Date: Wed, 24 Mar 99 22:46:09 EST From: Dwight McKay (The Moderator) Reply-To: Suns-at-Home@net-kitchen.com Subject: Suns-at-Home Digest V12 #9 To: Suns-at-Home-List Suns-at-Home Digest Wed, 24 Mar 99 Volume 12 : Issue 9 Today's Topics: for my next trick -- sawing a type 4 keyboard in half getting HP Laserjet III to work on SS20 parallel port key performance figures PCI Video cards for UltraAXi? Sparcs expensive? Well.... SUN LX screen freezing problem Sun Roadrunners (2) for sale or trade Suns-at-Home Digest V12 #8 (6 msgs) using a SunSwift adapter on an IPX +--------------------------------------------------------------------------+ | Submissions: suns-at-home@net-kitchen.com | | Requests: suns-at-home-request@net-kitchen.com | | WWW Archive access: http://www.net-kitchen.com/~sah | +--------------------------------------------------------------------------+ ---------------------------------------------------------------------- Date: Tue, 16 Mar 99 14:34:44 PST From: lakin@pgc.com (Fred Lakin) Subject: for my next trick -- sawing a type 4 keyboard in half To: suns-at-home@tigger.net-kitchen.com Emacs elbow has finally gotten to me and I think one of those tilty split kbds would help. Before buying an alternative PC kbd and then an interface adaptor to hook it to my SS20, seems simpler to just try sawing one of my old type 4s in half. Anyone every tried this? My plan is to number the 40 odd solder lines on each side of the cutline (between the 6-7 T-Y G-H B-N keys), then make the cut, and then solder 40 18" pieces of wire on each side to reconnect the circuits. And, oh yeah, deal with all the ICs above the number keys ... hmmm, may be more complicated than I thought. But nonetheless will probably give it a try. So any advice or suggestions welcome. tnx, -f [ Didn't I see David Copperfield do this trick? Or was it Penn and ] [ Teller? ;-) --ddm ] - ------------------------------ Date: Tue, 23 Mar 99 14:46:21 PST From: lakin@pgc.com (Fred Lakin) Subject: getting HP Laserjet III to work on SS20 parallel port To: suns-at-home@tigger.net-kitchen.com I have an old HP Laserjet III hanging off a serial port on my SunOS 4.1.4 SS20 (Integrix clone) and it works fine. Now I am trying to shift the printer to the parallel port, but no luck so far ... print jobs just hang, never get sent to the printer. 1. Got the proper cable I think: older Centronix 36 to keyed slimline 26 variation of DIN. 2. I used the menu interface on the printer to switch from serial to parallel. 3. I edited /etc/printcap and changed :lp=/dev/ttyb: to :lp=/dev/bpp0: (and checked to see that /dev/bpp0 already exists) 4. Then booted the machine. And then print jobs just hang. After waiting for several minutes, I gave up and lprmed the jobs. Oddly enough, however when I then halted the machine, on the way down it informed me that CAUTION: process(es) wouldn't die The print jobs? Dunno, they were gone after I rebooted. Also tried printing as root, no differnce. And, poking around /dev, I note that bbp0 has not been touched by all my experiments: crw-rw-rw- 1 root 107, 0 Aug 3 1996 bpp0 crw-rw-rw- 1 root 107, 1 Aug 3 1996 bpp1 ... crw-rw-rw- 1 root 12, 0 Jul 3 1997 ttya crw-rw-rw- 1 root 12, 1 Mar 15 17:34 ttyb So, what did I forget? Or will this not work at all? tnx, -f PS Here is the full princap file that works fine now when the printer is on a serial port: lp|hp|HP|laserjet|HP LaserJet III with Filter:\ :lp=/dev/ttyb:af=/var/adm/lw.acct:\ :sd=/var/spool/lw:lf=/var/spool/lw/lw-log:\ :br#19200:rw:fc#0000374:fs#0000003:xc#0:xs#0040040:mx#0:sf:sb:sh:\ - ------------------------------ Date: Sun, 14 Mar 1999 07:44:12 -0600 From: Reagen Ward Subject: key performance figures To: suns-at-home@tigger.net-kitchen.com >From what I've been told, RC5 cracking is usually better on Intel / Motorola platforms because the Rotate Left instruction is in silicon, rather than microcode. As a result, an HP PA-RISC 8500 CPU will be slower at RC5 cracking than a high-end PII, and a SS5 will be slower than a 486. Try running a better benchmark, like bytebench (still not great) like, and you'll see the SS5 do much better. Reagen Ward - ------------------------------ Date: Sat, 20 Mar 1999 14:48:09 -0500 From: Brian Neal Subject: PCI Video cards for UltraAXi? To: Suns-at-Home@tigger.net-kitchen.com I'm looking at building a PCI-based UltraSPARC system with an UltraAXi motherboard. I'm interested in what graphics options are available for PCI-based Sun boxes. Sun only lists the Permedia 2 (Techsource Raptor) and the ATI Rage IIc (Charger). I know the Creator series is also available, but what about any others? Matrox, Diamond? I'm looking for something that will do well in Orcad and AutoCAD, but also Quake II. Any suggestions? Thanks, -Brian Neal - ------------------------------ Date: Sun, 14 Mar 1999 00:13:53 -0500 From: "Sheldon T. Hall" <76701.103@compuserve.com> Subject: Sparcs expensive? Well.... To: Blind.Copy.Receiver@compuserve.com Folks- Recently, "Cruz0025 @ The U", writing about, oh, I don't know, said ... >Yet from all of the advertisements that I've been reading I get the distinct >impression that Sparc computers are extremely overpriced. Well, maybe. Certainly, you can buy a generic PC clone for a _lot_ less than a new Sparc-whatever that has roughly equivalent "MFLOPs" or other numbers. On the other hand, most of us here are buying _new_ Sparcs. We're not even buying very recent Sparcs. Some of us aren't buying them at all. We rescue them from abusive homes. My IPX, for example, probably cost about $10,000 in 1992, the same year I did buy a generic '486-66 PC clone for $2,200. Now my "loaded" IPX is worth about $400 ... and so's the '486. So, in depreciation terms, the Sparc was a _lot_ more expensive. But not for me. I paid actual money for the PC in '92, but a friend gave me the IPX after he pulled it out of a dumpster in '96. It was just the box; no memory, but it had SunOS 4.1.3 on a 424 MB HD, and once I put some memory in it, it booted right up. I've since acquired a keyboard, monitor, CD-ROM, and tape drive (all very cheap), and I've been using it to renew and expand my Unix skills, as well as to learn PERL and whatnot. Now, I could run Linux or Solaris x86 on that '486, and it, instead of the IPX, could be my DNS server, NTP server, SMTP server, Intranet Web server, and general play-toy. I'm sure it would do fine, but it just wouldn't have the same panache. I'd miss the cold aluminium optical mouse pad, for one thing. Besides, I do PCs for a living. The Sparc's my hobby, and no gentleman allows crass commercial considerations to govern his hobby. -Shel - ------------------------------ Date: Sun, 14 Mar 1999 12:19:36 +0000 (GMT) From: Richard Skelton Subject: SUN LX screen freezing problem To: suns-at-home@tigger.net-kitchen.com Hi Managers, In the last week my SparcStation LX display has frozen three time since I applied the latest recommended patches. I can remotely login to the LX and can see the Xsun is in a spin from top:- 139 processes: 137 sleeping, 1 running, 1 on cpu CPU states: 0.0% idle, 93.2% user, 6.8% kernel, 0.0% iowait, 0.0% swap Memory: 64M real, 984K free, 47M swap in use, 87M swap free PID USERNAME THR PRI NICE SIZE RES STATE TIME CPU COMMAND 344 root 1 15 0 10M 1816K run 93:42 87.26% Xsun 2726 root 1 10 0 2056K 1432K cpu 0:02 8.94% top If I kill Xsun the machine is totally frozen and I have to power cycle the system. Do I have a GX hardware fault or do I need the latest Xsun patch . I have all of the recommended PATCHES from the "Solaris 2.6 Patch Report Update as of 17/Feb/99" but not many of the security patches. === [4a] MISSING SECURITY PATCHES ======================================= This is the list of security patches which are not currently installed. ========================================================================= PatchID Rec Sec Description --------- --- --- ----------- 105802-07 * OpenWindows 3.6: ToolTalk patch 106112-03 * CDE 1.2: dtfile patch 106123-04 * SunOS 5.6: sgml patch 106415-02 * OpenWindows 3.6: xdm patch 106522-01 * SunOS 5.6: /usr/bin/ftp patch 106569-01 * SunOS 5.6: libauth.a & libauth.so.1 patch 106648-01 * OpenWindows 3.6: libce suid/sgid security fix 106649-01 * OpenWindows 3.6: libdeskset patch 106650-03 * OpenWindows 3.6: mailtool attachment security patch 106834-01 * SunOS 5.6: cp/ln/mv patch 106894-01 * SunOS 5.6: /usr/bin/uux patch === [4b] SECURITY PATCHES REQUIRING UPDATES ==================== This is the list of security patches which need to be updated. ================================================================ PatchID Cur Rec Sec Description --------- --- --- --- ----------- 105338-16 -14 * CDE 1.2: dtmail patch 105633-17 -06 * OpenWindows 3.6: Xsun patch 105667-02 -01 * SunOS 5.6: /usr/bin/rdist patch 106027-03 -01 * CDE 1.2: dtsession: patch for screenlock I will summaries my findings. -- Cheers Richard. = Richard Skelton | e-mail : Richard.Skelton@brake.demon.co.uk WWW : http://www.brake.demon.co.uk/ - ------------------------------ Date: Sun, 14 Mar 1999 08:18:04 -0500 From: db_computers Subject: Sun Roadrunners (2) for sale or trade To: Suns-at-home I have a pair of Sun Roadrunners (386i's) for sale, one is a 150, the other a 250. They are both look and are working just fine, and the pair comes with complete original documentation (4 binders) and the complete OS on diskettes. (4.1.1) I believe they have 8 and 16MB respectively, and 130 and 350MB drives. I also have the video card and cable for one of them, although I've just been using them via a dumb tube and telnet. Is there any interest in these? I'd rather sell them as a pair due to only having 1 set of docs and diskettes. Drop me a line! David david_barber@pobox.com PS - I'm located near London Ontario. - ------------------------------ Date: Sat, 13 Mar 1999 20:23:24 -0900 (AKST) From: "Christopher E. Brown" Subject: Suns-at-Home Digest V12 #8 On Sat, 13 Mar 1999, Dwight McKay wrote: > Date: Thu, 4 Mar 1999 10:24:29 -0500 (EST) > From: Rick Leir EPS > Subject: kkey performance figures > To: Dwight McKay > > This surprises me, I thought a SS5 was faster than a 486 , even for purely > integer loads. What OS? Was it paging wildly? What other factors? > cheers -- Rick Jumping in at the wrong point here, but if you are refering to the RC5 chal kkeys you are badly shorting the sparc. Comparing apples to oranges again, my current box does 2.2Mkeys/sec. :) >From the rc5 FAQ Why are Intel and PowerPC computers so much faster than other platforms? Integral to the mathematics of the RC5 algorithm are 32-bit rotate operations. For whatever reason, the designers of the x86 and the PowerPC architectures decided to implement the rotate function as a hardware instruction. Many other CPUs do not have built-in hardware rotate instructions and must emulate the operation by (at the very least) two shifts and a logical OR. This handicap is why many non-Intel and non-PowerPC computers run RC5 slower than one might expect based on real-world benchmarks. It is also the main reason why the RC5 client is a poor benchmark to use in determining the speed or performance of a particular CPU. - ------------------------------ Date: Sun, 14 Mar 1999 10:17:19 +0000 From: Bob Hoekstra Subject: Suns-at-Home Digest V12 #8 To: Suns-at-Home@tigger.net-kitchen.com, "Cruz0025 @ The U" wrote: > ... > I just inherited a SparcStation II+ clone and I would like to upgrade to > Solaris 7.0 or preferably Linux... I'm running Solaris 2.6 on my SPARC2 (upgraded with a Weitek PowerUP) and this is stretching the machine a little. Unless you have a good reason to use Solaris, I'd recommend Linux or NetBSD or SunOS 4 on a SS2 or clone. All these OSses are a slightly lighter load. Oh - if you have SunOS 4 on the machine, consider just leaving it -- it isn't necessarily un "upgrade" to go to Solaris 2. > ... In my endeavers to get to know this new > piece of sophisticated albeit antiquated equipment I've started pondering if > it is worth it. True, I like the big 16 inch monitor and the 2 big SCSI > drives I inherited and I like Linux and would like to experiment with it. No. 16 inch is small. 21 inch is big. And unless the disks have been changed, they should not be "big" either (in form factor or capacity). > Yet from all of the advertisements that I've been reading I get the distinct > impression that Sparc computers are extremely overpriced... Suns are expensive, but overpriced? I don't think so. PCs are built to a price: only the cheapest ever get sold, thus everything is made and assembled as cheaply as possible. Recently we have been seeing machines that are long lasting as well, but the older machines suffered from the cost cutting. Sun (and other manufacturers, e.g. Silicon Graphics, HP, Sequent) didn't suffer from quite the same pressures. In fact, somewhat the opposite, as their clients wanted machines that would run for months/years without problems or needing repairs. Some of these machines will have been up and running for several years without a reboot, never mind a repair. This kind of quality does not come cheap. > ...I just went to the site (www.memoryx.com thanx for the link BTW) to check up > on prices since the boards on Cycle-something -- another SS2 upgrade > manufacturer, were priced at around $1000 plus new. I figured that a > workstation from the early '90s ought to be around $150 or even less -- the > price was ~$400! Given what you've said about being 'slow as honey' and > ... Bear in mind that these machines really date from '386 days - theyu predate the '486, so the comparison is not strictly fair. Of course, no-one in their right mind would consider up-grading a '386 (or a '486 for that matter) so the fact that you were willing to consider this must tell you something. Slow? of course they are. But just try running Windows NT4 on a 80386 machine sold with 4MB RAM (i.e. high-spec) and see how slow that is. > ... why > would anyone spend such an enormous amount of cash? > ... Read the above again please. > ... I just picked up a > Pentium socket 4 mb less than 2 months ago for $10!! And if that wasn't > enough, I fried the BIOS and went to pick up another one and the vendor gave > me a used replacement for FREE! > ... Yeah. And when a client upgraded their 4/490 to a 4/690, I managed to get the boards as a no-cost swap for those in my 3/160 (Note to all readers: this works pretty well). This machine never got switched off in my garage (all those fanse made it too noisy for the house) and I'd still be running it if the electricity bill wasn't quite so high. My current SS2 (got it for free) made it superfluous, so I sold it. You will not be writing like this about your Pentium machine in 10 years time. > ... > I'm beginning to think that I would be better off salvaging what I can from > the parts while I'm still ahead... > ... Suit yourself. What about doing both? Get your own network! Run Solaris 2, SunOS 4, Linux (Debian), BeOS, and (if you insist) Windows 98 or NT at home, all communicating with each other :-) -- -----BEGIN GEEK CODE BLOCK----- Version: 3.1 GO/! d- s++:+ a+ C++(++++) US+++$ P+ L+ E--- W++ N++ w--- O- V- PS+ PE- Y+ PGP- t+ 5++ X+ R* tv+ b+ DI++ D G e(*) h++/-- r+++ y? ------END GEEK CODE BLOCK------ ----------------------------------------------------- Bob Hoekstra: APL & Unix Consultant Tele: +44 (0)1483 771028 http://www.khamsin.demon.co.uk Home email: Bob.Hoekstra@khamsin.demon.co.uk ----------------------------------------------------- - ------------------------------ Date: Sun, 14 Mar 1999 13:10:05 -0500 (EST) From: woods@most.weird.com (Greg A. Woods) Subject: Suns-at-Home Digest V12 #8 To: Suns-at-Home@tigger.net-kitchen.com > Date: Thu, 4 Mar 1999 00:20:34 -0800 (PST) > From: davras > Subject: Suns-at-Home Digest V12 #7 > To: Suns-at-Home@tigger.net-kitchen.com > > The scsi bus length must be 10ft or shorter. All 50 pins must be used. > Do not use an active terminator. Sun machines do not supply termination > power and expect none to be supplied. You'll know if you do, because it > will blow the termination fuse (if there is one). The last device on the > bus should have a passive terminator (dime a dozen, use a sun terminator > if possible). If you don't terminate, don't run the bus longer than 2ft. I'm afraid I can't approve of any of that advice. While it might work in a given situation it drastically violates the SCSI electrical specifications, which at least for the most common single-ended SCSI implementations are very unforgiving in the first place. Sun SCSI is a bit unforgiving at times, though often it works far more reliably and fails far more "reliably" than any PC-based SCSI sollutions. I.e. when it's marginal it often decides not to work at all instead of working partially with poor performance or even giving the opportunity for data corruption. The maximum cumulative length of a SCSI-2 bus, using proper cabling, is 6.0m, or about 19 ft. I've never actually tested a Sun of any kind with a bus this long though.... Note that the cable requirements for SCSI-2 and SCSI-2-FAST are much different, as are the connector requirements. (Obviously "wide" devices also have different cable and connector requirements too! ;-) You *do* want to use an active terminator on any SCSI-2 bus. Every real Sun(tm) HD-50 terminator I have here is an active terminator (and has the little LED on the end so you can see that termination power is available when the bus is connected). Not using active termination will *drastically* reduce the reliability of your SCSI bus, especially if you have any "fast" devices. As the standard says: "Any SCSI device may supply terminator power. Interface error rates are lower if the termination voltage is maintained at the extreme ends of the cable." This means you might want to ensure that the device at the end of the bus is also supplying TERMPWR, and indeed as you start to reach the maximum cable lengths this becomes increasingly important, especially if the terminator power line in the cable is only the minimum 28 AWG copper. If you don't terminate you shouldn't try to run the bus any more than four inches (i.e. 0.1m) -- that's the maximum length of a "stub", and is what you'd use on an internal drive, for example. Luckily most modern SCSI-2 and newer disks will do active termination of the bus at the request of an on-board jumper, and if the cable from that last disk on your bus to the second external connector on the disk chassis is less than "stub" length (i.e. < 0.1m) then you can probably get away with enabling termination on that last disk and not using an external active terminator (eg. if you don't have one). -- Greg A. Woods +1 416 218-0098 VE3TCP Planix, Inc. ; Secrets of the Weird - ------------------------------ Date: Sun, 14 Mar 1999 13:35:41 -0500 (EST) From: woods@most.weird.com (Greg A. Woods) Subject: Suns-at-Home Digest V12 #8 To: Suns-at-Home@tigger.net-kitchen.com > Date: Sun, 7 Mar 1999 01:44:17 -0600 > From: "Cruz0025 @ The U" > Subject: Suns-at-Home Digest V12 #7 > To: > > [[ someone, presumably , wrote: ]] > > > > Using the distributed.net client for SPARC/Solaris, the SS5 gives a rate > > of 30 kkeys/sec, while my slowest PC (a 100 MHz 486) gives a 90 kkeys/sec > > rate and the fastest (a 300 MHz Cyrix M-II) gives over 300 kkeys/sec. > > Continuing this apples and oranges comparison, the SPARC Ultra 1 that I am > > typing on now gives 321 kkeys/sec and it gives pretty good response > > as a user. > > I just went to the site (www.memoryx.com thanx for the link BTW) to check up > on prices since the boards on Cycle-something -- another SS2 upgrade > manufacturer, were priced at around $1000 plus new. I figured that a > workstation from the early '90s ought to be around $150 or even less -- the > price was ~$400! Given what you've said about being 'slow as honey' and > kkeys/sec (whatever that is) especially in comparison to a 486/100 (!!), why > would anyone spend such an enormous amount of cash? I just picked up a > Pentium socket 4 mb less than 2 months ago for $10!! And if that wasn't > enough, I fried the BIOS and went to pick up another one and the vendor gave > me a used replacement for FREE! There are several factors at work here, and the final result isn't quite so black&white. ;-) First off, you must remember that Sun(tm) equipment has that "Made in America" pricing, while almost all clone PCs are made in Asia in volumes 10 times as high, and of course with much cheaper labour, than is usually possible in North America. Often the inflated price of North American equipment carries on through the used equipment market too. Secondly you do get better quality for your money with Sun equipment. Not only are the parts and manufacturing often of visibly higher quality and certainly of proven higher reliability, but the design of Sun equipment is vastly superior to PC equipment and you do end up paying for that too. My little ss-1's and ss-2 have been running almost steadily since they were manufactured, and other than blowing dust out of them occasionally (and adding more memory!) they continue to run flawlessly, including the power supply fans. My sun-3/60 ran steady for over 10 years and probably only failed because I didn't clean the dirt out of the fan often enough and as a result it got nearly smoking hot inside. This quality and reliability and good design is why even used equipment buyers are willing to spend a bit more on old Suns than anyone in their right mind would spend on old PCs. Finally there's the difference between how a good server-class Sun machine responds to a heavier multi-tasking load than an average PC will. Until the Pentium, and especially the Pentium-Pro (and now its incarnation as the P-II) came along, even a sparcstation-2 can often blow the doors off a 386 or 486 in terms of total aggregate system throughput under heavy load. As has been shown the 486 can do math faster in a single-tasking system than even a sparcstation-5 can (and maybe even faster than a slower ss-20), but that's not the best measure of overall system performance. I am really happy with my new surplus-priced IBM Pentium-II 300MHz "Made in the U.K." system though -- it's well designed and well made with high-quality parts in it, and it runs like a bat-out-of-hell (it's at least three generations newer than my ss2 though). I wouldn't ever use it as a workstation though. We'll see if it lasts 10 years too. Meanwhile I'll probably buy a used ss5 soon, now that the prices have begun to fall to the point where I can't afford not to buy one! ;-) -- Greg A. Woods +1 416 218-0098 VE3TCP Planix, Inc. ; Secrets of the Weird - ------------------------------ Date: Tue, 16 Mar 1999 18:27:18 -0800 (PST) From: David Ras Subject: Suns-at-Home Digest V12 #8 To: suns-at-home@tigger.net-kitchen.com here's the official word from the field eng's handbook: Determining the SCSI Bus Length Your desktop system performance is reliable with a maximum SCSI bus length of 20 feet (6 meters), as illustrated below. The internal signal path of your system unit and the external SCSI cables must not exceed this maximum length. If this length is exceeded, the system will not run in a reliable manner. Figure 2-6 Example of a Daisy Chain Configuration To determine if the total SCSI bus length of the configuration you want to use with your desktop system is 6 meters or less, refer to Appendix B for the internal SCSI bus lengths of specific SPARCŪ systems, devices, and cables, and examples of how to figure the bus lengths for various configurations. For example, suppose you wish to connect a DDP and a Desktop Storage Module to the SPARCstation(TM) 2 system. This configuration would have a total internal SCSI bus length of 3 meters (as indicated below). SPARCstation 2 System (internal signal path) 0.50 meters Desktop Storage Pack (internal signal path) 0.30 meters Desktop Disk Pack (internal signal path) 0.30 meters Desktop Storage Module (internal signal path) 0.30 meters Micro-miniature SCSI Cable (for DDP/DBP/DSP) 0.80 meters Micro-miniature SCSI Cable (for DSM) 0.80 meters Since this total - added to the length of the external cables - is well below the maximum of 6 meters, this configuration is acceptable. Regarding SCSI Termination Recently released Sun systems and peripherals are now supplied with new regulated (active) SCSI terminators. Previously, Sun supplied standard SCSI terminators. Both types are shown below. Figure 2-7 Regulated SCSC Terminator All SCSI systems must be terminated at the end of the daisy chain that is attached to the SCSI bus. A terminator is built in to all bus SCSI cards and to all host systems. For the last unit attached to the SCSI bus, the regulated terminators provide the improved impedance matching required for Fast SCSI. The regulated terminators must be used for all 50-pin SCSI busses having Fast SCSI drives on a Fast SCSI host. Devices with the 3-row 50-pin D connector or the 50-pin ribbon connector (old-style connectors) should not be used on the same bus with Fast SCSI devices. Note - The mixing of Fast SCSI devices and old-style connectors in the same daisy chain is not recommended since errors may be created and performance may be degraded. If Fast SCSI devices and old-style connectors must be used in the same system, the old-style connectors should be connected to a separate SCSI port that doesn't contain Fast SCSI devices. If you are running SunOS 4.1.2 and earlier, and use a Fast SCSI bus, there shouldn't be any problems with system performance. If you are using SunOS 4.1.3 (Solaris 2.1) and later with a Fast SCSI bus, you must turn off the Fast bus or performance will be degraded. does that help? -dr Later, \|/ o o ? David Ras \_/ Sun Microsystems +--------------------------------------------+ | Device Inside Outside | | ------ ------ ------- | | e-mail davras@yamaha david.ras@sun.com | | phone x36079 (510) 574-6079 | | pager davras@pager (408) 989-7440 | | | | "Java done here!" | +--------------------------------------------+ - ------------------------------ Date: Sun, 21 Mar 1999 12:42:56 -0600 From: "Cruz0025 @ The U" Subject: Suns-at-Home Digest V12 #8 To: Thanks Bob and Greg for your replies. While I certainly did not intend to flame anyone in this group, Dave's comment about benchmark comparisons of 486's and SS2s seemed to warrant a reality check and your replies supplied it; it would seem that I was hoping to put a new engine in my old Mercedes and expecting an old Toyota price. The fact that I consider a 16-inch monitor and a 1 Gb harddrive 'big' and would consider a 486 to a Pentium 60 upgrade should be indicators as to my financial situation. Your advice regarding 'upgrades' is valuable. I probably would've been more receptive to the original OS that came preinstalled had I received the original CDs with it and had not followed a blind impulse to purchase a 'free' copy of Solaris 7.0 personal edition which I regretted soon afterwords. It was suggested by a co-worker that a native OS (e.g. Solaris) would probably fair better on it's home turf. Now your comparisons make sense: I didn't much care for the only 386 SX machine I ever saw running Windows 95! Incidently, your askant remark (if you insist) regarding Windows did not escape me. However, although I share your spirit, I don't have the solid background in Unix to put the muscle behind my words. What I've learned of computers in the last 4 years wasn't learned in places where one routinely finds Sun workstations to get one's hands dirty. But I'm still here! Justin Vanderziel zielot@thebestisp.com - ------------------------------ Date: Sun, 14 Mar 1999 18:14:33 +0200 (EET) From: Prevelakis Vassilis Subject: using a SunSwift adapter on an IPX To: suns-at-home@tigger.net-kitchen.com Hello, I have an IPX and would like to add a SunSwift Sbus card. This card has a 100MHz Ethernet interface (hme) and a Fast-Wide SCSI port (fas). Unfortunately although the card is certainly operational (I have tried it on an Ultra1) the IPX won't boot with the card attached. I suspect that the PROM monitor does not support the card but I hope that some modification can be made to the EEPROM configuration so that the card is recognised. Thanks **vp Vassilis Prevelakis vp@unipi.gr Network Management Center University of Piraeus - ------------------------------ End of Suns-at-Home Digest ******************************