Image Map Image Map
Page 1 of 4 1234 LastLast
Results 1 to 10 of 35

Thread: PDP 11/45, Part 6

  1. #1
    Join Date
    Jul 2009
    Location
    Boulder , Colorado USA
    Posts
    2,900

    Default PDP 11/45, Part 6

    Hi All;

    It looks like it's that time of Year Again..

    I am going to see what Instructions work and what Instructions do not work..

    Also, Does anyone know of any Document that had what mirco-instructions, go with what Instructions ??

    Here is the link for part 5 --

    http://www.vcfed.org/forum/showthrea...P-11-45-Part-5

    http://www.vcfed.org/forum/showthrea...-Part-5/page30

    Dave and FritzM, I hope to hear from both of You..

    THANK YOU Marty

  2. #2
    Join Date
    Dec 2011
    Location
    Oakland, CA
    Posts
    116

    Default

    Nice to see you back to the thread again, Marty!

    Standing by help out if I can. My 11/45 is working pretty well now, so I can run experiments if you need me to.

    Regarding microcode: there are a couple of tables and some narrative in the KB11-A maintenance manual to help navigate the microcode flow forks for each type of instruction. And then of course the microcode flows themselves in the 11/45 engineering drawings.

  3. #3
    Join Date
    Jul 2009
    Location
    Boulder , Colorado USA
    Posts
    2,900

    Default

    Hi All;

    FritzM, Thank You for the Encouraging words..

    I have the Information that You Reference, I just need to figure it all out..

    I can try an Instruction and see if I can figure out the complete flow for it to work..

    I think the Microcode board is still broke, I will start from there..

    THANK YOU Marty

  4. #4
    Join Date
    Dec 2011
    Location
    Oakland, CA
    Posts
    116

    Default

    So, much of the front panel behavior on the 11/45 (examine, deposit, etc.) is implemented in microcode; when the machine is “halted" it is actually looping in microcode state CON.00 (sheet 14 of the 11/45 microcode flows). That state loads the settings of the front-panel switches into BRA in preparation for potential use in a subsequent load-address or deposit. As a consequence, when the halt switch is down and the data display is in “Bus Register” mode, the data lights should follow along live with the front panel switches. This is the first and simplest test on an 11/45 to verify that the CPU is actually sequencing microcode. Also in this state, if the data switch is set to "u Addr" the lower eight bits of the data display should display constant octal 170, the address of the looping microcode state.

  5. #5
    Join Date
    Jul 2009
    Location
    Boulder , Colorado USA
    Posts
    2,900

    Default

    Hi All;

    FritzM, Thank You for the Reminder..

    If I remember correctly, the last time I used it, I had Deposit and Examine working, but it either wouldn't execute code from Memory or it wouldn't execute some instructions from Memory..

    First I need to check the Power Supply and its wiring, before I do any powering up of the Machine..

    THANK YOU Marty

  6. #6
    Join Date
    Jun 2012
    Location
    UK - Worcester
    Posts
    1,211

    Default

    Hi Marty/Fritz,

    Are we on Part 6 already?!

    I have been following your progress Fritz. Looking a nice system! I am cleaning my voltage regulators up at the moment and then will move on to reforming the electrolytics. There only seems to be one voltage regulator that I need to replace the capacitors on (they are definitely bulging/leaking). I will see how many of the original electrolytics I can save by reforming them.

    The 11/45 microcode is described in the KB11-A Maintenance Manual (as Fritz has already identified). They are also in the 1145 System Engineering Drawings (in flowchart form).

    I did find some slight errors/discrepancies within these two documents.

    If you remember, I produced an EXCEL spreadsheet of the microcode for use with my 11/45 LOGISIM simulator (which I have still not completed yet). I tried to take the discrepancies out (which I think I achieved).

    If you need some help to 'kick start' your memory - ask away!

    As I remember it - didn't you have a contact problem with either the EXAMINE or DEPOSIT key? If so, it would be sensible to repair/replace the switch - or 'tack solder' another switch in parallel with it and use that. You need to get EXAMINE and DEPOSIT working correctly first.

    Don't forget, you can use a KM11 to 'single clock step' the microcode. Didn't you make your own KM11?

    Dave

  7. #7
    Join Date
    Jul 2009
    Location
    Boulder , Colorado USA
    Posts
    2,900

    Default

    Hi All;

    Welcome back, Dave..

    "" As I remember it - didn't you have a contact problem with either the EXAMINE or DEPOSIT key? If so, it would be sensible to repair/replace the switch - or 'tack solder' another switch in parallel with it and use that. You need to get EXAMINE and DEPOSIT working correctly first. ""

    Yes, I had a switch Problem, I don't remember which one at Present, and I took out the Register Deposit switch to use as a Replacement, and Later I found a Replacement for the Register Deposit Switch, So, I think I am good on the Switch Department..

    And If I remember correctly I got Examine and Deposit to work correctly with Memory, it's making the op-codes work from Memory..

    "" Don't forget, you can use a KM11 to 'single clock step' the microcode. Didn't you make your own KM11? ""

    Yes, I had made my own KM11, and Yes, I got a replacement KM11.. Currently, I am using the Replacement look alike KM11.. My Homemade one I took apart..
    I also have a Timing Board that I can put my 'crystal divide by jumper' board and if I need it it can really slow things down, and that is besides the on board R-C time constant type of Frequency setting..

    THANK YOU Marty

  8. #8
    Join Date
    Jun 2012
    Location
    UK - Worcester
    Posts
    1,211

    Default

    I never went away Marty... Just over on other threads...

    In that case, try depositing a sequence of 000240 instructions (NOP) followed by a 000000 (HALT) and executing the first NOP instruction. See if you HALT where you expect to - or does it go off into infinity and beyond?

    If it doesn't do what you expect, re-deposit the program and single step through the NOPs. Do they do what you would expect (i.e. nothing) but just incrementing the PC (R7)?

    If that doesn't work, re-deposit the program and go into 'microcycling' and see what happens. You should (hopefully) get a clue form the microcode address it branches to.

    If this is all 'gibberish' I will try and expand a bit more tomorrow.

    Regards,

    Dave

  9. #9
    Join Date
    Jul 2009
    Location
    Boulder , Colorado USA
    Posts
    2,900

    Default

    Hi All;

    I have tried it once, and it isn't correct..

    With "Halt sw down, I Load Addr at '0000..

    I Deposit '000240 at '0000 thru '20..

    I deposit '00 at address '22 thru '40..

    Reload address '00, do an examine it shows '000240..

    If I do a "start" it shows Addr at '0000 and data at '177777..

    If I do a "Cont" it shows Addr at '00002 and Data at '177777..

    If I do a 'UAddr/cpu, Cont four time it goes from '170 to 355..

    And it is gone to the weeds..

    THANK YOU Marty

  10. #10
    Join Date
    Jul 2009
    Location
    Boulder , Colorado USA
    Posts
    2,900

    Default

    Hi All;

    I got out my Extender Card and put my KM11 on it, and I got it to Single Step and have the Display on UAddr/CPU, and so I can see it go thru the various steps..
    I have confirmed Examine, and I need to do Deposit next and then Load Address, before going to the unknown switches, like 'Cont' and 'Start' etc..
    And I can see if they follow the correct Micro-op code..

    THANK YOU Marty

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •