Home

This note is intended to answer one question in the guestbook and to let anyone who still might use MNT know where things stand at present. An update is very much overdue after 3+ years but I certainly have not forgotten MNT. About 18 months ago I got hit by a triple whammy when my G4 bit the dust forcing me to switch to OS X full time on the G5 that replaced it. Then Apple announced the switch to Intel processors followed by Metrowerks decision to abandon CodeWarrior (or more properly not to provide support for Intel processors) which is the development system I have used for the past 10 years. I nearly threw up my hands in disgust and gave up on software development entirely. After getting very lazy for 18 months I decided to give it one more try and to try to learn XCode, Apple's own system which is used by most developers these days. It is nowhere nearly as intuitive as CodeWarrior and somewhat klunky to use but should serve the purpose once I manage to figure it out more fully. The problem still remains that Apple continues to reinvent the wheel faster than I can keep up but I am going to try to make the transition in the same stepwise manner that I have always taken. An old dog like me can only learn so many new tricks and it is simply not possible for one single developer to keep up with Apple's 10,000 member staff. But still I am going to try to get enough up to date to make MNT useful once again.

Several changes to MNT will become necessary due to the changes imposed by Apple and Metrowerks. First of all, MNT will no longer be able to support OS 9. This is unavoidable simply because XCode does not target OS 9 at all - it only supports development for OS X. Secondly, I will probably remove all PAR support for two reasons:

• First of all, this capability was added to provide PAR support under OS 9 which did not exist otherwise. MacParDeluxe does a far better job of this under OS X than MNT ever did (MPD is what I use myself now) so there is no reason for providing PAR support in a program that is destined to run only under OS X.

• Secondly, PAR processing code involves a LOT of endianness changes to support Intel processors. By eliminating PAR support it will make it much easier to make the eventual transition to a future universal binary.

• It is possible that PAR support could return to MNT later on as a Unix executable (same way that MPD does it) once I figure out how to do that in XCode. This should be much simpler to maintain than the highly modified PAR code as it now exists in MNT. This would also allow for support of PAR, PAR2 and PAR3 once that becomes available. For those of you who have never heard of PAR3s see the QuickPar website for PCs (Google it - I don't have the link handy at the moment). From what I understand, PAR3s use the same Galois math that makes PARs possible inside of a FFT of sorts and is supposed to be something like 100 times faster than PAR2s. It was a year ago when I first read about it so it should be close to becoming a reality soon.