the fms fortran monitor system preceded and led to the ibsys operating system. both were tape based, while still being locked into the punchcard framework. a tape was in some sense a ‘virtual’ card deck, simply a newer faster way to represent a stream of cards, without the physical paper cards themselves. each line of the file written to tape is a card, and the complete file is a deck.

looking at the old jobs archived at rich’s site, many begin with the following lines/cards. ‘execute fortran’ means ‘enter the fortran monitor’.

$job           name1
$execute       fortran

and here’s the later and in some sense more capable ibsys sequence. this enters the ‘ibsys ibjob processor’, sets the job processing for a ‘core memory storage map’, and then enters the ‘ibftc fortran compiler’. assembly and cobol are other options for the last card.

$job           name2
$execute       ibjob
$ibjob         map
$ibftc  deck1  list,sdd,nodeck

another difference is that in an fms tape/deck the fortran code main body has to end with the non standard ‘call exit’ instead of ‘stop’. and there are special monitor command lines/cards beginning with ‘*’. it seems that the ‘classic’ fms was associated with fortran ii. it’s not clear whether fortran iv was possible within fms or not. there’s a later ‘fortran iva monitor system’ within the ibsys framework that may be the iv descendant of the classic ii fms.

for the ibsys sequence, the ibftc card is for the ‘fortran iv compiler’ component of the ibjob processor, which is a component of the ibsys operating system. this can serve as a baseline/reference fortran compiler. on page eighteen of the linked manual is an interesting note concerning ‘carriage control’. The WRITE (i,n) list statement prepares a BCD tapе that can be used to obtain off-line printed output. The PRINT n, list statement prints on-line during execution. this confirms that print is important in the simh context. and the same paragraph seems to explain why the ‘0’ appears in holleriths like ‘5h0hello’. it seems to be a carriage return control character specifying double spacing on the line printer.

further experimentation has finally cleared up the mystery regarding write. the ‘bcd tape for off line printed output’ is on seven, at least with simh config that came from rich. so what works is write(7,n) and then ‘./listtape -b sysspp1.tp’ after the run. at last, printed output both on the ‘console’ and an ‘output tape’! what’s often shown in documentation is write(6,n), but simh 7094 is write(7,n), and simh dec10 is write(5,n).