Sun-Spots-Request@RICE.EDU (William LeFebvre) (07/07/88)
SUN-SPOTS DIGEST Thursday, 7 July 1988 Volume 6 : Issue 131 Today's Topics: Re: Monitor eyestrain Re: 8-bit shelltool DB-15 slide latch Expansion Memory for Sun3/xxx's Problems with mixed language programming Questions about 3/260 memory expansion postscript printers under 4.0? Clock Synchronization for the Suns? conversion to sun raster format? 3rd party drives? Ringing the bell (source) Send contributions to: sun-spots@rice.edu Send subscription add/delete requests to: sun-spots-request@rice.edu Bitnet readers can subscribe directly with the CMS command: TELL LISTSERV AT RICE SUBSCRIBE SUNSPOTS My Full Name Recent backissues are available via anonymous FTP from "titan.rice.edu". For volume X, issue Y, "get sun-spots/vXnY". They are also accessible through the archive server: mail the request "send sun-spots vXnY" to "archive-server@rice.edu" or mail the word "help" to the same address for more information. ---------------------------------------------------------------------- Date: 24 Jun 88 22:56:46 GMT From: novavax!proxftl!bill@bikini.cis.ufl.edu (T. William Wells) Subject: Re: Monitor eyestrain Reference: v6n105 I have 20/200 vision and have had to contend with all kinds of problems with monitors. Here are some suggestions; some of these I have used, others just heard about. Good luck, and d**m flourescent office lights. 1) Make a shield of some material, attach it to various parts of the monitor to cut down on glare. This also improves contrast by reducing the amount of light incident on the screen. 2) Cut down on the flourescent lighting in your office. The amount of light in the average office is far more than is needed for most work. You can (or your office maintenance workers can) remove some of the bulbs from your fixtures to cut down the light level. Another possibility is to rig a shield of some translucent material over the diffusers for your lights. A judicious combination of 1) and 2) can almost completely eliminate screen glare. 3) Project incandescent light from a portable lamp onto your screen. While this decreases the contrast somewhat, it can help cut down on the flicker effect from flourescents. 4) Keep a close eye on your viewing distance. The distance you keep between your eyes and the screen is a compromise between various physical factors and the optimal seeing distance. Things like office furniture and the length of your arms might be conspiring to place you at an awkward distance for reading the screen. 5) Go back to white on black, after doing any of 1) - 3). The bright background can do horrible things to one's ability to focus properly. You might also look into different fonts for the screen. Fonts ideal for black on white are often not very good for white on black. ------------------------------ Date: Fri, 24 Jun 88 17:33:35 PDT From: rprobst@sun.com (Richard Probst) Subject: Re: 8-bit shelltool Install 4.0 and you will have a version of shelltool that displays 8-bit characters (but you will need a font with 256 characters). Also 8-bits wide in 4.0: cmdtool and textedit. --Richard Probst (rprobst@sun.com) SunView Engineering Manager ------------------------------ Date: Fri, 24 Jun 88 11:41:50 EDT From: gotham!marine!fish@sun.com (Lorenz Fish) Subject: DB-15 slide latch Those latches are abysmal!! With the thickness and weight of the cables involved and the light weight of the latch and minimal (far from positive) retaining capability of the latch any movement of the machine or the cable will result in a disconnect. ANY movement. This is a matter of repeated frustration at our site and I would like to hear about alternatives. Whenever there is an ethernet problem message all hands head for the machine room to check the DELNI and to the backs of machines to see what got kicked out, or just fell out because the temperature changed and the cable tension shifted. In short the slide latch is a toy and never assures protection against even the most feeble adverse event. Fortunately the back of the Cabletron MT-800 has a frame you can add to help guide the cables and to which the tranceiver cables can be bound with cable ties. Unfortunately no such aid is provided at the backs of most machines and we have had to resort to exotic techniques to prop tranceiver cables up using boxes and duct tape and just-so positioning of machines. A little screw-down housing ala RS-232 connectors would be very welcome. Please feel free to use this message to promote progress in this regard and please post a summary to sun-spots when all the votes are in. Thanks! Lorenz Fish Technical Analysis Group Marine Midland Bank Manhattan UUCP: ...sun!gotham!marine!fish ------------------------------ Date: 24 Jun 88 03:38:02 GMT From: vsi1!tim@uunet.uu.net (Tim Richardson ) Subject: Expansion Memory for Sun3/xxx's I keep seeing people on the net refer to Clearpoint as a good source of expansion memory for their Sun's. I have nothing against Clearpoint but there is another vendor with a quality product. This vendor is PARITY SYSTEMS, located in Los Gatos, CA. Vicom evaulated both the Clearpoint and Parity products and ended up with the Parity memory board. We did this because Vicom is a CPU board OEM customer of Sun and we integrate their CPU's into our image processor. Many of our customers want additional memory and often ask us if we can recommend memory boards other than the one's Sun sells. The Parity Systems board was: 1. Cheaper 2. More robustly constructed 3. Less marginally timed (i.e. a more robust design) 4. AND, they have a memory board into which you can mount your SCSIbus host adaptor, thusly taking NO ADDITIONAL slots to expand your workstations' memory! I have no affilation with Parity systems, but they must not be marketing their product as effectively as Clearpoint so here I am chiming in with my 50 cents worth (inflation adjusted). Tim Richardson VP Engineering VICOM Systems Inc. sun!pyramid----\ 2520 Junction Ave. uunet!ubvax-----!vsi1!tim San Jose, CA 95134 ucbvax!tolerant/ +1-408-432-8660 ------------------------------ Date: Fri, 24 Jun 88 17:41:55 EDT From: rmckinst@hstbme.mit.edu (Robert McKinstry) Subject: Problems with mixed language programming I am having trouble linking f77 subroutines with C programs when using mathematical functions on the SUN-4. I get: ***************** ld error message *************** Linking: Undefined: _class _ir_class_ when using the makefile: ***************** makefile *********************** SOURCE = cont.o cidf_read.o contsubs.o lm.o HEADER = cont.h LIBS = -lc -lm -lF77 -lI77 -lU77 CCOPTS = -g -fsingle cont: $(SOURCE) $(HEADER) cc $(CCOPTS) $(SOURCE) $(LIBS) -o cont cont.o: $(HEADER) cc $(CCOPTS) -c -fsingle cont.c cidf_read.o: $(HEADER) cc $(CCOPTS) -c -fsingle cidf_read.c contsubs.o: f77 -g -c contsubs.f lm.o: f77 -g -c lm.f ************************************************** Man class informs me that these are IEEE floatingpoint calls and that I should be including ****************** from "man class" ************** SYNOPSIS #include <floatingpoint.h> #include <math.h> FILES /usr/include/floatingpoint.h /usr/include/math.h /usr/include/values.h /usr/lib/libm.a ************************************************* When I include floatingpoint.h and math.h, cc tells me that cc -g -c -fsingle cont.c "/usr/include/sys/ieeefp.h", line 20: redeclaration of fp_direction_type "/usr/include/sys/ieeefp.h", line 21: redeclaration of fp_nearest "/usr/include/sys/ieeefp.h", line 22: redeclaration of fp_tozero "/usr/include/sys/ieeefp.h", line 23: redeclaration of fp_positive "/usr/include/sys/ieeefp.h", line 25: redeclaration of fp_negative "/usr/include/sys/ieeefp.h", line 38: redeclaration of fp_precision_type "/usr/include/sys/ieeefp.h", line 39: redeclaration of fp_extended "/usr/include/sys/ieeefp.h", line 40: redeclaration of fp_single "/usr/include/sys/ieeefp.h", line 41: redeclaration of fp_double "/usr/include/sys/ieeefp.h", line 43: redeclaration of fp_precision_3 "/usr/include/sys/ieeefp.h", line 46: redeclaration of fp_exception_type "/usr/include/sys/ieeefp.h", line 47: redeclaration of fp_inexact "/usr/include/sys/ieeefp.h", line 48: redeclaration of fp_division "/usr/include/sys/ieeefp.h", line 49: redeclaration of fp_underflow "/usr/include/sys/ieeefp.h", line 50: redeclaration of fp_overflow "/usr/include/sys/ieeefp.h", line 52: redeclaration of fp_invalid "/usr/include/sys/ieeefp.h", line 55: redeclaration of fp_class_type "/usr/include/sys/ieeefp.h", line 56: redeclaration of fp_zero "/usr/include/sys/ieeefp.h", line 57: redeclaration of fp_subnormal "/usr/include/sys/ieeefp.h", line 58: redeclaration of fp_normal "/usr/include/sys/ieeefp.h", line 59: redeclaration of fp_infinity "/usr/include/sys/ieeefp.h", line 60: redeclaration of fp_quiet "/usr/include/sys/ieeefp.h", line 62: redeclaration of fp_signaling "/usr/include/floatingpoint.h", line 25: illegal type combination "/usr/include/floatingpoint.h", line 26: illegal type combination "/usr/include/floatingpoint.h", line 26: syntax error at or near symbol [ "/usr/include/floatingpoint.h", line 30: illegal type combination "/usr/include/floatingpoint.h", line 36: illegal type combination "/usr/include/floatingpoint.h", line 38: syntax error at or near type word "sigfpe_handler_type" "/usr/include/floatingpoint.h", line 87: illegal type combination "/usr/include/floatingpoint.h", line 87: syntax error at or near symbol [ "/usr/include/floatingpoint.h", line 87: fatal error: too many errors *** Error code 1 I don't include ieeefp.h explicitly, floatingpoint.h does that for me. What am I doing incorrectly? If anyone has experience linking f77 subroutines to C programs please let me know how successful you have been. Thanks. ...bob ------------------------------ Date: Thu, 23 Jun 88 15:42:10 BST From: Andrew Watson <mcvax!harlqn.co.uk!andrew@uunet.uu.net> Subject: Questions about 3/260 memory expansion I'm investigating adding more memory to our twin 3/260s, regarding which I'd like to pose a couple of questions: 1) Suppliers. I've been talking to Clearpoint, whose SNX2RAM seems to be competitively priced and has the advantage that the cards (including 16MB & 32MB types) can be used in both 3/2xx & 4/2xx boxes, useful against the day that we upgrade. Does anyone have any experience with these? What other suppliers should I consider (bearing in mind that I'm on this side of the pond)? 2) How much memory can I get into a Sun 3? From the fact that the Sun 4 upgrade is a CPU board swap, I imagine that the bus can cope with 128MB. Sun only support configurations up to 32MB, and have implied to me that this is a Sun 3 CPU limit. Others tell me it's just that you can put only 4 x 8MB Sun memory boards in one 3/260. Yet other sources tell me that the machine has 4MB on the CPU and uses up to 28MB from the bus. What's the story? What about power supply load considerations? All advice/information gratefully received. Our news link is a little erratic, so I'd appreciate it if people post replies to me & I'll summarise. Andrew Watson, Harlequin Limited, andrew@uk.co.harlqn Barrington Hall, Barrington, Tel: +44 223 872522 Cambridge CB2 5RG, UK Fax: +44 223 872519 ------------------------------ Date: Thu, 23 Jun 88 09:16:55 HST From: Neil Frazer <neil@akolea.hig.hawaii.edu> Subject: postscript printers under 4.0? Under SUNOS 4.0, troff output is in postscript, rather than CAT/4 commands. I have a postscript printer (QMS-PS810) on ttyb of a SUN4-260. Should I put anything in my printcap file, or should I just put troff output in an intermediate file and then cat it to /dev/ttyb ? I would really like to run the printer in an emulator mode, with an appropriate printcap entry, and then tell it to go postscript (without hitting buttons on the printer and powering it up again) whenever I send it troff output. Documentation that came with the printer is wonderful, if you have a Mac or a PC, but hasn't a word about UNIX. Sun 4.0 documentation (e.g. printcap(5) man entry) is almost as useless. Sorry to bother you guys with this kind of low level stuff. Any suggestions would be appreciated. Especially those that don't involve buying something. Neil Frazer (neil@akolea.hig.hawaii.edu) ------------------------------ Date: Fri, 24 Jun 88 08:45:33 EDT From: beck@cs.cornell.edu (Micah Beck) Subject: Clock Synchronization for the Suns? Has anyone successfully ported the Berkeley Clock Synchronization daemon to a Sun? Failing that, is any good clock syncronization software available for Suns? How can I get it? Micah Beck beck@cs.cornell.edu Cornell Computer Science Dept. ------------------------------ Date: Fri, 24 Jun 88 09:47:14 EDT From: Scott Thomas <13501SGT@MSU.BITNET> Subject: conversion to sun raster format? We are going to be purchasing a color scanner that will be hooked to a PC or more probably a Macintosh II Computer. It scans into a TIFF format as well as PIC format. I am looking for a conversion program that will convert one of these formats to SUN-RASTER format. Have you heard of anything? Scott Thomas. Systems Analyst Michigan State University 517-353-1800 ------------------------------ Date: 27 Jun 88 13:11:22 GMT From: bennett@beach.cis.ufl.edu (Paul Bennett) Subject: 3rd party drives? We are looking at finding a source for drives for our Sun 3/50-60 machines. I would appreciate it if someone could provide us with company names/phone numbers, quality of service stories, reliablity experience. As I recall, there are several 3rd party vendors. Any assistance will be appreciated. Paul R. Bennett ------------------------------ Date: Thu, 23 Jun 88 12:57:07 PDT From: Don Dwiggins <pyramid!dbase!ashtate!atsun!dwiggins@sun.com> Subject: Ringing the bell (source) A while back a message appeared in comp.windows.news giving a couple of routines for ringing the console bell; they work even if the "visual bell" is turned on, so you can get "beeped" even if you're not watching the screen. I wrapped a little main program around them, which folks may find useful. Try "sunbell 50 200 50 100 50 50 50 200 50 400 50 200 50". Don Dwiggins {scgvaxd,crash}!ashtate!dwiggins ----------------------------------------------------------------------- /* * Sound the Sun console bell in a particular pattern. Its arguments are * taken as numbers of milliseconds to have the bell on and off, alternately, * e.g. "sunbell 200 100 200" will give two .2 second beeps separated by * .1 second. If the number of arguments is even, the last argument has * no effect, since the program concludes by turning the bell off in any * case */ #include <sys/time.h> #include <sys/types.h> #include <sys/ioctl.h> #include <sundev/kbio.h> #include <sundev/kbd.h> #define NULL ((int *) 0) long atol(); turn_on_bell() { int bell, bell_on = KBD_CMD_BELL; bell = open("/dev/kbd",1); ioctl(bell, KIOCCMD, &bell_on); close(bell); } turn_off_bell() { int bell, bell_off = KBD_CMD_NOBELL; bell = open("/dev/kbd",1); ioctl(bell, KIOCCMD, &bell_off); close(bell); } main(argc, argv) int argc; char *argv[]; { struct timeval howLong; int i; howLong.tv_sec=0; for (i=1; i<argc; i++) { if ((howLong.tv_usec = atol(argv[i])*1000) <= 0L) { char *errstr = "sunbell: Bad argument; giving up\n"; write(2, errstr, strlen(errstr)); exit(1); } if (i%2 == 0) turn_off_bell(); else turn_on_bell(); select(0, NULL, NULL, NULL, &howLong); } turn_off_bell(); exit(0); } ------------------------------ End of SUN-Spots Digest ***********************