Google
 


Subscribe to The Tech Night Owl Newsletter

*Required



Archive for November, 2006


The Mac Software Report: A Eulogy for PowerPC Applications

Friday, November 17th, 2006

This entire article may seem a tad premature, but I see the handwriting on the wall. You see, history is about to repeat itself.

Back in the mid-1990s, Apple ditched the aging 68K chip in favor of the PowerPC. When native applications first arrived, they were offered in “Fat Binary” versions, which meant that code for both chips was included. It didn’t take long, however, for companies to excise the 68K code, as more and more PowerPC Macs were sold. In a few years, Fat Binary versions were history.

Now in those days, it took quite some time for PowerPC software to appear at first, nothing like it is today, where, less than a year after the first MacIntel was released, there are well over 4,000 Universal applications. At the same time, less than year after the Intel switch began, many of its products are already in their second generation.

This accelerated release schedule is going to have some huge implications for those of you using older Macs. Perhaps the biggest salvo in what will be a growing trend was the release of the public beta of Adobe Soundbooth.

Contrary to what you might have expected, the application will be available for Windows and Intel-based Macs. PowerPC models need not apply. Of course, the final release probably won’t happen until some time next year.

So why is this happening? Well, Adobe talked about the work required to optimize a sound-editing application for specific hardware. Once it was done for Windows, it wasn’t so hard to port the application to Macs that used the same processor family. There would be extra costs and development time involved in producing a PowerPC version, which would reach a declining user base, so they decided not to do it.

But didn’t Steve Jobs tell us that making a Universal version of an application is a piece of cake, that all you had to do was select a couple of checkboxes in Xcode? Well, that may be true for some applications, but in most cases, a company has additional work to do to optimize performance for both processors and address compatibility issues. The job becomes all the more complicated if a developer has to first import all the code from a different programming environment, as Adobe and Microsoft had to do.

That, and the fact that they’re dealing with applications with millions of lines of code, explains, at least in part, why the Universal version of Adobe’s CS applications won’t appear until next spring and the new version of Office for the Mac won’t arrive till months later. It’s not that they are just sitting back and doing next to nothing, as some conspiracy theorists might want you to believe.
Now I am not about to suggest that the successors to these two application suites, which probably won’t appear until 2009 or later, will be Intel only, but that would seem perfectly logical in light of this trend.

This doesn’t mean that you should prepare to toss your PowerPC Macs right now. The movement away from Universal will be slow, and will first involve software that writes directly to the hardware. In addition to a sound-editing application, for example, don’t be surprised if games become part of this growing trend. Of course, when it comes to games and multimedia development, you always do best with the latest, most powerful hardware. Besides, games ported from Windows never worked quite as well on the PowerPC.

As for Apple, they will likely continue to produce Universal applications for several years. Operating systems? The same is true. I expect Leopard’s successor will also run on PowerPC hardware, but not 10.7, or whatever it’ll be called.

So, if you still like your old Mac, it’s not quite ready for retirement, but you should start looking for the gold watch.

Related Articles:


The Tiger for Intel Report: The Case for Rosetta

Thursday, November 16th, 2006

To be quite fair to Apple, they never promised miracles with Rosetta, the PowerPC emulation environment for MacIntels. You’d be able to run many of your legacy applications with decent performance, but you really wanted your favorite applications to be produced in Universal form.

When the tech sites began to test the first round of Intel-based Macs, they pronounced Rosetta workable, but reported about a 50% speed hit compared to native benchmarks. Photoshop would launch much more slowly and then take its own sweet time processing complicated rendering filters, so be forewarned.

At the same time, most of the people who were buying these new Macs were upgrading from much older hardware, so even in emulation, Photoshop would run at a pretty decent clip. It was all about your expectations. Besides, everything will be right in the world when Adobe releases CS3 next spring, and everything becomes Universal. Get ready to start your engines.

What some of you might have forgotten, of course, is that Apple wasn’t standing still. Almost every incremental update for Mac OS X Tiger included improvements to Rosetta as part of the package. In fact, the 10.4.8 update, as some observed, offered as much as a 30% performance boost. Even better, if you were among the folks who acquired a Mac Pro, you might find Rosetta’s performance level to approach that of a PowerMac G5. And remember, it’s still emulation.

You can also be certain that Mac OS 10.5 Leopard will contain further enhancements. True, Apple hasn’t made any specific claims yet, or even said much about Leopard beyond a few core feature additions and enhancements. But don’t be surprised if there are precise promises about how much Rosetta will be enhanced by January, at the Macworld Expo.

But that’s just one factor in making emulation more and more useful.

Do you recall when Apple switched from 68K to PowerPC? You had emulation there, too, and it seemed awfully slow at the time, at least for a while. But as hardware got faster, and the software improved, it reached a point where emulation was actually faster than the real thing!

However, PowerPC speed bursts were somewhat slow in coming, especially in recent years. Intel, on the other hand, seems to release new chips every few weeks. Well, it’s not that fast, but they aren’t standing still, particularly with AMD in their rear view mirrors, or just ahead, depending on the chip family involved.

Suddenly, the fastest Mac on the planet becomes an also-ran in a matter of months. As folks still extol the virtues of the 3.0GHz Mac Pro, Intel already has released Xeon chips with four processors. Since the chips are available at some retail outlets, it’s actually possible to buy the chips yourself and, if you’re willing to undergo a little pain and some cuts and bruises, you can turn your Mac Pro into an eight-processor powerhouse.

But hold on, as it’s only a matter of time before Apple offers an official version.

As you might imagine, it won’t just be a few multithreading applications that benefit from doubling the processor allotment. Rosetta will benefit as well, and it is poised to surpass the native speeds, even on a Power Mac G5 Quad. That has, in fact, already begun to happen in some cases even before the arrival of any new hardware.

Back in the PowerPC days, it took a few years before emulation blew away 68K. With the accelerated upgrade process hastened by Intel, it’ll happen some time in the first half of 2007. It’ll be a combination of ever-faster Intel chips and the arrival of Leopard.

Then you won’t have much reason to complain about Rosetta anymore. Of course, if you want something to complain about, I suppose there’s always the loss of Classic. Some third parties are trying to bring that back, but it’s not something you can depend on yet.

Related Articles:


Is Zune Getting Too Much Attention?

Wednesday, November 15th, 2006

Consider that Microsoft has billions and billions of dollars in the bank, ready to fight to the death to gain ascendancy in the digital music player arena. As a practical matter, you have to expect that Apple ought to fear that sort of clout. Also consider how Microsoft has been able to dominate the operating system, office suite and browser arenas and you can surely bet it’s a force to be reckoned with.

At the same time, Microsoft hasn’t done so well when it strays beyond its core business. Yes, the Xbox has been fairly successful when it comes to sales, but at what cost? Every single unit is sold at a loss, with the hope that enough Microsoft games will be sold to make up the difference. Or perhaps they’ll come to dominate the market someday and be able to raise prices yet still retain market share.

Of course, if I had some Microsoft stock, I’d be seriously considered about this kind of business plan. I’m in it for the money, and that’s not a way to make money. On the other hand, I’m not a stockholder in Microsoft or any other company.

Now it’s clear that Zune has earned a lot of attention simply because of where it comes from, and because Microsoft has embraced much of Apple’s strategy in creating an ecosystem around the new music player. Giving its PlaysForSure partners the brush off, the Zune won’t even support that software or hardware scheme. Songs you acquire from Napster, the former MSN Music and other PlaysForSure-complaint services won’t work. Period. If you bought music from those services, you have to buy them all over again. Not even a converter?

What about the poor companies that invested millions of dollars with the expectation that their partnership would pay off? In theory, PlaysForSure is still in operation. In practice, they have a right to be seriously concerned, and they will think twice about ever partnering with Microsoft on any other misbegotten ventures.

As to the Zune player itself, it has gotten so-so reviews from nearly every major technical resource, except for CNET, which seems to adore the player for some unaccountable reason. Sales of the black version appear to be decent at Amazon, where it has been in the top 10 for the past day. The other colors have fared much worse, and sales at retail channels are also reported to be tepid.

Unlike Apple’s new products, few are lining up to acquire a Zune. Except for a few so-called “fanboy” sites devoted to the product, not many are paying serious attention. If it doesn’t take off right away, I rather suspect the media will look elsewhere for stories.

This isn’t to say Zune must be a failure. The first product indeed seems rushed, of course, probably because Microsoft had to have something ready for the holidays. But if they lick their wounds and go back to the drawing boards, they certainly have the talent to build something that would truly compete with the iPod in looks and perhaps even in the interface department. But how many people will cut them enough slack to allow them to perfect Zune? Besides, can they really turn on a dime like Apple and gain an advantage beyond a couple of questionable extra features that seem to have been added by committee?

But maybe Microsoft is just getting far more attention than it truly deserves. If any other company came out with the Zune player and the Zune Marketplace, it would get a couple of stories and that would be it. With Microsoft, even after all the failures to extend its reach beyond its businesses, they’re still large enough to capture attention.

So, yes, you have to take Microsoft seriously. Surely Apple does, and if Zune were to gain any traction at all, you’d quickly see changes in the iPod line to regain its advantages. For now, let me confess that I have not seen a Zune player in person. I have not examined Microsoft’s Zune Marketplace site either. To be perfectly selfish about it, if they carried my Podcasts on the site, I’d pay attention. Otherwise, I have better things to do.

Related Articles: