- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I'm still in learning mode with today's fortran.
Why does the IMSL routine LIN_SOL_GEN want arguments declared like X(N,0) when being called to calculate a matrix inverse? This seems really strange. It must have something to do with the argument not being needed in the IMSL for the purpose of determining the inverse.
Brian in Austin, TX
Link Copied
1 Reply
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
This particular IMSL routine does more than one job: (i) solving equations, (ii) inverting a matrix, (iii) others (?). Therefore, a convention is required to tell the solver which particular job is involved in this particular call. Note that it is only the size of the matrix b, along with the value of nrhs, that are used in finding the matrix inverse. The matrix b is neither set before calling nor are its values used after the call.
In other words, the authors of IMSL devised a convention to enable distinguishing between Job (i) and Job (ii). If you use their library, you follow their convention. If you were writing the library routine, you might choose an entirely different convention.
Reply
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page