home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!zephyr.ens.tek.com!gvgpsa!gvgspd!omega!akh
- From: akh@empress.gvg.tek.com (Anna K. Haynes)
- Newsgroups: comp.os.ms-windows.programmer.tools
- Subject: MSC/SDK Beginner's Luck (BAD) -- Reality Check Needed
- Message-ID: <2010@gvgspd.GVG.TEK.COM>
- Date: 4 Sep 92 19:54:45 GMT
- Sender: news@gvgspd.GVG.TEK.COM
- Reply-To: akh@empress.gvg.tek.com
- Distribution: na
- Organization: Grass Valley Group, Grass Valley, CA
- Lines: 81
-
- I just got Microsoft's Windows 3.1, C/C++ 7.0, SDK 3.1, & have been
- trying to use them, & would like to 1) ask a question & 2) let
- off some steam lest I go critical.
-
- (
- I have a 486 with 8 MB RAM, & am running under MS_DOS
- from within Windows (i.e. clicking on the "MS-DOS Prompt" icon);
- I do _not_ have 386Max installed.) Windows is on C:, C/C++ and SDK
- are on D:. Here is my config.sys:
-
- buffers=10
- files=30
- dos=high,umb
- DEVICE=D:\sdk\bin\himem.sys /machine:at
- DEVICE=D:\sdk\bin\emm386.exe NOEMS
- DEVICE=D:\sdk\bin\smartdrv.exe /double_buffer
- lastdrive=v
- stacks=9,256
- shell=c:\command.com /p /e:1024
- )
-
- 1. Question.
-
- I started working through the MFC tutorial (Class Libraries Users Guide,
- \sdk\mfc\samples\tutorial). Sample program #1,
- dmtest.exe, has a few quirks:
-
- When I nmake it without the debug flag set,
- the output shown in the book(p.59) does not match what I get. (Both
- are wrong--in the book, Brian Meyers gets lost in the ether,
- & on my 486, Banipuli is found although s/he does not exist.)
-
- When I nmake it _with_ the debug flag set (nmake -F dmtest DEBUG=1),
- then run it, it dies (Assertion failed: file memory.cpp, line 334,
- AfxAbort called). If I try to run it under CodeView (cv dmtest --
- I haven't tried to figure out how the $%^& CodeView is supposed to
- work yet, I may be doing it wrong) it comes up in assembly language
- & I don't know how to get it to show source code.
-
- These things (the wrong nmakes) also happen on a co-worker's machine
- (I'm not sure what his config is; he has a 486 too.)
-
- WHY do we get these results???? I sure hope it is pilot error.
-
- 2.
- Steam, low pressure:
-
- a) Is it true that the Petzold book (Programming Windows, 2nd ed.) is
- out of print because (for 1 thing, at least) the examples won't work
- with 3.1? (they don't work for me) If so, what does this
- say about backward compatibility or lack thereof?
-
- b) In Microsoft Windows 3.1 Guide To Programming, the first Makefile (on p.16)
- generates error messages when I try to nmake Generic ("too many rules for
- generic.exe"). The 2nd one (p.56) is missing the "all: ..." line & so
- will _not_ "...properly compile and link the files created for Generic",
- despite the manufacturer's claims.
-
- c) It would be nice if we got more than 1 page (in Microsoft Windows 3.1
- Programming Tools, SAME 1 page in M.W. SDK Programming Tools) of
- documentation on the Dialog Editor, especially since it _does_ exist
- (27pp in QuickC for Windows Toolkit). Yes, I know online Help exists,
- but it has major limitations( since everything's in little chunks I
- don't know if I've missed something (AND NEITHER DO YOU--where does
- it explain which of the cute little Toolbox symbols is associated with
- which object?), & my brain needs the continuity of 1-dimensional
- documentation, otherwise it loses things too quickly).
-
- STOP READING HERE if it's me or my system configuration that's at fault.
-
- Steam, higher pressure:
-
- I thought Microsoft was supposed to be a reputable company, but if
- they can't even get the BASIC BASIC BEGINNER MORON stuff to
- work, .... does it get any better or is this what I have to
- look forward to? Are the proofreaders & debuggers awake up there?
-
- -------------------------------------
-
- Anna K. Haynes The Grass Valley Group, Inc. Grass Valley, CA (916) 478-3691
- <akh@empress.GVG.TEK.COM> -or- <...!tektronix!gvgpsa!empress!akh>
-