Is it just me, or is anyone else having a hard time getting excited about the new Intel iMac and MacBook Pro? Aside from the laptop’s terrible name, both bits of kit look like they should be sure fire winners. I’m told that the processors used in the new Intel Macs are clock for clock faster than the G5 processors they replace (and kick the G4 all over the park to boot).
A dual core laptop? Yes please. Granted it has some shortcomings - the DVD drive is not dual layer, the screen is slightly small than the PowerBooks, and it’s missing some output dongles. But for the speed increase I can live with that - DL media is still pretty expensive and regular DVDs are pretty cavernous, 60 pixels isn’t that much and you can buy the dongles separately while only have a vague feeling of price gouging.
A desktop Mac that’s potentially faster than a current dual core G5 PowerMac and comes with a gorgeous 20″ screen? Tasty. Kodawarisan has just posted pictures of a disassembled Intel iMac. From this picture it looks like the CPU is socketed rather than soldered onto the motherboard which could make for pretty easy upgrades using cheap commodity parts - no more expensive processor cards. What’s not to like?
Software. That’s what. By releasing the Intel Macs six months early, it seems to have taken most developers by surprise. There are a lot of small applications that have already been released as universal binaries, but where are the Adobe CS suites, the Microsoft Offices and the Macromedia Studios? Even Apple’s own ‘pro’ applications (Final Cut, et al) are a couple of months off, as is Firefox. I was told that Macrodobia would get their apps out by the end of the year, which is great, but we are at the beginning of the year. Coincidentally that is when the Intel PowerMacs (MacMacs?) are to be released, although going by last weeks announcements, that could change.
I think that part of the problem is the secrecy Apple like to shroud their development products in. Developers have had (to my admittedly rather limited knowledge) no access to pre-release machines. Normally this would not be a problem as from revision to revision not much changes, but we are talking a totally different processor architecture. There’s an article on MacNN where Microsoft say they are waiting for delivery of shipping machines to ‘better evaluate … the final release date of Mac Office’. This may be spin/playing for time, but if they had access to them before they were announced, at least that would be an excuse they could not use.
Development kits have been available but these do not contain the same hardware as found in the new Macs. For example, they are single core Pentium 4 machines. One advantage Apple have in this switch is that they do not have to support legacy hardware - such as the Pentium 4, so no developer has been able to code CPU instructions specific to the Yonah processor family (which would lead to speed increases usually around 10% like setting the -march GCC flag to your CPU type) as they just haven’t been available. Sure you could work from Intel’s documentation but without a means to test what you had just written, it would be a bit like fumbling around in the dark.
Another thing that bothers me is that in July of last year, Apple froze development for the Cocoa-Java bridge - no new features of Cocoa would be added to it, essentially removing the point of developing Cocoa apps in Java. For the uninitiated, a big thing Java has going for it is that you can write platform independent code. You write your application, compile it, and it will run on anything (Intel, PPC, Alpha, Sparc, your telephone, your fridge, whatever) that has a Java Virtual Machine available for it, which is pretty much everything. Cocoa is the framework that lets you use things like OS X specific buttons and scrollbars, etc and is very much tied to OS X and is usually invoked using Objective-C. The Cocoa-Java bridge then, allows you to use Cocoa classes in Java for that authentic Mac look and feel. Of course this robs you of being able to run your Java program on Windows, but who cares as if it’s well designed, you should be able to port the core of it (written in pure Java) and just re-write the user interface in Swing or SWIG or whatever. The release of the Intel version of OS X will obviously have the Cocoa framework implemented natively on Intel hardware. Consequently, if you’ve used the Cocoa-Java bridge to write your application, it should run on the new Intel Macs with no tweaking whatsoever. This would obviously greatly speed the deployment of Intel OS X applications, but, with no more development being done on the Cocoa-Java bridge, it’s not going to happen as who wants to be stuck using a dead API? What’s really a kick in the teeth is that they froze development after announcing the move to Intel.
It all looked so promising last year, the transition looked like it was going to go very smoothly. Now we have amazing hardware - that PowerBook update that everyone was waiting for - but no software to run on it, for now. All eyes to the WWDC in June…
Popularity: 8% [?]