My old machine is slowly becoming sensible. Turns out I had three bad RAM chips in my Seattle 16K boards. They're TI 9044's, I assume are 2147 type. Jameco sells them for $6 each; is there a better source? (They're also 55nS claimed, about 3X faster than what's in there now!) I juggled chips such that the duds are at 9000 - 9fff, so I can boot CP/M (62K system) and run some programs (memtest, sid...). I've got to figure out why the floppy controller won't talk to side 2 (sectors 27 - 52) and I can do that in 16K easily. My B: drive is loud OH MAN IS IT SO LOUD I have to power it off after a program load! Need to find another double sided drive. Out of sheer laziness, to avoid even the faint effort it took to find the bad chips, I yanked all my nice static RAM and swapped in my other RAM cards, one by one: * Central Data 64K * MSC 64K * mumble brand, some acronym, no manual (I'm not at home right now) My EPROM monitor resides at 7800 - 7fff, uses a dozen bytes RAM below 7800. It self-copies into RAM and turns itself off (for a clean address space after boot). Central Data: set the jumpers right, plugged it in, Nothing Happens. (Boring story & expletives deleted) it shorted +16V, popped the fuse. Another one of those !@#$%W^% tantalums; 47uF/35V. Replaced with an electrolytic (it was just down near the card edge). Card works now, but flake-ey results, failed address-contention tests, couldn't boot CP/M. It's all 4116's, lots of jumpers for buss timing, ugh. Needs Work. Reference separate rant on capacitors. MSC: Dislike this thing's RC filters on the data/address lines! Crock-detect! I forget the problem, but had problems with one bank (32K). Mumble: Inert. All the static cards back in, debugged the problem. For the S-100 buss, I'll stick with static memory, thank you!