I had this problem -- anytime I booted in OS X, my PowerMac 7500/Crescendo G3 could not unstuff any archives that had many small files without *subtly* corrupting some of them (without notification!) ... I wouldn't find out until I tried to read the specifically corrupted file. I discovered the perfect way to replicate/test the problem with more certainty ... take a large archive full of tiny small files (15.2MB was the stuffed-size of my test case), and decompress it *repeatedly*, and then compare the sizes of the resulting folders ... they should all be identical. But mine were always different. If you haven't already, you should do this test yourself! Make sure your test archive contains a LOT of little files, some big ones, and at least some files with resource forks -- I'm not saying those things are at all the issue, but my failed archives had all of those attributes and you are trying to replicate my results. The decompressed folders would turn out sometimes a few percent bigger, sometimes a few percent smaller. You could have this problem and never know it because 95% of the files in the archive decompress fine. On OS 9 though it still all worked perfectly. I tried everything to solve this ... swapped out every RAM chip, unplugged every PCI device, unplugged absolutely everything on my motherboard I could do without. Played endlessly with cache control. I even took out half my video RAM (and then swapped it with the other half) (and then did it all again in the other pair of slots). No luck.
This was an unacceptable situation. If I'm going to switch into OS 9 whenever I'm going to run Stuffit, well that's just not workable. Almost anything but that. Can't lose Stuffit. This is what gave me the motivation to take the trouble to try all over again (because I was at the point of giving up on X), but this time for Jaguar instead of 10.1.5 ... it was a hail mary pass. Maybe the problem would be gone when everything was changed... Well, it is gone now. The same version of Stuffit now expands perfectly. Never ANY variation in the decompressed size, no matter what my cache control settings, and even when I plug everything back into the motherboard. The only things that changed were 10.2.3, and the fact that I was forced to figure out how to do it as a clean install (with hard drive format) after XPostFacto (see the thread 'wherefore iTunes' for how I did). This latter thing could very well be important too, because doing a clean install of Jaguar allowed me to update my hard disk driver. (At least, I think it updated it.) And that could affect file corruption I think as much as the OS update. I would love to know some of the reasons why all these things stop working and start working, and stop guessing at them, but it looks like that's the world we're in here... My PCI slots are still a problem, by the way. A lot of cards don't work. File corruption abounds when I use them. But then, that was a problem for me on OS 9, too. Accelerator cards and PCI architecture seem to mesh rather imperfectly in any OS. Paul. P.S. This is all with XPostFacto 2.2.5 ... (2.5.5? whatever, it's the very latest one as of now) ... PM7500, 80MB RAM, 1GB drive, Crescendo G3/350, and Power Logix Cache Control X 2.1b4. I have a PCI FireWire card and it doesn't work but having it in or out has never altered my system's behaviour in any way, and I have swapped it out in response to almost every problem. Hope this rambling attempt to communicate has imparted something usable to you all. -- Unsupported OS X is sponsored by <http://lowendmac.com/> Support Low End Mac <http://lowendmac.com/lists/support.html> Unsupported OS X list info <http://lowendmac.com/lists/unsupported.html> --> AOL users, remove "mailto:" Send list messages to: <mailto:[EMAIL PROTECTED]> To unsubscribe, email: <mailto:[EMAIL PROTECTED]> For digest mode, email: <mailto:[EMAIL PROTECTED]> Subscription questions: <mailto:[EMAIL PROTECTED]> Archive <http://www.mail-archive.com/unsupportedosx%40mail.maclaunch.com/> Using a Mac? Free email & more at Applelinks! http://www.applelinks.com
