> And I still can't work out how to use MAKE.  No, not quite - every
 > MAKE I use seems to be different!  I don't have enough hours in
 > the day.

Yes, they are different, which is why I've only used one at a time for all
compiling for some years.

Make's I have used are ZMAKE (Zortech's), Polytron's POLYMAKE, NDMAKE for
some 3-4 years and switched from that to DMAKE in 1990 and have been using
it ever since. I've looked at plenty of others, and even used them in
employment or on contract (including bsd make, pmake, sysv make, and of
course MS's nmake and Borland's make), but for my own work on all operating
systems I commonly use, dmake is my 'old friend'.

I haven't yet found anything it can't do or can't be made to do in one way
or another and it has saved more hours in the day than it has cost. I've
often kicked myself for not rereading its manual and discovering even more
features that I could have been using years ago that would have saved
considerably more time and convenience.  Availability of source was the
real reason I originally swung to it; that, and the way it very elegantly
handles response files under MSDOS (in fact, any operating system). I
didn't even know that the msdos & os/2 bound version had parallel make
capability until I tripped over the option in the manual and tried it.

As an example, one time-saving idea I tried only recently was to get the
make to first write a batch & response file for a single invocation
compile and link for OS/2 DLLs with several hundred individual modules.
Previously, dmake used to take around 45 minutes to compilete a full
compile and link in separate steps using BCOS2 (over two hours using
C++Set/2, but that's another story :-().  It now takes less than 10 minutes
since dmake writes a response file with all the files to compile, supplied
libraries and options, and BCC does it all in one hit.  I haven't tried
this with C++Set/2, but I suspect that the difference will be even more
marked.


  cheers,
  david

--- MaltEd 1.0.b5