🚀 go-pugleaf

In article <1992Nov6.102017.17897@sequent.com> furballs@sequent.com (Paul Penrod) writes:
>>Q: Is compiling under C++ inherantly slower than C??
>
>Yes, always...
>C++, because of its very nature, contains quite a bit of overhead,
>since the compiler must resolve such cuties as polymorphism,
>inheritance, and the like.

But Turbo Pascal has all of those, and is much faster than any C or C++ 
compiler I've ever seen.  

>With your compile times under a minute, I dont think you can
>optimize all that much, nor should you have much to complain about.
>Now, on one of the project I am working on, it takes about 45
>minutes to compile 7 megs or source, broken up into 15
>subdirectories, and several hundred files. 

I don't have any projects that big to compare, but the compile + link time 
for a project with 350K of source (1/20th of yours) is 9 seconds
under TP (1/300th of your time, assuming you were including 
linking, i.e. 15 times as fast).  

>The fastest optimization came when I switched away from a 386 and
>went to a 486/33 from a 386/25. The compile times are now down to a
>more reasonable 15 minutes.

My timing was on a 486-25 system, without paying any special attention to 
optimization. Still it was 5 times faster than your compiler.

Why do people accept such slow compile times?  Turbo Pascal demonstrates 
that fast compiler technology exists.  Is C++ five times harder to compile 
than TP?

Duncan Murdoch
dmurdoch@mast.queensu.ca