card to the 7800GT (without raising prices) while at the same time offering existing owners a chance to buy the 7800. Instead they offer nothing. Way to screw the cash cow Steve!
Apple has had years to get Rosetta working, I wouldn’t count on getting more than a few extra percentage points from further optimization. I firmly expect a G5 to remain on the price list until the end of the year to satisfy the Adobe users out there who don’t want to pay top dollar for emulated speed.
I also expect the XServe to remain G5 powered until a 64-bit Intel replacement and compatible version of OS X Server to be ready.
I think Apple should skip the Intel desktop chips, which even with Woodcrest will
not be that great, and go with dual-core and quad-core Opterons instead. In the
desktop and server space, AMD’s lead is commanding, with better architectures.
Apple should be free to pick the best chips from both AMD and Intel, not tied
solely to Intel just for nostalgia’s sake.
Then keep a quad-core G5 in production for another year just to backstop
customers who need apps like Adobe’s that are still three or four quarters
out.
If the new Intel boxes are as fast as they say than yes, people should buy them and wait for CS3. I mean it might be worth working in emuation mode for a while. But keeping G5s in production is a good idea as well
While Intel chips are a good fit for the consumer and laptop models, Apple should continue Universal Binary support and offer the most powerful processors for the Pro and Server machines, whether Intel/AMD, Power (check out PA Semi), Sparc, MIPS, Cell or the unknown.
Microsloth dropped the ball on multi-architeccture WinNT. Apple should tap into this opprortunity.
No way should Apple delay. Productivity — even for web and graphics professionals — doesn’t begin and end with Adobe apps. Even aside from Apple’s pro apps, the speed of web browsing, file searches, and general interaction with the machine is a huge part of the equation. To seal the deal, Parallels offers a cheap and wicked fast way for Mac professionals to do testing on Windows platforms, which is critical for web and multimedia work.
Considering the fact that a Mac Mini can already outperform my dual G5 in processor-intensive tasks, whatever Apple comes up with for the Mac Pro is going to blow it clear out of the water.