pdevries@watserv1.waterloo.edu (Peter DeVries) (05/04/90)
In article <2453@dataio.Data-IO.COM> bright@Data-IO.COM (Walter Bright) writes: >In article <1990Apr19.153450.23508@cs.columbia.edu> baker@cs.columbia.edu (Michelle Baker) writes: > ><(3) any opinions on the debugging facilities.... e.g. how do they compare >< with the Microsoft C debugger? >It's better! :-) (You asked for an opinion!) Wait a second! Has anyone managed to debug a C++ program with the zortech debugger that is any bigger than a TOY problem? We have found that although the debugger is pretty neat, we are really stuck with Nothing at all once we try it on our real application. Peter DeVries Mutual Life of Canada c/o mgardi@watdcsu (519) 888-3523 (416) 972-0594 My opinions/comments are mine, and mine only, and have nothing to do with what Mutual Life of Canada thinks (now isn't that an understatement!) p. -- Peter DeVries Mutual Life of Canada c/o mgardi@watdcsu (519) 888-3523
cs326ad@ux1.cso.uiuc.edu (05/05/90)
I must include my two cents here..... I recently completed a small front end pascal compiler for class using Zortech's C++ and ran into several problems with the debugger. As a whole, the program and all modules were 2500+ lines. I went to debug a certain problem (actually several), and had mixed results. At one point, I was viewing the C++ code that I was using, and tried to move down the program listing to set a break point at a certain case label, and I ran into all assembled code....Yes! all the code after a certain point was assembly. In fact, after a certain point, I couldn't move any further down the program listing. The source file was around 1500 lines (the parser file), and I could only get to around line 500, when I really needed to be at line ~1300. It helped for several other small minor bugs, but I never did find the problem with this one. It seems that a certain member of a structure was getting redefined somewhere along the parsing, and I could not find it anywhere, and that was when I became somewhat disenchanted with the debugger. I suppose someone could flame me for RTFM, but I checked over all the relevant parts in the manual, and could come up with no logical explanation for what happened. After the semester is over, I plan to take up the problem with Zortech...... =========================================================================== = = = Mark Allender University of Illinois = = cs326ad@ux1.cso.uiuc.edu Urbana, Illinois = = ma20449@uxa.cso.uiuc.edu = = = ===========================================================================
Bob.Stout@p6.f506.n106.z1.fidonet.org (Bob Stout) (05/06/90)
In a message of <May 04 16:49>, Peter DeVries writes: >><(3) any opinions on the debugging facilities.... e.g. how do they compare >>< with the Microsoft C debugger? >>It's better! :-) (You asked for an opinion!) > >Wait a second! Has anyone managed to debug a C++ program with the zortech >debugger that is any bigger than a TOY problem? Yes! how many lines are in your source files? Assuming reasonable modularity (a reasonable, though not always valid, presumption when discussing C++) the debugger should run OK - also assuming you read the manual and have set the switches so it has enough memory to work. >We have found that although the debugger is pretty neat, we are really >stuck with Nothing at all once we try it on our real application. For truly mammoth applications, it's already been publicly announced (on BIX) that the 2.10 debugger will include variants to run with DOS extenders, in protected mode, and to do remote debugging. Since this much has already been leaked publibly, I can say as a beta tester that these do exist and seem to work as advertised. The bottom line is to compare the alternatives. CodeView is brain-dead (CV 3.0 which comes with MSC 6.0 is a little better, but not enough - mostly they just tacked on a new interface). Turbo Debugger is the state of the art, but the new ZDB has almost all the same features and works with ZTC++. In any case, programs that run out of memory with ZDB will also run out with TD (this is at least true with the new ZDB).