anderson@vms.macc.wisc.edu (Jess Anderson, MACC) (08/23/89)
(Sorry about the empty post, I can't believe I did that!) If you have a Longshine EEMS memory card or know about such things in general, I'm hoping you can advise me regarding a problem I've been having. The setup is this: AT clone (Everex 1800B) with full 1 MB on the motherboard. This is a 10MHz machine, so it should not need DRAMs faster than 120 ns. But when I bought the Longshine card (model LCS-8661), I didn't know that, so I populated it with 2MB of 100 ns chips (this was well before prices went nuts). The jumpers are set to configure the whole 2MB as expanded (not extended) memory. I don't use the Turbo Spooler supplied by Longshine, but I do use the EMS (3.2) driver, called LEMS.SYS and the ramdisk driver TDISK.SYS (usually with 512K buffer size), leaving 1.5MB available as expanded memory. The documentation is sparse and not entirely intelligible at that. Everything worked great for a long time. But then I updated my MS Word from version 3.1 to 4.0 and later to 5.0. Under both 4.0 and 5.0, *if* I configure the Longshine card (device=c:\dev\lems.sys /n in config.sys), then load Word, it loads, but as soon as loading finishes, the machine hangs with one beep. Under 4.0, the beep is accompanied by an error message (off board parity error) *in 40-column mode* (odd?). Under 5.0, no message, just the beep. Well, it's not much fun having 2MB you can't use. Other software that accesses the board (e.g., Ventura Publisher Prof. Ext.) works, apparently. If I run memory diagnostics I find no bad chips. Recently, I installed nine new chips in the first bank, and tried again, repeating the test by moving chip sets up. Unless there are *two* bad chips, that should have found it. It didn't. There is a jumper that turns parity checking off, and if I do that, then (no surprise) everything works, but naturally that may not be such a good way to work, since it would seem to invite undetected errors. It could be a couple other things than the DRAMs, I guess, maybe the board developed an open trace (impossible to find, I suppose), or some other chip on it is busted. I guess it's also possible that it's the software itself that leads to the conflict. I have no problem with Word 5.0 or 4.0 on my office machine, which is also an 1800B, but with an Intel AboveBoard. Could be it's a problem of getting what you pay for. But before I shell out for a new board, maybe somebody can yet help me find a way to find the problem or offer other advice. That would sure be nice. ==Jess Anderson===Academic Computing Center=====Univ. Wisconsin-Madison===== | Work: Rm. 2160, 1210 West Dayton St., Madison WI 53706, Ph. 608/263-6988 | | Home: 2838 Stevens St., 53705, 608/238-4833 BITNET: anderson@wiscmacc | ==ARPA: anderson@macc.wisc.edu========UUCP:{}!uwvax!macc.wisc.edu!anderson==