(Topic ID: 122521)

system9 and memory check strangeness

By Spudgunman

9 years ago


Topic Heartbeat

Topic Stats

  • 12 posts
  • 5 Pinsiders participating
  • Latest reply 9 years ago by thedefog
  • Topic is favorited by 1 Pinsider

You

Linked Games

No games have been linked to this topic.

    #1 9 years ago

    I got a leon in a system 9 flashing 0/7 if I run the memory test on 7 it will complete, if I run the memory test on 0 it wont. threw a another ram chip in same results.. thoughts?

    #2 9 years ago

    12h bump no one has seen this?

    #3 9 years ago

    Look at the items related to Error 7
    From PinWiki:

    System 9 Games
    The following table lists the error codes displayed on the System 9 MPU 7-segment display.
    Code Description
    0
    Normal game boot with no problems detected. The game should be in attract mode.
    1
    CPU board locked up. Possible cause is memory protect circuit and U18 CMOS RAM "stuck bits".
    That is, an actual failure of the CMOS RAM chip or the 6808 processor's ability to communicate error free with the RAM.
    2
    U20 Game ROM 1 faulty
    3
    U20 Game ROM 1 faulty
    4
    U19 Game ROM 2 faulty
    5
    Blanking signal stuck, coin door closed, memory protect circuit faulty, or the U18 CMOS RAM faulty
    7
    System failure. Occasionally, the following components can contribute to this problem: U21 (4MhZ crystal); components in the IRQ circuit; broken leads on C9 (22uF) in the reset section; loosely seated ICs on the CPU board
    Other or no indication
    U20 Game ROM 1 faulty

    #4 9 years ago
    Quoted from Spudgunman:

    I got a leon in a system 9 flashing 0/7 if I run the memory test on 7 it will complete, if I run the memory test on 0 it wont. threw a another ram chip in same results.. thoughts?

    Weird. Maybe burn another sys9 EPROM to rule out an issue with the EPROM itself? I don't know exactly what Leon's test does to check the RAM, other than writing to it and doing a checksum. So it shouldn't matter whether it is on 0 or 7, perhaps some logic or something else is screwy on the board. That's a tough one.

    #5 9 years ago

    Leon's memory tests were not intended to check the whole ram area. He did a check if the memory chip is present by writing $aa and $55 to the last address and reading it back. Then he did a simple check if all addresslines were present by checking the presence for the same data at other addresses. If the last address was $7ff, the checks were at $6ff, $5ff, $3ff, $77f and so on (simulating a missing addressline each time). It is effective but an SRAM can still have defective cells at other addresses, it is a weak spot in Leon's test ROM's. It also doesn't work if 2 addresslines are missing (due to a bad repairjob /broken traces for example).

    There is a transistor in the /CE (chipselectline), you might check that one.

    #6 9 years ago

    I will try reburning leon as well..

    yea the transistor 3904 test ok on board. I will check it further anyway.

    #7 9 years ago

    Good info MarAlb.

    #8 9 years ago
    Quoted from thedefog:

    Good info MarAlb

    Thanks, you are welcome. I was in contact with Leon several years ago, when he was still alive. I asked him to rework the Gottlieb system80 test ROM because i found the results to be unreliable. He was aware of that but he told me he was ill and didn't have the time for that anymore.... I asked him if he didn't have an objection if I was going to write a test ROM myself. His response was something like "yeah, right...." I did learn to program 6800 assembly in the eighties so I freshed up my brains and started assembling. I showed Leon a piece of my code and he liked that and gave me a few tips. He asked me "is your program passing the memory test if you lift an addressline??". Well, it did..(#$#!)....because if an addressline is missing you are indeed writing and reading an address but not the address you think you are writing to and reading from. Then he told me about his "missing addressline strategy". I took another approach myself because i found it to be important to check the whole memory range in a memorychip. After I completed my system80 testrom (and I am still improving it when having new ideas) I created many more, GTB system 3, Williams system3-6, 7, 9, Data East, Gottlieb sound, Bally 2518, ..... It is really a shame Leon passed away, it would be nice to discuss with him about testcode.....

    Sorry for being off-topic....

    Marco

    #9 9 years ago
    Quoted from MarAlb:

    Thanks, you are welcome. I was in contact with Leon several years ago, when he was still alive. I asked him to rework the Gottlieb system80 test ROM because i found the results to be unreliable. He was aware of that but he told me he was ill and didn't have the time for that anymore.... I asked him if he didn't have an objection if I was going to write a test ROM myself. His response was something like "yeah, right...." I did learn to program 6800 assembly in the eighties so I freshed up my brains and started assembling. I showed Leon a piece of my code and he liked that and gave me a few tips. He asked me "is your program passing the memory test if you lift an addressline??". Well, it did..(#$#!)....because if an addressline is missing you are indeed writing and reading an address but not the address you think you are writing to and reading from. Then he told me about his "missing addressline strategy" I took another approach myself because i found it to be important to check the whole memory range in a memorychip. After I completed my system80 testrom (and I am still improving it when having new ideas) I created many more, GTB system 3, Williams system3-6, 7, 9, Data East, Gottlieb sound, Bally 2518, ..... It is really a shame Leon passed away, it would be nice to discuss with him about testcode.....
    Sorry for being off-topic....
    Marco

    OP could probably use your Sys9 ROM Marco...

    Marco has picked up the torch where Leon sadly left off. I've used his tools and he continues to evolve them. It's good to have him around here.

    viperrwk

    #10 9 years ago

    if you have a new test tool for sys9 happy to test it out here

    #11 9 years ago
    Quoted from viperrwk:

    OP could probably use your Sys9 ROM Marco...
    Marco has picked up the torch where Leon sadly left off. I've used his tools and he continues to evolve them. It's good to have him around here.
    viperrwk

    Thanks!

    Quoted from Spudgunman:

    if you have a new test tool for sys9 happy to test it out here

    Please send me a PM with your e-mail address. No problem to send you the images.

    Marco

    2 weeks later
    #12 9 years ago
    Quoted from MarAlb:

    Thanks, you are welcome. I was in contact with Leon several years ago, when he was still alive. I asked him to rework the Gottlieb system80 test ROM because i found the results to be unreliable. He was aware of that but he told me he was ill and didn't have the time for that anymore.... I asked him if he didn't have an objection if I was going to write a test ROM myself. His response was something like "yeah, right...." I did learn to program 6800 assembly in the eighties so I freshed up my brains and started assembling. I showed Leon a piece of my code and he liked that and gave me a few tips. He asked me "is your program passing the memory test if you lift an addressline??". Well, it did..(#$#!)....because if an addressline is missing you are indeed writing and reading an address but not the address you think you are writing to and reading from. Then he told me about his "missing addressline strategy". I took another approach myself because i found it to be important to check the whole memory range in a memorychip. After I completed my system80 testrom (and I am still improving it when having new ideas) I created many more, GTB system 3, Williams system3-6, 7, 9, Data East, Gottlieb sound, Bally 2518, ..... It is really a shame Leon passed away, it would be nice to discuss with him about testcode.....
    Sorry for being off-topic....
    Marco

    This is a great story. People that are good at programming in machine code blow my mind. Much respect for assembly heads (ass-heads)! I try to, the best I can do are some cheesy hacks to NES ROMs and some other simple stuff. Just shifting from big endian to little endian is enough to make my head hurt. I understand how to read assembly, just don't understand it well enough to write it. I did get the original Metroid Source code to compile with some modifications, that was my only proud moment. Every other attempt has been a fail.

    Reply

    Wanna join the discussion? Please sign in to reply to this topic.

    Hey there! Welcome to Pinside!

    Donate to Pinside

    Great to see you're enjoying Pinside! Did you know Pinside is able to run without any 3rd-party banners or ads, thanks to the support from our visitors? Please consider a donation to Pinside and get anext to your username to show for it! Or better yet, subscribe to Pinside+!


    This page was printed from https://pinside.com/pinball/forum/topic/system9-and-memory-check-strangeness and we tried optimising it for printing. Some page elements may have been deliberately hidden.

    Scan the QR code on the left to jump to the URL this document was printed from.