XPost: comp.lang.c++   
   From: lynnmcguire5@gmail.com   
      
   On 2/5/2026 1:23 AM, Thomas Koenig wrote:   
   > Lynn McGuire  schrieb:   
   >> On 2/4/2026 1:20 PM, Thomas Koenig wrote:   
   >>> Lynn McGuire  schrieb:   
   >>>   
   >>>> I am swinging huge datasets for simulation models from 1 MB to 1,000 MB.   
   >>>>     Nothing besides C++ has the oomph and speed to make this happen.   
   >>>   
   >>> Speed differences between compiled languages are reltively small.   
   >>> But Fortran (where I am reading this) can certainly do so.  A pity   
   >>> you missed out on Fortran's development in the last 35 years.   
   >>   
   >> We will see.   
   >   
   > We will not see, because you made the decision to switch to C++   
   > instead of gradually introducing modern Fortran features in your   
   > code.  There will be no benchmark to test against.   
   >   
   >> When I finish converting the 800,000 lines of F77 code to   
   >> C++ in my calculation engine, I will run a timing test using my 659 +   
   >> 100 = 759 benchmark files.  I just restarted this project after a year   
   >> or two of working on customers problems and demands.   
   >   
   > That could be interesting.  With such a conversion test, it would   
   > also be interesting to see how many bugs were uncovered in the   
   > conversion process, and how many new ones were introduced.  You   
   > may know the first number (although I totally understand if you don't   
   > want to share), but you may never know the second one.   
   >   
   > Do you use tools like valgrind and sanitizers?   
   >   
   >> I have over 100,000 lines of F77 converted to C++ now and the input   
   >> parser of my chemical data regression tool is fully working.   
   >   
   > Parsers are a field where Fortran does not shine, although allocatable   
   > characters have made this much easier.   
      
   I have tried valgrind in the past.  It alerts too much in my F77 code   
   since we have a homegrown dynamic memory system dating back to 1977.   
   The pointers are stored in integers and valgrind really hates those.   
      
   The only conversion problem that I am worried about is that I have   
   Fortran 66 interpreter built into my software for my customers to write   
   custom calculations embedded in their datasets attached to any type of   
   object.  I will need to write a converter for the F66 format statements   
   to C formats.  Not a big deal though.  Some of my customers have over   
   10,000 lines of F66 code in their datasets.  And yes, the F66 parser is   
   written in F77 code, we wrote that back in 1982.   
      
   Lynn   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)