banshee@ucscb.UCSC.EDU (Wailin Through The Nets) (02/06/90)
I have a three year old Z-158 which has intermittent occurances of memory parity failure. This crashes the machine. I've tried a few ram testing software programs, but they either report nothing as being wrong, or (nmi) scream about the parity chip being bad on the first bank (of three). I only have memory failures once a week or so and usually when I am doing something that is using almost all available memory on the machine. Along similar lines, having 3 banks of 256K ram should give me a total of 768Kb of memory -- yet 128 of that is unavailable as DOS can only deal with 640Kb. Is there any method to utilize that RAM? Or is the XT architeture to blame? -- |========================================================================| | banshee@ucscb.UCSC.EDU | YOU CAN BE ONE WITH SATAN, JUST CALL | | ucbvax!ucscc!ucscb!banshee | HIS NEW NUMBER (976) MR SATAN CALL NOW | |-------------------------------| OPERATORS ARE STANDING BY TO TAKE YOUR | | Smail to: Banshee | SOUL TO THE DEPTHS OF HELL WHERE YOU | | UCSC, Cowell College, 705 | WILL BURN AND WRITHE FOREVER! HAHAHAHA | | Santa Cruz, CA, 95064 | --No Subliminal Messages Here | |========================================================================|
kent@sun.ufnet.ufl.edu (Kent Phelps) (02/06/90)
The early Z-158 CPU boards have been infamous in developing intermittant
parity problems. All chips on these boards were socketed and there were so
many mods that there is a rats nest of wires on the back of them. Most of these
parity failures defy locating because they are so erratic. Sometimes reseating
the chips helps. Best solution is to exchange the board with ZDS for the new
version. The new version card is P/N 181-6792. One old version is 181-5857
and I believe there is one other part number that is even older. You can
exchange the card through a service center or through Heath parts for about
$200.00.
Hope this helps.
--
Kent Phelps | Internet: kent@sun.ufnet.ufl.edu
University of Florida | UUCP: ... ????
Digital Design Facility| Phone: 904-335-8322bobmon@iuvax.cs.indiana.edu (RAMontante) (02/07/90)
About getting "all" your 768K in a z-158 (and a possible memory board solution): I added an expanded memory board to my z-158. (As is often the case) the board is capable of backfilling the main-memory space before starting the EMS. So I took the third bank of 256K chips off my memory board and added them to the EMS board, then let it backfill from 512K up to 640K. It added the "unused" 128K to its total EMS. So in effect the EMS board gave me an eighth-Meg more memory than I bought for it... As for the "rats' nest of wires" memory board problem, I discovered in the course of playing around that I could take _all_ the chips off my memory board and put them on the EMS board --- it successfully backfilled the entire 640K. I don't do this because it seems kind of suspect to me, and my memory board is okay (actually, I seem to have a chip failure in one of my EMS chips, but I can't diagnose that! argh). But it might get you out of a problem without having to spring for a replacement mem. board. And a final neat feature: Chris Dunford/Cove Software distributes a free program called EEMRAM that can map Enhanced expanded memory into the address space between 640K and a CGA adaptor. No hardware mods needed, and my machine reports 736K of main memory. I love it! My board: RAMPARTNER 2-Meg EEMS/EMS board PC Technologies, Inc. Ann Arbor, MI 48108 313/996-9690 What I got is LIM 3.2/EEMS compatible, they may be compatible with LIM 4.0 by now. Or they might be out of business; I haven't been following along to know. bob,mon.
jackson@acf4.NYU.EDU (Robert Max Jackson) (02/11/90)
Kent Phelps [Internet: kent@sun.ufnet.ufl.edu] writes:
The early Z-158 CPU boards have been infamous in developing intermittant
parity problems.
............................
Damn. 3 of 12 Z-158s I have (about 2 1/2 yrs. old) have
started to show intermittant parity errors in the past several months.
I thought we must be doing something wierd. This makes it sound as if
we should just consider it a lost cause and junk the memory boards
when the problem gets annoying. Is that true or is there another
reasonable course?limonce@pilot.njin.net (Tom Limoncelli) (02/12/90)
When I got that error, I found that shutting off the computer for a while (5 minutes... really drain those caps) would solve the problem. In fact, I thought the problem was that I had been power cycling too quickly. Our repair department says that a ROM replacement fixes the problem on the Z157. Hope some of that info was helpful, Tom -- Tom Limoncelli The computer industry should spend more time in front of tlimonce@drew.uucp their computers. Remember when "Look & Feel" tlimonce@drew.Bitnet was what you tried to do on a date? limonce@pilot.njin.net