I encountered the same exact situation a few months back and sent a few emails (maybe to John as he had some suggestions) at the time. I also built a NOP fixture for the Centipede baord and was using a 5004a signature analyzer. I could probably dig up the recorded results and they will probably agree 100% with your results. I dug around for a while looking for reasons but ended up generating my own list of signatures based on a known good board. One thing I did find in some documentation ( http://www.spies.com/~arcade/TE/SigAnalNotes.pdf - on page 14 of the pdf document) an interesting table that lists a table with different processors ans some signatures for each address line. As for the 6502, it lists to separate values, just the 6502 and another 6502-A footnoted with a phrase of 'Atari games will use a KK 6502-A NOP. The signatures will be the same as for an 8080.' If you compare the two bits of information for each type to values you can record, it is simple to see what you have built (as I did) as opposed to what you need to match the other values throughout the analysis guide furnished by Atari. If I remember correctly, my results matched the 6502 list, but it has been a while. 6502 A0 - UUUU A1 - FFFF A2 - 8484 A3 - P763 A4 - 1U5P A5 - 0356 A6 - U759 A7 - 6F9A A8 - 7791 .... 6502-A A0 - UUUU A1 - 5555 A2 - CCCC A3 - 7F7F A4 - 5H21 A5 - 0AFA A6 - UPFH A7 - 52F8 A8 - HC89 ... I eventually just built my own signature list for a known working board with the fixture I had built as the Signature analysis numbers are not really a true measurement other than a certain number of clock cycles that pass before an signal is encountered - correct me if I am wrong on that anyway. There are some interesting articles on Al's sight regarding the use of signature analysis if anyone is bored. So I guess to help the question along, I think what we need to know is what do you do differently to build a NOP ficture to produce results for a 6502 (same as the 6800 signatures) as opposed to the 6502-A (same as the Z80 and the 8080 signatures)? I sure am glad someone duplicated my results. I look forward to any answers that can set this straight. On a separate note - is anyone interested in a HP5004a signature analyser? I haven't used it since. Greg SteveT wrote:
I just got a hp5004a , trying to use it to help repair a sick
centipede board. I built a NOP adapter (processor and socket)
It works perfectly. Signatures match the docs for the NOP adapter but they do NOT match what is published on the centipede schematics starting right at the micro. What Am I missing here? Is there something about the cat box I'm not taking into account here. Anyone out there using the HP5004a with centipede?
Steve