XBR2D96D@DDATHD21.BITNET (Knobi der Rechnerschrat) (10/28/88)
This message is intended for Mr. McCalpin at FSU. I've problems to
reach him directly, so I use the list. I'll be happy about any comments
on the contents of the message. MK
--------------------------------------------------------------------------------
Dear Mr. McCalpin,
here are some more answers/opinions from me on the 4D f77 compiler:
1) -C is available and seems to work.
2) -pg is available (derived from documentation), but I have never used it.
3) The dbx interface (f77/4D) is much better than that in the 3.5 release of the
3130 software. I don't want to comment to much on the 3.6 release for the
3130 or the current release for the 4D, because we are more C oriented
programmers.
4) I'm not shure about the argument checking, but I suppose the answer is no.
5) They have still problems (and that is my main concern about the f77-stuff).
You have to ensure by yourself that the values that you pass to the library
functions fit into the allowable range. Btw. what is the PARANOIA test suite
and where can I get it from ???
6) The optimisation is quite good. I'm not shure about -O3 (global optim. on the
object code), but -O2 the default works ok (at least on our problems).
Regards
Martin Knoblauch
PS: I'm going to send this message to all of your e-mail addresses, because
I seem to have trouble with "mccalpin@fsu".