[offlist]
On Jan 26, 2008, at 11:56 AM, Charlie Dickman wrote:
I appreciate everyone's replies. New architecture requires new
learning and I'm learning.
It occurred to me that the order of evaluation of arguments is not
guaranteed; thanks for the reminder.
It was also a head slapper that the G$ and Intel chips are
different endians. Again, thanks for the education.
The only thing I can say in defense of my own ignorance is that
they worked when coded on the G4.
There really is no defense IMO for being unaware of the single best
known issue encountered in the transition to Intel processors. Apple
has documentation specifically for people porting PowerPC code to
universal binary code, and it's been around since they announced the
switch to Intel processors:
http://developer.apple.com/documentation/MacOSX/Conceptual/universal_binary/index.html
or you can have your own copy for reading on airplanes:
http://developer.apple.com/documentation/MacOSX/Conceptual/universal_binary/universal_binary.pdf
There's an entire section are architecture differences.
At any raye I have fixed my errors.
If you didn't know about endian issues, there's a chance there are
other issues about which you are not aware. Learn first, code
second. It's really in your best interest because it will save you a
lot of time wasted scratching your head, tracking down bugs, and
rewriting incorrect code. Unless your code is very, very simple,
there could well be places where problems might not show up for a
while if you rely on testing to identify problem areas in your code.
Larry
