[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

strange problem with dual IDE card

I'm using a CompuPro box (20+ slots) with a CompuPro CPUZ and RAM 17 (64k static card).  The console is via a N8VEM serial card @ 38.4k baud.  Very nice set-up.  I'd like to get a compact flash based system going, but the N8VEM IDE card is causing all sorts of problems.  There is no problem reading and writing to both compact flash modules on the card (i.e. no errors).  And I have used several compact flash modules, including the 4 GB Kingston (the one with the flower on the front that seems to be used a lot in John's build photos).  Same thing happens with all of them.

Here's the strangeness.  If I write all ASCII numbers in each of the, say, first ten sectors (so, the first sector will have 512 31h's in it, and the second 512 32h's in it, and so on), everything seems to work OK.  If I power off the system, and start it up again, when I look at the content of the first sector it is 512 32h's, *NOT* what I wrote.  It's like I'm looking at the content of the second sector, BUT the hex display on the card says that I'm looking at the first sector.  Each sector of the first 10 that I've written appears to be "shifted" up by one.  BUT, if I write to any other sector (just one write), the content of the first 10 sectors "shifts" back to what it was supposed to be:  31h in sector 1, 32h in sector 2, etc. etc.  I'm using the MYIDE utility from John's build pages (re-jiggered to match my port addressing (50h, I think)), as well as a home grown utility.  Same thing happens with each.  Not only that, but if I move the IDE card to a Cromemco box (ZPU, 256k DRAM card, 16FDC, and of course the IDE card) the same thing happens.  But I only have the re-jiggered version of MYIDE on this system, not my home grown utility.

I'm pulling my hair out with this.  It seems very, very consistent and reproducible.  Anybody seen anything like this?  Ever?  Please help before I go bald!!!!