Path: utzoo!utgpu!watmath!att!dptg!rutgers!usc!elroy.jpl.nasa.gov!ames!think!bloom-beacon!mit-eddie!uw-beaver!blake!oregon!joe From: joe@oregon.uoregon.edu (Joe St Sauver) Newsgroups: comp.lang.postscript Subject: Postscript code needed to recover LaserWriter VM Message-ID: <6318@oregon.uoregon.edu> Date: 9 Aug 89 15:34:42 GMT Organization: University of Oregon Lines: 25 Can anyone point me to Postscript code that will allow me to do the software equivalent of turning an Apple Laserwriter off and back on again, or otherwise recover needed virtual memory? Essentially, in case someone can suggest a better alternative, here's the story: Postscript output is being generated by TeX/dvips (on a NeXT sitting on an ethernet). The Postscript is then being transferred (using binary FTP) to a PC with a LocalTalk card. LWPRINT is then used on the PC to send the Postscript to a shared Laserwriter (sitting on the LocalTalk network). [Before you die laughing at this arrangement, we *will* be striving to replace this rather kludgy "solution" with CAP before too long, but for now, here's what happens...] The Laserwriter performs like a real soldier *IF* the NeXT output is sent to the LaserWriter just after it has been powered down and back up. However, if the normal hodgepodge of jobs from Macintosh and Novell-network have been pumped through ahead of it, the NeXT Postscript output croaks with a VMerror when sent from the PC with LWPRINT. Incidentally, before anyone asks, yes, I have reduced the amount of memory declared to be available to PS to only 35000 bytes, so the NeXT is *not* trying to send Postscript suitable for a memory-rich NeXT laser printer. (I've tried lower values, but 30000 or below hangs dvips). Remember, it works fine if the NeXT output is the first job onto the newly powered-up LaserWriter.