It is unlikely that a 6 bits in a byte will fail, caused by a RAM, since each RAM chip is only one bit, not a byte. It is more likely that one of the ROM chips is being doubly decoded. Since these are decoded as 1 K blocks, it is likely a problem with the address block decoder, the 74LS145. Just my guess. Dwight
________________________________ From: Cameron Kaiser via cctalk <cctalk@classiccmp.org> Sent: Friday, May 12, 2023 8:51 AM To: Jon Elson via cctalk <cctalk@classiccmp.org> Cc: Cameron Kaiser <spec...@floodgap.com> Subject: [cctalk] Re: KIM-1 stuck bits from $280 to $29f > If the failure affects the EPROM monitor,then any results you get from the > monitor are suspect. No, what I mean is, the appearance of the upper six bits being dead was because of how the monitor shifts in data from the keypad. Since bit 2 was always zero, it would look like everything above it was zero too because the bit shifts carried the error forward. A direct brute-force step through showed the actual issue and I should have just done that in the first place. The monitor works properly everywhere else outside of those locations, including from the TTY. -- ------------------------------------ personal: http://www.cameronkaiser.com/ -- Cameron Kaiser * Floodgap Systems * www.floodgap.com<http://www.floodgap.com> * ckai...@floodgap.com -- Ninety-nine percent of lawyers give the rest a bad name. -------------------
