bathurst@phoenix.Princeton.EDU (Bruce Bathurst) (04/18/91)
Has anyone else been having problems with "name association" on STSC's APL v.10 for the IBM PC? On my old vanilla IBM PC and my Toshiba T1000, the function "examples" in their provided workspace gives syntax errors; and, worse, my own functions load and attach properly, but hang the machine when they run. One briefly appearing error said the coprocessor was sent a nan. "Name association" is why I upgraded to v.10. It allows subroutines in Fortran and C, compiled with certain specific compilers, to be brought into the interpreter as locked APL functions. The possibility of using the LINPACK and EISPACK routines on the fly is very attractive. Before I send STSC a floppy disk, I would like to know if anyone has gotten "name association" to work properly on their PC. Thanks very much! Bruce Department of Geological and Geophysical Sciences Princeton University, Princeton, NJ 08544 bathurst@phoenix.princeton.edu bathurst@pucc.bitnet !princeton!phoenix!bathurst Department of Geological and Geophysical Sciences Princeton University, Princeton, NJ 08544 bathurst@phoenix.princeton.edu bathurst@pucc.bitnet !princeton!phoenix!bathurst
bathurst@phoenix.Princeton.EDU (Bruce Bathurst) (04/23/91)
(OK all you J only people) I'll have to answer my own question. I couldn't get "name association" to work on STSC's v.10 APL PC. STSC's hotline staff has always proved unusually intelligent and polite. Two weeks ago "Bill" had no trouble running the examples, but said he would check with the developer. He phoned me back a few moments ago. The example workspace doesn't run on my IBM PC or Toshiba T1000 because these have 8088 processors, and the executable C functions were mistakingly compiled by STSC with the 81888 rather than the 8088 switch on. I think it's very odd that I should be the first to notice this, because "name association" adds extraordinary power to the use of APL in calculations that require numerically stable algorithms. Bruce Department of Geological and Geophysical Sciences Princeton University, Princeton, NJ 08544 bathurst@phoenix.princeton.edu bathurst@pucc.bitnet !princeton!phoenix!bathurst
dave@visual1.jhuapl.edu (Dave Weintraub) (04/23/91)
In article <8575@idunno.Princeton.EDU>, bathurst@phoenix.Princeton.EDU (Bruce Bathurst) writes: ... |> |> ... "name association" adds extraordinary power to the use of |> APL in calculations that require numerically stable algorithms. |> |> Bruce If you think it's nice on STSC's APL, you should see that way that APL2 supports QuadNA, which is *much* more powerful, IMHO. Now, if they would only announce this for their PC product. Dave dave@visual2.jhuapl.edu