Sun-Spots-Request@RICE.EDU (Dave Johnson) (12/25/85)
SUN-SPOTS DIGEST Friday, 25 Dec 1985 Volume 3 : Issue 16 Today's Topics: administrivia Re: Second source for Sun-3/75 and Sun-3/160 memory expansion boards? Sun Printers Re: SUN TIP locking woes Only a FEW? Re: Voice recognition on SUN Re: multibus adaptor for the VME bus removing batteries Performance enhansements for Sun Unix? bell in suntools? bleep on sun Dvorak keyboard in suntools sun-3 draws faster? Loading up TeX on a Sun-3 ttysw_input more memory for SUN-1/150U RS-232/EIA Flow control on SUN2s Public domain NFS daemon? Screen saver? Dhrystone benchmark - DN330 vs. Sun-3 Need help selecting Apollo vs. Sun and Relational DBMS for hospitals ------------------------------------------------------------------------ Date: Fri, 25 Dec 85 1:28:56 CST From: Dave Johnson <sun-spots-request@rice.EDU> Subject: administrivia Scott Alexander, the regular moderator for SUN-Spots, is vacationing away from Rice during the Christmas holidays and has asked me to take care of the moderation duties while he's away. Hopefully, everything will run smoothly. Please continue to send all submissions for the digest to Sun-Spots@rice.EDU and all additions/deletions/problems to Sun-Spots-Request@rice.EDU. Dave Johnson ----------------------------- Date: Sat, 14 Dec 85 14:44:14 CST From: oddjob!kaos!sra@lbl-csam.ARPA (Scott Anderson) Subject: Re: Second source for Sun-3/75 and Sun-3/160 memory expansion boards? LCF says that they will be selling a complete line of VME memory boards, from 1 to 16 MB. They promise this sometime in 1986, although I don't know exactly when. [Since the VME SUNs use a private bus for their memory (like the Multibus SUNs), these LCF boards might not be useful. Does anybody know if these really are VME memory boards, or if they're SUN private-bus memory for the VME SUN models? --dbj] ----------------------------- Date: Sat, 14 Dec 85 14:44:14 CST From: oddjob!kaos!sra@lbl-csam.ARPA (Scott Anderson) Subject: Sun Printers > We wish to use a laserprinter on both a Sun, and a Macintosh (1 of each), > and it is not obvious to us whether to buy an Apple Laserwriter, and an > interface kit to tie it up to the Sun, or to buy a Sun Laserwriter. > Can anyone comment on > (a) Price advantages in doing either > (b) Any technical diffculties either way > (c) Any other reasons to go one way or the other Buying the laserwriter from Apple is cheaper, especially if you are in the University Consortium. However, buying the interface kit from Sun jacks up the price to the point where only the Consortium price would be cheaper, and then not by much. The alternative is to get the "interface kit", i.e. Transcript, direct from Adobe; I believe that their price is about $500. I am not sure how different these are, but I think that there are only a few Sun-specific things, such as the raster filter, which you could probably write yourself if you are so inclined. We bought our laserwriter for our Suns, but there are a few dedicated Macintosh users here who are pressuring me to hook them up to it. Rather than have a mechanical switch, I would prefer to always spool the output from a Sun. This would require some program to catch Macintosh output and send it on to the lpr daemon. If anybody has written such a program or has another alternative, I would be interested in knowing. Scott Anderson ihnp4!oddjob!kaos!sra ----------------------------- Date: Sun, 15 Dec 85 17:49:53 EST From: Barry Shein <bzs%bostonu.csnet@CSNET-RELAY.ARPA> Subject: Re: SUN TIP locking woes The bug described, being unable to remove the lock file at the end of a tip session due to lack of privs, is an old 4.2 problem. One straightforward solution we used here was to attack the sources and use flock() on the file rather than existence of the file as a method of locking. This allows the file to always exist (or not exist, it creates it if necessary) eliminates needing to clean things up across a crash (or some other tip abort.) If you do not have the sources I would suggest writing a little setuid C program which execs TIP (I would move the ucb tip to /usr/lib/tip or some such and call this thing /usr/ucb/tip so users won't even know what happened) and waits for completion retaining its setuid status (it can setuid(getruid()) as it execs.) On completion it can insure the lock file gets removed thus curing the immediate problem. No big deal. Oh yeah, if you use the same tty lines for TIP and UUCP you will have to consider fixing both (I believe they both use the same routine, ulockf() or some such) if you opt for the source fix. -Barry Shein, Boston University ----------------------------- Date: Sun, 15 Dec 85 19:35:25 EST From: Sid Stuart <linus!sid@mitre-bedford.ARPA> Organization: The MITRE Corp., Bedford, MA Subject: Only a FEW? First I would like to thank Bill for setting us straight on the Sun 2 versus Sun 3 compatibility problem. I was begining to wonder if anyone from Sun read this digest. Second I would like to comment on something I keep hearing from people at Sun, which Bill mentioned in his article also, I quote: "...Note that only a few old Sun-2/120s and Sun-2/100s and Sun-2/170s used 3Com controllers." Flame on What do you mean by few? From what I can see 100% of the 120's and 170's shipped to Mitre have 3Com boards. Would you please show us some figures showing systems shipped with 3Com boards versus systems shipped with Sun Ethernet boards to support your claim? What do you mean by old? I have worked on 10 different Sun 120's and 170's at MITRE, some that were shipped as late as January of this year and they ALL have 3Com boards in them. Even in this busness, we should wait a year before something is called old. When did Sun start shipping systems with Sun 2 ethernet boards? Flame off Boy, that felt good. sid at linus ----------------------------- Date: Sun, 15 Dec 85 13:46:03 EST From: snk%osiris.UUCP@harvard.ARPA (Steve Kahane) Subject: Re: Voice recognition on SUN I suggest you call Kurzweil Applied Intelligence - (617) 893 5151 snk ----------------------------- Date: Tue, 17 Dec 85 22:48:39 EST From: allegra!sjuvax!jss@ucbvax.berkeley.edu (J. Shapiro) Organization: Haverford College, Haverford, Pa. Subject: Re: multibus adaptor for the VME bus I am not a regular contributor, but I hope this will be useful: Someone had asked if there was a multibus adaptor for the VME bus. Talk to your Motorolla Rep. Motorolla has in fact built one, and he/she should be able to guide you to it. Jon Shapiro Jonathan S. Shapiro Haverford College ----------------------------- Date: Tue, 17 Dec 85 00:42:33 EST From: mark@markssun.cs.umd.edu (Mark Weiser) Subject: removing batteries Sun has apparently told everyone to remove the batteries from the clock on the SCSI board before the battery blows up. I have done this, but it is now real inconvenient each time I re-powerup my sun. I'd love to put the battery back. How serious is this problem, and what does anyone hear about a more permanent solution than disabling the battery clock? -mark ----------------------------- Date: Mon, 16 Dec 85 11:38 PST From: lerner@isi-vaxa.ARPA (Mitchell Lerner) Subject: Performance enhansements for Sun Unix? Does anyone know of past, present or future performance improvements and/or functional enhansements for Sun Unix. We recently installed 4.3 bsd and noticed very significant improvements in overall performance of the system. With the distribution came a document about the actual performance improvements and functional enhansements made by Berkley to their Unix and I might prove interesting reading to some of yall. In it they are specific about what they improved and how they improved it. Im wondering if Sun has or will benifit from similar improvments to their Kernel, drivers, libraries and Utilities, C Run-time Library, Csh, Signals and security. ----------------------------- Date: Mon, 16 Dec 85 17:25:14 EST From: mark@markssun.cs.umd.edu (Mark Weiser) Subject: bell in suntools? How can I ring the bell from suntools? Obvious things like sending ^g to /dev/console don't work. -mark ----------------------------- Date: Tue, 17 Dec 85 19:48:51 GMT From: Richard Tobin <richard%aiva.edinburgh.ac.uk@cs.ucl.ac.uk> Subject: bleep on sun Can anyone tell me of a way to produce an audible bell while running suntools? And can the internal speaker be persuaded to produce any other sounds? Richard Tobin, JANET: R.Tobin@uk.ac.edinburgh AI Applications Institute, ARPA: R.Tobin@uk.ac.edinburgh@ucl-cs.arpa Edinburgh University. UUCP: ...!ukc!edinburgh.ac.uk!R.Tobin ----------------------------- Date: Fri, 13 Dec 85 07:53:25 EST From: John D. Ramsdell <linus!ramsdell@mitre-bedford.ARPA> Organization: The MITRE Corp., Bedford, MA Subject: Dvorak keyboard in suntools I enjoy using a Dvorak keyboard when using standard UNIX. A Dvorak keyboard layout looks like: ! @ # $ % ^ & * ( ) [ + 1 2 3 4 5 6 7 8 9 0 ] = " < > P Y F G C R L ? ' , . p y f g c r l / A O E U I D H T N S _ RETURN a o e u i d h t n s - RETURN : Q J K X B M W V Z ; q j k x b m w v z When I log into UNIX, I create two processes that translate my tty input and output via pseudo terminals. All works fine except that sometimes ^S gets translated in a process that should not process it. My problem comes when I attempt to use suntools. Suntools gets keyboard input from /dev/console instead of from the tty of the process that started suntools. I think that is odd and our word processing software, which does its own windowing, does take its input from the tty that started the word processing program. The command "suntools -k /dev/tty" will crash the sun. Does anyone know how to specify ones own keyboard translation in software for use with suntools? For non-Dvorak users, this would be helpful so that you can map the "| \" into RETURN, making the keyboard much more reasonable to use. Is there something wrong with allowing user translation of keyboard input without modifications to the kernel? ----------------------------- Date: Tue, 17 Dec 85 00:40:48 EST From: mark@markssun.cs.umd.edu (Mark Weiser) Subject: sun-3 draws faster? I have heard rumors that the sun-3 has a faster screen. Anyone else heard this, and what does it mean? Of course, the sun-3 has a faster cpu an so will be be able to rasterop faster. Is there more than that? -mark ----------------------------- Date: Mon, 16 Dec 85 22:09:21 PST From: Pierre MacKay <MACKAY@WASHINGTON.ARPA> Subject: Loading up TeX on a Sun-3. In my capacity as Unix-TeX site coordinator I recently got a query from a Sun-3 user who was successful as far as INITEX and VIRTEX, (uppercase for emphasis only), but could not make undump work. (undump takes the header information off virtex and applies it to a core dump to create a pre-loaded program with a considerable amount of processing already complete.) The process will be familiar to anyone who has received our Unix distribution. Has anyone else had this trouble with Sun-3. The machine complains that there ``is no 68881 available'' or something similar. Is there a cure? does it require a modification of undump? Pierre MacKay ----------------------------- Date: Wed, 18 Dec 85 18:22:17 GMT From: Richard Tobin <richard%aiva.edinburgh.ac.uk@cs.ucl.ac.uk> Subject: ttysw_input I'm writing a tool which uses a panel subwindow and a tty tool subwindow, and I want to simulate input to the tty when a button is selected. The way to do this seems to be to use ttysw_input. When I try it, with a call like this: ttysw_input(tsw->ts_data, "*UP*", 4); the characters aren't actually received until a 'real' character is typed (for example, if I type ! the characters *UP*! all appear at once). Is this behaviour correct? I can get round the problem by using this instead: write(((struct ttysubwindow *)tsw->ts_data)->ttysw_pty, "*UP*", 4); but I thought the idea of ttysw_input was to avoid having to look at the internals of the ttysubwindow structure. Richard Tobin, JANET: R.Tobin@uk.ac.edinburgh AI Applications Institute, ARPA: R.Tobin@uk.ac.edinburgh@ucl-cs.arpa Edinburgh University. UUCP: ...!ukc!edinburgh.ac.uk!R.Tobin ----------------------------- Date: Thu, 12 Dec 85 07:28:20 CST From: fritz@ut-ngp.ARPA (Fritz Benedict) Subject: more memory for SUN-1/150U Has anyone successfully put 4 Mbytes on a SUN-1/150U? I have tried SUN supplied SUN-2 memory and gotten parity errors and crashes for my efforts. We have 2 Mbytes and desperately need 4. Are ther vendors selling P2 compatible memory boards for a SUN-1/150U? We might also be interested in trading SUN-2 memory for older, functional SUN-1 memory. Thanks for your help. Fritz Benedict (512)471-4461x448 uucp: {...noao,decvax,ut-sally}!utastro!fritz arpa: fritz@ut-ngp snail: Astronomy, U of Texas, Austin, TX 78712 ----------------------------- Date: Fri, 20 Dec 85 12:54:59 EST From: jas@mitre.ARPA (John Scott) Organization: The MITRE Corp., Washington, D.C. Subject: RS-232/EIA Flow control on SUN2s Help! I'm working on a printer driver which should (must) use EIA as flow control over a serial port. Before anyone suggests that I read the manual, let me say that the manual is unclear (that's an understatement, bet then what else is new) about whether and how EIA flow control is done on the serial ports. I should mention that the beast in question is a SUN 2 configured with the standard 2 serial ports (i.e. uses the zs device driver). I can already feel the heat of the flames so reply to me directly and save some network bandwidth. John Scott <scott @ mitre-gateway> Disclaimer: The views expressed herein do not necessarily represent those of any thinking being or management. ----------------------------- Date: Thu, 19 Dec 85 19:02:20 PST From: fluke!jeff@uw-beaver.arpa Subject: Public domain NFS daemon? Anybody have a public domain implementation of an NFS daemon which can run in USER MODE (some of these kernels around here, I just can't hack up). Yes, I know it would be slow (lots of context switching), but it's intended for light-duty usage, where the basic functionality is much more important than the efficiency. BTW, this is to run on a variety of machines which run something approximating 4.2BSD. Jeff Stearns John Fluke Mfg. Co, Inc. (206) 356-5064 {uw-beaver, decvax!microsoft, ucbvax!lbl-csam, allegra, sun}!fluke!jeff ----------------------------- Date: Sun, 22 Dec 85 14:21:49 EST From: mark@markssun.cs.umd.edu (Mark Weiser) Subject: Screen saver? Anyone out there have a screensaver suntool? This would be something which makes the screen go blank (usually just by covering it with a window), thereby illuminating no phosphors and extending the screen life. Screensavers usually also have little moving objects on them so they can be told apart from a dead machine, and they generally go away at the first keystroke. Any for the suns? -mark ----------------------------- Date: Sun, 22 Dec 85 22:18:23 CST From: Mike Caplinger <mike@bellcore.ARPA> Subject: Dhrystone benchmark - DN330 vs. Sun-3 I just got my 68020 upgrade, taking my Apollo up to a DN330. One of the first things I did was run the Dhrystone benchmark on it. I got a number somewhere around 1700 - about the speed of a VAX/11-780. Fine and good. Now I see some Sun-3 results that run in the 3000s! Are these for real? Would someone explain to me why the Sun apparently runs almost twice as fast as a DN330 with an identical processor? Mike Caplinger mike@bellcore.arpa ihnp4!bambi!mike ----------------------------- Date: Sun, 22 Dec 85 16:35:59 EST From: ihnp4!islenet!bigtuna!brianm@seismo.ARPA Subject: Need help selecting Apollo vs. Sun and Relational DBMS for hospitals I'm developing a patient medical record system for a community-based hospital built around Unix workstations. The medical record system follow's Weed's POMR medical database, and will be implemented by a relational DBMS. I'd like opinions as to whether I should use Sun or Apollo workstations, and why, and which relational DBMS to use. The system will tie together all components of a hospital which may enter information into a patients's medical record, including the laboratory, radiology, nuclear medicine, nursing stations, admissions, etc. The configuration requires a minimum of 30 nodes; the relational DBMS must support a least 12 different views of the medical database, which will have 68+ relations. I plan to embed the DBMS within a mouse/icon interface, to decrease the amount of typing necessary. Voice input would be a Major Plus. Also, conversion (via a frame grabber?) of radiographic images (X-rays, CT-scans, NMR images, PET scans, SCA, etc.) into a format suitable for on-line storage and display will eventually be required--is it feasable now? If so, do you know anyone who does it? AT&T makes a board for the IBM-AT (ugh) that converts images from VCRS, laser discs, and video cameras into 512x512 data representations--is there anything similar for the Sun or Apollo? Any and all replies are appreciated--from biased opinions to actual unbiased studies. Thanks in advance, Brian Martin University of Hawaii School of Medicine UUCP: ..!{dual|ihnp4|vortex}!islenet!bigtuna!brianm US Mail: 3420-A Hinahina Street Honolulu, Hawaii 96816 Phone: (808) 735-5661 ----------------------------- End of SUN-Spots Digest ***********************