I'm not sure that the Econoram IV was one of the boards that used those 64k-bit DRAMs. Did it? This alpha-particle issue didn't come up with the 4K and 16K dram devices. It only surfaced when the density went up to 64K-bits and the supply went to a single 5-volt Vcc. The first context in which I recall the incidence of packaging-induced alpha-particle-hits causing a problem was with 64K-bit DRAMs. There was a component of this problem that was characteristic of the device family, and was later mitigated by adding an XOR buffer (XOR-ing data with one of the address lines) to the data to equalize otherwise random charge distribution throughout the device. Needless to say, I was never convinced, myself, that all the information about the then-new 5-volt-only DRAMs was on the level, as much of it sounded like folklore. The fact that the industry went to a logic fix suggests it wasn't without merit, however. There were plenty of other reasons why various DRAM circuits didn't work terribly well, most dealing with timing and buffering. Once the S-100 became dominated by Z80's, the use of various features of the Z80, not present in the 8080 round which the bus was designed, began to cause interoperability problems on the S-100, as a consequence of which it became advisable to have DRAM boards that were either made or certified by the CPU maker. Several vendors made DRAM boards that worked and played quite well with others, but Godbout/CompuPro was not one of them. One couldn't even rely on those working with older Godbout products in the system. Dick ----- Original Message ----- From: "Douglas Quebbeman" <dhquebbeman(a)theestopinalgroup.com> To: <classiccmp(a)classiccmp.org> Sent: Monday, November 19, 2001 9:23 AM Subject: RE: Intel C8080A chip brings $565 on EBAY