Loading ...
Sorry, an error occurred while loading the content.
 

re: [midatlanticretro] Re: Using 2SIO to download BASIC into an Alair without a disk drive or cassette.

Expand Messages
  • Bob Applegate
    ... Whoa! Run a real memory test! I m assuming the memory is static, but there are lots of different kinds of tests that look for refresh problems on dynamic
    Message 1 of 15 , Feb 26, 2010
      Bill Degnan <billdeg@...> wrote :


      > I am not using a terminator.
      > My memory tests so far have all been manual read/write/read tests at each
      > page location 0000, 1000 - F000. I have swapped RAM cards into a number of
      > configurations, and I usually get the same result. It's possible that my
      > laptop serial port is not right? Not sure yet.
      > cd

      Whoa! Run a real memory test! I'm assuming the memory is static, but there are lots of different kinds of tests that look for refresh problems on dynamic memory, as well as shorted address/data bits, bad bits in chips, etc. You wouldn't believe how many memory problems slip through because of tests that don't run sufficiently complex algorithms.

      I'd also test one board at a time, or at least pull boards you don't need. You've got four 16K boards, right? Pull all but one and check the portion not used by the test code. Once it seems to work, put in another board and test all 16K of it using a really good (extensive) memory test. Now you've got a "golden board" that is known, or at least presumed, to be good.

      Bob


      ___________________________________
      NOCC, http://nocc.sourceforge.net
    • Bill Degnan
      ... might find some flaky memory chips. ... card during this test. ... I am not using a terminator. My memory tests so far have all been manual read/write/read
      Message 2 of 15 , Feb 26, 2010
        >
        >
        > Just checking if your terminator card is plugged in ?
        > If it is, it could still be acting up.
        >
        > Maybe if you run a memory tests on with all the Ram cards plugged in you
        might find some flaky memory chips.
        >
        > Even some suspicious memory location could be corrupted by the terminator
        card during this test.
        >


        I am not using a terminator.
        My memory tests so far have all been manual read/write/read tests at each
        page location 0000, 1000 - F000. I have swapped RAM cards into a number of
        configurations, and I usually get the same result. It's possible that my
        laptop serial port is not right? Not sure yet.
        cd
      • Bill Degnan
        ... each ... number of ... my ... there are lots of different kinds of tests that look for refresh problems on dynamic memory, as well as shorted address/data
        Message 3 of 15 , Feb 26, 2010
          >
          >
          > > I am not using a terminator.
          > > My memory tests so far have all been manual read/write/read tests at
          each
          > > page location 0000, 1000 - F000. I have swapped RAM cards into a
          number of
          > > configurations, and I usually get the same result. It's possible that
          my
          > > laptop serial port is not right? Not sure yet.
          > > cd
          >
          > Whoa! Run a real memory test! I'm assuming the memory is static, but
          there are lots of different kinds of tests that look for refresh problems
          on dynamic memory, as well as shorted address/data bits, bad bits in chips,
          etc. You wouldn't believe how many memory problems slip through because of
          tests that don't run sufficiently complex algorithms.
          >
          > I'd also test one board at a time, or at least pull boards you don't
          need. You've got four 16K boards, right? Pull all but one and check the
          portion not used by the test code. Once it seems to work, put in another
          board and test all 16K of it using a really good (extensive) memory test.
          Now you've got a "golden board" that is known, or at least presumed, to be
          good.
          >

          Agreed, but I am trying to get programs into the system - chicken and egg
          problem here. I will test the RAM using a system that can already load
          programs.

          Bill
        • Bob Applegate
          ... Maybe I missed something but the subject implies you re loading a BASIC interpreter. That s much larger than a memory test and it won t test memory. Use
          Message 4 of 15 , Feb 26, 2010
            Bill Degnan <billdeg@...> wrote :


            > Agreed, but I am trying to get programs into the system - chicken and egg
            > problem here. I will test the RAM using a system that can already load
            > programs.

            Maybe I missed something but the subject implies you're loading a BASIC interpreter. That's much larger than a memory test and it won't test memory. Use a smaller program that can test your hardware before trying to load something larger.

            You're still using unknown RAM, but a memory test is a lot smaller and will use less unknown RAM to run.

            The other thing is that I seem to remember your failure happened at a page boundary. How big are the RAM chips and what was the address again? Maybe it happened on a chip boundary.




            ___________________________________
            NOCC, http://nocc.sourceforge.net
          • Bob Applegate
            ... If the hex address ends in 00 then it s a page boundary. What was the address where it failed and what type of RAM chips are on the boards?
            Message 5 of 15 , Feb 26, 2010
              Bill Degnan <billdeg@...> wrote :


              > I can't tell if it's a page boundary or not.

              If the hex address ends in 00 then it's a page boundary. What was the address where it failed and what type of RAM chips are on the boards?


              ___________________________________
              NOCC, http://nocc.sourceforge.net
            • Bill Degnan
              ... interpreter. That s much larger than a memory test and it won t test memory. Use a smaller program that can test your hardware before trying to load
              Message 6 of 15 , Feb 26, 2010
                >
                > Maybe I missed something but the subject implies you're loading a BASIC
                interpreter. That's much larger than a memory test and it won't test
                memory. Use a smaller program that can test your hardware before trying to
                load something larger.
                >
                > You're still using unknown RAM, but a memory test is a lot smaller and
                will use less unknown RAM to run.
                >
                > The other thing is that I seem to remember your failure happened at a
                page boundary. How big are the RAM chips and what was the address again?
                Maybe it happened on a chip boundary.
                >

                I can't tell if it's a page boundary or not.

                I am working through this web site:
                http://www.solivant.com/altair_bootloaders/index.php?album=altair_bootloader
                s&pagen=1

                I will keep at it. I have followed every page closely including the
                troubleshooting section.

                Bill
              • Dan Roganti
                ... Anytime you add more loads on that bus, by adding more cards, it s wise to use a terminator card. The signal quality begins to degrade by the consequence
                Message 7 of 15 , Feb 26, 2010
                  Bill Degnan wrote:
                  > I am not using a terminator.
                  > My memory tests so far have all been manual read/write/read tests at each
                  > page location 0000, 1000 - F000. I have swapped RAM cards into a number of
                  > configurations, and I usually get the same result. It's possible that my
                  > laptop serial port is not right? Not sure yet.
                  >

                  Anytime you add more loads on that bus, by adding more cards, it's wise
                  to use a terminator card. The signal quality begins to degrade by the
                  consequence of electrical reflections between all the drivers from more
                  cards on the bus.

                  I uploaded some memory tests that I found online to here. One of them is
                  short that you can thumb-in. It's in assembly still, it's not a
                  comprehensive test but it's a start, under folder S-100 Test Code.

                  =Dan
                • Bill Degnan
                  I will take a look. I may not have a terminator that works. I have to check bd ... Alair without a disk drive or cassette. ... each ... number of ... my
                  Message 8 of 15 , Feb 26, 2010
                    I will take a look. I may not have a terminator that works. I have to
                    check
                    bd

                    -------- Original Message --------
                    > From: "Dan Roganti" <ragooman@...>
                    > Sent: Friday, February 26, 2010 2:24 PM
                    > To: midatlanticretro@yahoogroups.com
                    > Subject: Re: [midatlanticretro] Re: Using 2SIO to download BASIC into an
                    Alair without a disk drive or cassette.
                    >
                    > Bill Degnan wrote:
                    > > I am not using a terminator.
                    > > My memory tests so far have all been manual read/write/read tests at
                    each
                    > > page location 0000, 1000 - F000. I have swapped RAM cards into a
                    number of
                    > > configurations, and I usually get the same result. It's possible that
                    my
                    > > laptop serial port is not right? Not sure yet.
                    > >
                    >
                    > Anytime you add more loads on that bus, by adding more cards, it's wise
                    > to use a terminator card. The signal quality begins to degrade by the
                    > consequence of electrical reflections between all the drivers from more
                    > cards on the bus.
                    >
                    > I uploaded some memory tests that I found online to here. One of them is

                    > short that you can thumb-in. It's in assembly still, it's not a
                    > comprehensive test but it's a start, under folder S-100 Test Code.
                    >
                    > =Dan
                    >
                    >
                    >
                    >
                    >
                    >
                    > ------------------------------------
                    >
                    > Yahoo! Groups Links
                    >
                    >
                    >
                  Your message has been successfully submitted and would be delivered to recipients shortly.