The question isn't whether Apple can "win" with the G5 and OS X, the real issue is will Apple become a respectable second to WINTEL. The G5 is a nice processor and OS X is arguably the best OS, but even Steve Jobs has admitted that the "OS wars are over and Microsoft has won" (that's a near direct quote from Mr. Jobs, made a few years ago).
As far as the 3.0GHz G5, I believe that the statement was that they (Apple and IBM) had made a commitment to be at 3GHz in another 12 months. Whether that 12 months started at WWDC or when the first G5 actually ships is probably uncertain. Thus, it may not be until next September that we actually have a 3GHz PowerMac G5.
The truth, as I see it, is that a __single__ 2.0GHz G5 is roughly comparable to a 3.0GHz Pentium IV. The single Pentium will have advantages in integer operations, but in floating point and SIMD operations the G5 will be fairly close. Thus, I don't think there is any question that for many tasks (but certainly not all) a dual G5 will outperform any available Pentium IV system. However, Apple needs to ship the G5s in quantity before the end of September. If they miss their end of August or early September target dates and/or if there are serious problems with the new G5 PowerMacs then things could look pretty bad for the remainder of the year.
Also, remember that this is the year of the notebook computer (Mr. Jobs said it and it's true -- PC notebook sales during last quarter have surpassed desktops). Thus, there is a further problem ahead for Apple since the Pentium M and Intel's Centrino products have pretty much closed the quality-of-experience gap that once existed between PowerBooks and PC notebook computers. It could even be argued that the Pentium M has taken the lead from Apple. So, now that the G5 has more-or-less leveled the playing field on the desktop Apple needs to be working very, very hard to bring better performance to their entire line of notebook computers. If they aren't able to do that soon, then I believe that Apple might begin to see some very serious erosion in notebook market share. It may be too late and unreasonable to expect an improved G4 from Motorola (for notebooks), so it may be a long and "slow" ride until PowerBooks based upon the G5 bring something new to Apple's notebook line. Long because I really don't expect to see G5-based PowerBooks until January 2004 (earliest?).
So, is Apple's "glass" half empty or half full? Right now I think it's pretty uncertain. It's going to be an interesting race, but there is no question that Apple will need to be working double-time in order to keep up with WINTEL.
