On 2000-10-14 [EMAIL PROTECTED] <John> said:
>> Odd thing about it is that as long as I'm working along with the
>> machine, nothing happens -- it's only when the machine is idle for
>> some period of time that it acts this way.
>Sounds like your "dynamic refresh" of RAM is failing. (The 8088
>- in fact, all Intel CPUs - automatically produce a refresh cycle
>for RAM. (A dummy address-refresh cycle reads all addresses
>sequentially, but does nothing with the information read. It is
>transparent to the user, and is there ONLY to refresh dynamic RAM.)
Thank you for explaining the dynamic address-refresh cycle. I suspect
that, if this same problem occurred to the previous owner of this
machine, it may have been the reason why it was taken out of service.
That's too bad, though, because even with that problem, one can just
be sure to save material frequently while working. [Just a moment
ago, as a matter of fact, this machine -- which is the XT in question
-- experienced a freeze while I was writing this too you; I had
saved enough material to resume later.] Does this problem indicate
that more is necessary than just re-seating the CPU chip and RAM
chips?
>I have no idea what the "(s)" means, but the "60000" is hex address
>where parity went bad. Bring up debug, then WRITE "00" (not "FF")
>at 6:0000 - 6:FFFF. Wait 30 seconds or so, then DUMP 6:0000.
>If they don't show "00" in all locations, that block is losing it.
>(Don't use "FF" - that's the response of empty memory blocks, as in
>"E000:0"...)
>Best bet: yank out all of the RAM chips (use a tiny, tiny
>screwdriver to pry them up), then carefully reinsert them.
>(Straighten any bent pins before sticking them back in, then insert
>ALL the pins in one side before squeezing the second side for
>insertion. Check all the pins with a magnifying glass before
>finally pressing straight down with your thumb... This check is
>the MOST IMPORTANT part; never push them back in without doing so -
>because you are guaranteed to bend at least a few pins on the way
>in... especially the thinner ones sometimes used at the chip's
>corners.) This removal/reinsertion will "wipe" the pins clean.
>There is no need to remove the dark tarnish from the upper parts of
>the pins.
Thank you for these careful instructions! I follow through with them
as soon as I find the time (probably in a few weeks).
Jerry -- on his PC/XT
To unsubscribe from SURVPC send a message to [EMAIL PROTECTED] with
unsubscribe SURVPC in the body of the message.
Also, trim this footer from any quoted replies.
More info can be found at;
http://www.softcon.com/archives/SURVPC.html
