home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!stanford.edu!rutgers!news.columbia.edu!cunixa.cc.columbia.edu!sll2
- From: sll2@cunixa.cc.columbia.edu (Steven L Levitt)
- Newsgroups: comp.sys.mac.programmer
- Subject: Bad Floating Point bug in Think C 5.0
- Message-ID: <1992Sep1.212923.24833@news.columbia.edu>
- Date: 1 Sep 92 21:29:23 GMT
- Sender: usenet@news.columbia.edu (The Network News)
- Distribution: na
- Organization: Columbia University
- Lines: 23
- Nntp-Posting-Host: cunixa.cc.columbia.edu
-
- HEELLLLLLLPPPPP!
-
- I just moved a fairly large Think C 5.0 project from a Mac II running system
- 6.0.4 (very old, I know) to a Mac IIci running system 7.0.1. Both macchine
- were running the same version of Think C, 5.0.2.
-
- I have begun to encounter a problem when one of my functions returns a value
- of type double (I have the Native Floating Point option on). Typically,
- the value should be between 100 and -100, yet when the value is printed out
- in a window it is always 0.0. Now, here's the funny part. When I use the
- source-level debugger to set a breakpoint at the line where the function
- returns and then step past it, the value is correct, and eventually prints
- correctly on the screen. The only difference is that a breakpoint was set,
- and the program stopped just before the function executed.
-
- Could some wise and wonderful person enlighten me as to just what the hell is
- going on here? Does anyone know of incompatabilities Think C has with System
- 7 that might cause this? Thanks in advance.
-
-
- ----------------
- Steven L. Levitt sll2@cunixa.cc.columbia.edu
- An Undergrad
-