JB said:
Oh they'll notice when they buy a new x86 Mac, go to install Quark Express and Photoshop, and realize that they have to spend another grand upgrading their software. Hold outs will notice when the new versions of apps that they need only run on x86.
Apple is saying that rebuilding for x86 is easy, and they are probably right. But releasing new products targeted at both ppc and x86 Macs will cost the software vendors. They'll probably punt on PPC soon after the x86 Macs are released.
I'd imagine that many s/w vendors would offer a reasonably priced upgrade to migrate from a ppc to x86 app. Reason being the mac will be flooded with a host of new competitive apps from companies that didnt want to bother with making a ppc and x86 version of their s/w. This could be a real boon for mac users. The recross pollination might mean x86 windows users might also get better, easier to use apps that come over from the osx/ppc platform.
As far as vendor cost, quite a few are already building and maintaining both osx/ppc and windows/x86 versions. This should make their lives a little bit easier rather than harder, but theres a hump to get over. Hopefully apple gave their key developers the wink and nod at least a few months ago so they could get going on the work.
There may also be a reasonable emulation solution. I would imagine that a 2GHz Dothan Pentium-M with 64 bit extensions would run a full-on osx/ppc emulation with reasonable enough speed to get someone by for a little while until the native s/w becomes stable. If that doesnt do it, a 4GHz pentium 4 will probably do.
SG - your points are well made. While there have been some small performance disparities from time to time there hasnt been enough of a 'kick' from the risc camp to compel application developers or businesses/users to want to jump. Which in a way is a little pathetic if you really know what an incredible kludge the x86 architecture is. And processors have never been the big deal for either IBM or motorola. Its just about the only deal with intel. And considering intels track record with non-processor stuff, you could remove the words 'just about' from the last sentence and not feel like you really misstated.
My first comments on the PPC architecture were something like "they're going to have a chicken and egg problem with product acceptance/penetration and applications availability, and they're also never going to get enough of a performance edge on IA to really make people take a hard look. It'll be a niche product at best.".
Considering I once said that MS-DOS would never knock over CP/M, that the IBM PC would never sell and that there was no way we'd ever get high speed networking running over unshielded twisted pair wire, I'm pleased that I got it right at least once.
Maybe John is right, maybe I am a moron in disguise
Perhaps I *should* have listented to my friends siren song about sexy mensa women and joined the club to help bolster my own self image and validate my intelligence...