I feel such an idiot! I had minicom set for hardware flow control on the PC that was connected as the console :-)
I'm sure I remember that as a symptom of no bank 0 memory! Years ago it would have been the first thing I checked, well maybe not, I would have had a real VT100 to connect to it. I am getting old! Now I start adding modules, thanks for all who made suggestions! cheers, Nigel On 2021-10-23 4:47 p.m., Noel Chiappa wrote: > > From: Nigel Johnson Ham > > > an 11/23 will not work without bank zero memory > > It depends on what you mean by 'work'. If you mean 'ODT does not operate > correctly', or 'the CPU won't run', I can assure you that neither of those > is correct. > > Here is a recorded log from a session earlier today on a system here next to > my desktop: it contains a KDF11-A (in slot 1Left of a Q-Q backplane) > and a console DLV11-J (in slot 1Right), and no other cards: > > ^@^@ > 000000 > @777560/000000 > @777564/000000 > 777566/000000 > 777570/? > @777640/177777 777 > @/000777 > @777640G > Cmd: > > 177640 > @^@^@ > > As you can see, ODT is working,and the console registers are all responding > OK. > > The thing at the bottom is that hack of putting a small program in the PARs > (UIPAR0, in this case). The "Cmd:" prompt is where I told my console program > to send a 'break' down the serial line to the -11, stopping the CPU. (The > 'run' light had been on, after the "777640G".) > > (Note: The PAR hack doesn't work in J-11 CPUs; they won't do instruction > fetches from PARs.) > > To confirm that I'm not BSing you, can someone else on the list > please do this on their KDF11-A, and confirm that it produces the results > I show about? Thanks. > > > > All I get is output of 173000 and the no ability to input. > > If that's with just a non-LSI11 CPU card and the console serial line, the > machine has a problem. As you can see above (the "777560/", etc were my > type-in), you should be able to talk to ODT on a KDF11 (and I tried a > KDJ11 yesterday, it worked too). > > (I don't recall if the LSI-11 types _anything_ when started with no working > memory at 0; if anyone wants to know, I can fairly easily try it, I have a > bunch of working LDI-11's, both quad and dual cards). > > Noel
