MacEdition Logo : Click here
 

Does one megahertz per day keep the analysts at bay?

By Tony Leggett (pocwactso@macedition.com; Post a message about this story), July 24, 2001

Feedback Farm

Have something to say about this article? Let us know below and your post might be the Post of the Month! Please read our Official Rules and Sponsor List.

Forums

Want to dig even deeper? Post to the new MacEdition Forums (beta)!

Since this time last year, Apple has bestowed upon the Mac faithful an extra 367MHz of processing goodness. That’s roughly one megahertz per day. Given that at this time last year Apple was stuck at the 500MHz barrier, 867MHz actually seems in line with Moore’s Law, give or take. So that’s good, isn’t it?

Well, no, not really. The problem is that before this one-year snapshot Apple had over one year of supercomputer stagnation. Apple’s been making up for lost time and lost opportunities ever since, and to do that you not only need to keep pace, you need to turn it up a notch. This year’s Stevenote was very much “steady as she goes,” and the markets reacted accordingly – giving the stock a post-keynote 17-percent beating.

Did Apple deserve it? Shouldn’t we be looking at more than just the dry, boring hardware specs and get to the big picture? Of course we should – looking at just the megabytes and megahertz is quite myopic and treats Apple as just another boxmaker. Like every Macworld keynote, there were lots of announcements: Some good, some bad and some best left unspoken.

With a few notable promising exceptions, this was not a good keynote. Even some announcements that appear positive at first glance are not. Let the postmortem begin:

Mac OS X – now slipping at a store near you!

First things first; let’s give credit where credit is due. The modifications and enhancements demonstrated in Mac OS X 10.1 are an impressive improvement over Apple’s initial rushed botch-up. (Only contented users of Mac OS X 10.0.0 are entitled to flame me for that comment.) Faster performance, better connectivity – out-of-the-box NT networking! Oh, pure tears of joy! – and improved user customisability are all advances not to be scoffed at. Nor is CD burning, DVD playback (even if it barfed during the keynote demo) or the slew of upcoming applications displayed in the keynote.

OS X still has a way to go, but 10.1 gives the first hint that we may yet get an advanced operating system that actually takes advantage of the PowerPC, as opposed to taking it hostage. Let me reiterate: That is a massive achievement that deserves more than a polite clap at a keynote.

But the ghosts of previous WWDCs haunted this keynote – such as WWDC 1999 whispering how OS X would be available by May 2000, and even the ghost of WWDC 2001 congratulating itself for releasing OS X “on time.” Hopefully, they settled down the fanboys in the front row. “Shipping is a product feature” – arguably the most important feature of all, and it’s what’s missing from all of these announcements. OS X 10.1 isn’t shipping until September, and many of the “10 on X” applications demonstrated are at early beta stage only. Don’t blame developers for this; blame Apple and its wayward set of APIs, which (to quote Roxio, makers of Toast) have run into “severe delays” in getting to developers.

Let’s be honest – Mac OS X 10.1 is what the original Mac OS X offering should have been and it’s a damning indictment of the computer industry in general that so many users tolerated the flaws in Apple’s initial offering. The final insult, however, may be that the "free" upgrade is only made available as a CD update package for $19.95 (for more details, read Apple’s press release; note the absence of the phrase, “free download”). Whatever happened to the good old days where Apple let the Mac print magazines bundle software updates on their demo CDs? If there isn’t also the option of a free download, this certainly earns Apple the “Arseclown-of-the-Week” award.

Any mention of the up coming Mac OS 9.2 was notably absent in the keynote. Word from the showroom floor is that it will be shipping on Apple’s new systems but it’s surprising that there wasn’t even a one-second sound bite about it, given the promises of improved Classic compatibility. It’s understandable that Mac OS 9.2 shouldn’t steal any of Mac OS X 10.1’s thunder but some low-key announcement would have been nice for Mac OS 9’s massive installed base.

High-end stuff or high-end fluff?

If you pretend for a moment that Apple exists within a vacuum, the revisions to Apple’s pro line aren’t too shabby at all as a stopgap measure. The graphics cards that come standard with these models are a massive improvement on this time last year, and for that Apple deserves due credit. It is also important to avoid the folly of the “Megahertz Myth.” It’s not just an Apple marketing gimmick; there’s more than a kernel of truth to it. 11.8 gigaflops (the theoretical performance of the dual 800MHz G4), whichever way you look at it, is not to be sneezed at.

But the Megahertz Myth is a two-edged sword. While we snicker at the inefficiency of the P4’s mammoth pipeline, the G4 is wandering down the same track. As Apple’s site states, the dual 800MHz G4 is 20% faster than the previous dual 533MHz G4. Yet by raw numbers it should be more like 60% – this was the tradeoff Motorola made when they couldn’t squeeze anything more out of the original four-stage pipeline of the PPC 7410. Extrapolating from that, we get a processing power increase in real terms of approximately 694MHz between the PPC 7410 and the PPC 7450.

For a good comparison and explanation of the differences between the PPC 7450 and the PPC 7410 see Bare Feats’ test page. Also note that while the low-end configuration is now 733MHz, it is without the 1MB of L3 cache, so the performance will take an additional hit.

The new G4s aren’t duds – far from it. This simply illustrates the truly phenomenal performance of the original shorter-pipeline G4. However, just as only a simpleton would accept clock speed alone as an accurate performance measure, the same needs to be kept in mind with Apple’s controlled “bake-offs” as a true reflection of everyday real-world performance. It’s important to bear in mind before scoffing about the Megahertz Myth.

Proper benchmark tests in the next few weeks should yield some intriguing results.

The lamentable low end

With Apple’s refreshed iMac lineup it’s easier to list the positives before the negatives because there’s so few of them. On the plus side of the ledger it was immensely satisfying to see Flower Power and Blue Dalmation banished to the dustbin of “it-seemed-a-good-idea-at-the-time” abandoned technologies. Rest in peace, Flower Power – alongside the Cube, Microsoft Bob and the Apple ///.

Another positive was Apple’s upping of the RAM so all iMacs are OS X ready. The cost of the extra RAM is trivial for Apple and avoids the embarrassment of shipping desktops incapable of running Apple’s flagship OS out of the box. All that needs to be fixed now is the 64MB low-end iBook. Also good was the adoption of the ATI 16MB graphics card across the entire range – that’s quite respectable for an entry-level machine.

Don’t be overly crestfallen that the new lineup comes sans LCD or a larger screen. It’s almost certain that LCD (or some other flat-panel technology) iMacs will arrive. At what price points and feature sets? That’s the important “yet to be seen” element. What is of great concern are the model options and pricing of the current lineup.

When Apple released the CTO options for the iBook, I was impressed. It was straightforward and sensible with no bewildering tradeoffs between X MHz and Y MB depending on what optical drive and other bells ’n whistles you wanted. All models came in the same format with the same processor and graphics – did you want yours with just CD-ROM, DVD, CD-RW or a combo drive? That was a smart move as 182,000 customers (and counting) have demonstrated. So imagine my dismay when the new lineup comes not only with ho-hum specs, but CD-RW across the entire line.

There’s already a significant installed base of users who already have an external CD burner (previous iMac, iBook and PowerBook owners, to name a few). Offering one CD-ROM-only model (with otherwise identical specs) for, say, $100 less than the current low-end model surely would have encouraged sales from people not wanting to buy something they already have. There are also groups that specifically don’t want a CD-RW machine (such as education, which Apple has a stopgap fix for). And what about people who’d pay the premium for the combo CD-RW/DVD drive? Perhaps the one-speed-suits-all approach is not quite right for the iMac, but how about ditching the 500MHz model altogether and having two 600MHz models (one with CD-ROM, one with CD-RW) and two 700MHz models (one with CD-RW and one with a combo CD-RW/DVD drive). It’s just a humble suggestion.

The pricing, however, is mind boggling. In a weak economy where the PC market is in the toilet, raising the prices on your consumer offerings is not a smart way to increase market share. Neither is gouging your customers on what is quickly becoming a dirt-cheap commodity: RAM. Compare Apple’s price for a 256MB upgrade – $100 (after a 50% discount!) – with IBM’s $41 for a similar boost, and aftermarket prices of around $65 for 512MB . Apple leaving the original prices up on the site to make out that it’s a “bargain” with the discount borders upon being offensive.

The only explanation with any credence is that it’s a short-term strategy to prime the market for the LCD iMacs (which would be more expensive to produce) in time for Christmas. The fact that Apple has kept a batch of the old CD-ROM-only iMacs to appease education suggests the all-CD-RW lineup may not last for much longer.

Bottom line?

Us Mac users are a fickle lot. We expect Steve Jobs to pull a rabbit out of his hat every year or else we scream about the sky falling. No matter that Steve Jobs himself set up this expectation through Apple’s secretive and paranoid approach to product announcements. The real issue isn’t that there was no “rabbit,” but rather even that what was announced lacked the inspiration we’ve come to expect from a post-Jobs Apple. The ongoing delays in getting Mac OS X to the stability, performance and functionality it needs dog the company as well – the word “September” is coming up all too often.

However, Steve’s frustrated fumbling with the camera summed up his entire keynote. Try as he may, he just couldn’t press the right buttons. Still, I’m cutting him a tiny bit of slack; if that’s the worst keynote we ever see, Apple’s in for some promising times.

E-mail this story to a friend

Talkback on this story!

Cannot connect to the database.
Please contact the administrator.