su@dartvax.UUCP (Peter Su) (09/08/87)
Hi, in recent months I've lots of ads and hype about this product from Steve Jasik. After reading the article in Mactutor, I'm wondering if anyone out there has acutally used this thing, and if so, is it really that good? I generally develop in LightSpeed C, and I'm trying to see whether I should wait for THINK to come up with a debugger, or go out and get "The Debugger" instead. Thanks! Pete
stew@endor.harvard.edu (Stew Rubenstein) (09/10/87)
In article <7019@dartvax.UUCP> su@dartvax.UUCP (Peter Su) writes: >Hi, in recent months I've lots of ads and hype about this product from >Steve Jasik. After reading the article in Mactutor, I'm wondering if anyone >out there has acutally used this thing, and if so, is it really that good? > >I generally develop in LightSpeed C, and I'm trying to see whether I >should wait for THINK to come up with a debugger, or go out and get >"The Debugger" instead. Summary: I love and hate it; it has a lot of great features, but it is neither finished nor debugged, and the documentation is quite terse and, in many cases, incomplete. Personally, I use The Debugger when can, I leave TMON installed to catch the unexpected bus error, and keep MacsBug handy for when, for one reason or another (and the times do come up) the other two don't work. I've been using this for about two months now, and am of mixed feelings. On the one hand, it has a lot of great features, and the multiple window operation is a lot easier and more flexible to use than the rigid TMON windows (let alone windowless MacsBug). On the other hand, it is a royal pain to use because it doesn't stick around; you have to start it up every time you want to debug an application, and then reboot when you want to go back to something else. It is not quick starting up, either. Furthermore, the program (in its July incarnation) was full of bugs and documentation was about 20K of online notes in Jasik's inimitable terse style. The August version arrived last week and came with a (still incomplete but much better) printed manual. I have not had a chance to use it much, yet; some bugs remain, but I would rate this version as useable. The Debugger reads LSC project files (I program mostly with MPW). If you want to debug programs now, Apple's or Think's future plans don't matter much. You can't use The Debugger for anything but applications, at the moment. If you are working on INITs, DAs, or anything else, use TMON or MacsBug. Jasik says he's working on supporting DAs. Also, you can't use The Debugger while any AppleTalk activity is going on: no spoolers, AppleShare, LaserWriter printing, etc. The Debugger does support the full 68020 and 68881 instruction set, which TMON doesn't. Stew Rubenstein Cambridge Scientific Computing, Inc. UUCPnet: seismo!harvard!rubenstein CompuServe: 76525,421 Internet: rubenstein@harvard.harvard.edu MCIMail: CSC
jwhitnel@csib.UUCP (Jerry Whitnell) (09/10/87)
In article <7019@dartvax.UUCP> su@dartvax.UUCP (Peter Su) writes: >Hi, in recent months I've lots of ads and hype about this product from >Steve Jasik. After reading the article in Mactutor, I'm wondering if anyone >out there has acutally used this thing, and if so, is it really that good? > >I generally develop in LightSpeed C, and I'm trying to see whether I >should wait for THINK to come up with a debugger, or go out and get >"The Debugger" instead. I'm using LightSpeedC and decided to pick up a copy of The Debugger when it showed up at ComputerWare. I tried it out but ended up going back to TMON. Why? First the good points. It is a very powerful debugger with support for LightSpeedC symbols. It provides a large set of breakpoints, watchpoints and trap conditions. It is supported by Steve on both Delphi and CompueServe. It uses the standard Mac interface (windows and scroll bars). Now the bad points. It is a memory hog. It is still buggy (I was able to crash The Debugger twice using the example program). The user interface is up to Steve Jasik's standards which means it is very bad. It does not let you run direct from the LightSpeedC enviroment, you have to load the Problem Program (the program your testing) after the Debugger has been loaded (it's an application, not a resident program like TMON). The manual is scattered and not very well orginized. If you need the ultimate debugger, this is probably it. But be prepared for the high learning curve and the bugs. > >Thanks! >Pete Jerry Whitnell It's a damn poor mind that can only Communication Solutions, Inc. think of one way to spell a word. -- Andrew Jackson
tim@hoptoad.uucp (Tim Maroney) (09/15/87)
Thanks for the comments on Jasik's Debugger. I have been curious ever since
his article/ad in MacTutor, and now I know: stay away. It's slow, it's
completely useless for anything I do since it's incompatible with Appletalk,
and it has a Jasik user interface. Nosy has quite literally the worst user
interface I have ever used. Even if you make allowances for "first Mac
program", it's still amazingly bad. I hoped that he'd straightened up for
his debugger, but apparently not.
But I didn't fork an emacs to bitch about Jasik. Has anyone heard of any
plans to create a MultiFinder-based debugger? Now that Apple has finally
upgraded their system software to the state of the art at the time of the
Mac's introduction, a real debugger application should be possible, not one
of these TSR-type hacks. An application-based debugger shouldn't be any
more fragile than an above-BufPtr one, and the user interface benefits are
obvious. Not only that, it should be much easier to develop the debugger.
Any interesting rumors along these lines?
--
Tim Maroney, {ihnp4,sun,well,ptsfa,lll-crg}!hoptoad!tim (uucp)
hoptoad!tim@lll-crg (arpa)