--- Roger Ivie <IVIE(a)cc.usu.edu> wrote: And I've got the mounds of leftover crap to prove it! What an odd thing to make. That's a *lot* of instruments in a small amount of space (owing to cable-length limitations). Our drivers were a little different - first, all of our products were 68000- based. Second, we used 1/4th of the address space as shared memory to the VAX as a DMA cycle. Imagine this if you will... accessing 18 or 22 bits of 68K address space would return (or modify) the contents of a real physical memory location in the VAX. I have used one of our Qbus boards and a Fluke ICE tester to check LSI-11 DRAM chips... It works something like this... o Plug into a BA11-N an 11/03 CPU with a row of 4096 DRAMs socketed, a DL11-E, a COMBOARD-Q with the CPU removed and the Fluke plugged into the CPU socket. o Power on the BA11-N, then use ODT to flip a bit in the COMBOARD-Q CSR to enable DMA cycles (safety measure on our part) o Set up the Fluke to test 4K of memory in the "shared memory" quadrant of the 68K address space. o Check the results. I used this arrangement to test 4Kx1 DRAMs for my Z-80-based GORF machine. I have a DRAM tester for 64K and 256K chips, but it only tests single-supply parts. As far as I'm concerned, 100 is a handful. My old company made more than ten times as many Unibus cards as that. I think I get one inquiry per year about our old sync products; nothing serious. Who needs to move sync data over 3780, HASP or SNA to or from a VAX these days that doesn't already have a way to do it? -ethan ===== Infinet has been sold. The domain is going away in February. Please send all replies to erd(a)iname.com __________________________________________________ Do You Yahoo!? Thousands of Stores. Millions of Products. All in one place. Yahoo! Shopping: http://shopping.yahoo.com