Image Map Image Map
Page 5 of 8 FirstFirst 12345678 LastLast
Results 41 to 50 of 73

Thread: D.R.E.A.M. 6800 Build

  1. #41
    Join Date
    Jan 2011
    Location
    Vancouver, BC
    Posts
    4,374
    Blog Entries
    2

    Default

    Wow.. that's amazing. It was a 365 after all. I just didn't realize initially there were two of them. I pulled the second and put it in the Hazeltine and the Hazeltine went ape. Swapped in the known good from the Hazeltine into the DREAM... boom!

    Also have had some interesting emails with Michael Bauer. Apparently JR Systems was a company that the magazine contracted to make kit PCBs of the DREAM, and they also made some expansion boards like this 3 PROM expansion board I got. On their own accord (and with Michael's permission but not blessing), they created a 6802 variant. Michael says the 6802 wasn't such a great choice because it lacked the tri-state pins of the 6800, and this set up a fight between the MPU and video controller for the bus. Despite this, many 6802 DREAMs worked anyway - Michael figures the video controller somehow wins the battle.

    Anyway, now I've gotta put it all back together and repair some pins I broke on the connector.. ugh. I should've stuck with my gut, but I thought there was only one 365. It's interesting how similar the behavior was to the formerly broken Hazeltine.

  2. #42
    Join Date
    Jan 2011
    Location
    Vancouver, BC
    Posts
    4,374
    Blog Entries
    2

    Default

    Okay so with the DREAM working I've attempted to mess around with the functions and so forth. It doesn't quite work the way the manual says.

    For example, the manual says to go and edit memory contents, you hit FN and then 0, and then enter the four digit address you want to start from, and then the bytes for each address. On mine, when you hit FN, it immediately brings up byte entry at whatever address you're sitting at.

    The keypad is a little iffy to say the least, but I discovered by accident that if I reset, and then hit four digits (when it recognizes them being pressed), it'll take me to whatever address those four digits represent. If I then hit FN it'll start letting me enter digits. If I hit FN after that, it just increments by one memory address.

    This is all fine, if a little clunky. The problem is I don't know how you then Load or Save anything - since the second you hit FN it increments memory addresses.. it gives you no time whatsoever to enter tape mode.

    Not sure if this is a bug or feature -- ie. if someone modified the CHIPOS. I'll have to dump the EPROM tomorrow and compare.

  3. #43
    Join Date
    Jun 2012
    Location
    UK - Worcester
    Posts
    2,686

    Default

    I don't know what manual you are reading - but that is not the same as the manual I am reading.

    To edit the display buffer you enter the following sequence:

    [RST] 0 1 0 0 [FN] 0

    [RST] is the RESET key. 0100 is the hex address of where you want to look/edit in memory (i.e. the address is entered FIRST before the function key) and then hit the function key followed by the numeral 0. [FN] 0 means "enter memory display/edit mode".

    You can now either enter a 2 digit hex value (in which case the value entered will be stored and the memory address automatically incremented to the next address) or hit the [FN] key again to move on to the next sequential memory location.

    Incidentally, as you start to modify memory at 0100, you should see the bits in the upper-left of the video screen image change - as this is the memory you are playing with...

    You need to press [RST] to terminate the memory modify command.

    To save a block of memory to the tape, you first need to modify memory locations 2, 3, 4 and 5 with the starting address and (end address + 1) of the block of memory to save.

    Address 2 contains the most significant byte of the start address of the block of memory to save.
    Address 3 contains the least significant byte of the start address of the block of memory to save.
    Address 4 contains the most significant byte of the end address (+1) of the block of memory to save.
    Address 5 contains the least significant byte of the end address (+1) of the block of memory to save.

    To save/load the display buffer enter:

    Address 2 = 01.
    Address 3 = 00.
    Address 4 = 02.
    Address 5 = 00.

    The display buffer extends from 0100 to 01FF (0200 = 01FF+1).

    Enter the keystrokes:

    [RST] 0 0 0 2 [FN] 0.

    Enter the four bytes of the start and (end address + 1) followed by the [RST] key to return to 'command' mode.

    Enter [FN] 2 to save the block of memory specified to tape. Or (having entered the parameters in memory locations 2 through 5) hit [FN] 1 to load data from tape into the specified memory.

    To execute a program enter:

    [RST] C 0 0 0 [FN] 3

    I am still trying to work out why it is C000 - but I think I know... EDIT: C000 is the start of the CHIP8 interpreter to 'execute' the user program.

    Dave
    Last edited by daver2; January 28th, 2019 at 10:58 AM.

  4. #44
    Join Date
    Jan 2011
    Location
    Vancouver, BC
    Posts
    4,374
    Blog Entries
    2

    Default

    Thanks muchly for that Dave. I was reading the brief given in the original article. I probably just read it wrong but it seemed like they were saying hit FN and then one of 0, 1, 2, 3 or 4 to either memmod, dump, load or save data.

    I think complicating matters is my keypad has issues. I sometimes have to press a key several times before I get a beep acknowledging it. I'm not sure if its the keypad itself or something going on with the 6821 etc. It seems to get the characters right if I press them enough times.

  5. #45
    Join Date
    Jan 2011
    Location
    Vancouver, BC
    Posts
    4,374
    Blog Entries
    2

    Default

    Anyone have any thoughts on how you might service a Digitran keypad?

    Looking at the back.. it doesn't look like it was intended to be serviced. The plastic bits that pass through the PCB are sort of bent down to prevent it. digitran-keypad-back.jpg

    I've been pressing all the keys on the one my DREAM has and have gotten some to become responsive again... but a couple, like 5 and 9.. just won't budge. Actually I got 5 to work briefly, but then it stopped again.

    I'm wondering if there are any tricks, short of trying to disassemble it, that might get the contacts going again.

    I'm assuming it's the contacts anyway. I notice other keys, like C and D as examples, work just fine every time.

  6. #46
    Join Date
    Jun 2012
    Location
    UK - Worcester
    Posts
    2,686

    Default

    It took me a few times to understand the article as well...

    I would checkout the keypad with a multimeter set to continuity. Find the row and column for each key. Put your multimeter on the row and column for the key you wish to test, and poke the key to test it out.

    I think if the keypad is dead (or some of the keys) then I think it is probably toast and you will need to replace it.

    Dave

  7. #47
    Join Date
    Jan 2011
    Location
    Vancouver, BC
    Posts
    4,374
    Blog Entries
    2

    Default

    Yeah I think this keypad is right buggered. I can get every key except 5 and 9.. and 1 has also become flaky.

    I don't *think* this would be anything to do with the computer.. I'd assume if it was keys not being translated properly you'd lose everything in a row or column? For example, 1, 5, 9 and D all share a column, but D works and 1 sort of does. And I find some keys that weren't working at all before worked after repeated presses. I'm suspecting we've got dirty contacts in there -- this thing had a thin but perceptible layer of Aussie dust and a tiny bit of corrosion. I'm thinking some moisture might've rotted the keypad.

  8. #48
    Join Date
    Jan 2011
    Location
    Vancouver, BC
    Posts
    4,374
    Blog Entries
    2

    Default

    Also.. I don't think this keypad will be easy to replace. I have three other keypads here...but they have more pins... looks like they have one pin to each key. The one I have has 7 or 8 pins... I think it has some kind of matrix system?

  9. #49
    Join Date
    Jan 2011
    Location
    Vancouver, BC
    Posts
    4,374
    Blog Entries
    2

    Default

    Okay so I now know the answer to can you service a digitran xl keypad: yeah kinda.

    They definitely aren't meant to be repaired, but I did find by straightening or removing the little black plastic nubs that hold the pcb tight against the keys I could pry the pcb off, with sickening plastic snapping noises all the way for nubs that wouldn't let go.

    I was presented with this:

    20190129_215659.jpg

    Where that white stuff is is where the 5 and 9 keys are. Something was disintegrating in there.. a white plasticky kind of stuff.

    The keypad itself is dead simple: foil tabs make contact between the key pad and a series of pads on the left side.
    They even label on the back of the pcb which two of the 8 pins have to connect to get each key.

    Anyway... I did a light sanding on the contacts. Amazingly it snapped back together.. I guess there was just enough nub there to work... and voila.. it works. Too late tonight to actually try programming.. but I will. It's a good thing this worked out... none of my other keypads would have worked since the DREAM design assumes a matrix.

  10. #50
    Join Date
    Jan 2011
    Location
    Vancouver, BC
    Posts
    4,374
    Blog Entries
    2

    Default

    I successfully entered a program! Unfortunately the number 2 key gave out and started producing 1s for some reason. Have to take the keypad apart again. Program worked great... was a little space shooter. Kinda minimal but I get how people would have found it thrilling back in 1978.

    I then shut off while I went about other things. I thought before I fix the keypad again and try to enter another program, I'd try to save a short bit of data. I entered values from 0200 to 0210.. set the pointers for it and then made it do a save.

    What seems to be happening is the DREAM is continuously putting out that solid tone you normally hear as the intro a computer looks for before the data stream happens. The first time I told it to save, i hit record, the screen blanked for 1 second, and that was it... back to the normal data entry. I checked the tape... there was that continuous tone, and then a brief bit of the sound you usually hear when recording. Initially I was puzzled about the tone and its length. I thought maybe it hadn't recorded anything at all and that I had some other machine recording there. So I took care to blank the tape with the DREAM disconnected from the recorder.. then tried again. But yeah... it is basically generating that tone all the time. Anyway on my second attempt to record it only gave me tone... no indication of any data. I could be doing things wrong or not being patient enough. Not sure.

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
  •