Re: DECstation 2100: power trouble?

[Mouse <mouse%Rodents-Montreal.ORG@localhost>]

>> I installed [a mouse] jumper.  I had to reset the machine a few
>> times (and/or let it sit powered on; I did both, but don't know what
>> is responsible for it working), but eventually the LEDs went to 7f
>> instead of ef; it then POSTed successfully:
> Excellent.

Well...I may have spoken too soon.  See below.

>> It now netboots into some NetBSD/pmax, whatever I had in my leftover
>> netboot setup.

>> I don't know why I didn't succeed before.  I must have not done
>> whatever magic sequence got it past getting stuck at ef before.

> It might have just needed a little time to "wake up". :-)

I speculate that it's thermal.  There are two chips that get hot (more
than "warm") to the touch: the CPU and a (heatsinked!) chip on the
other side of the CPU from the RAM slots.  Them being hot seems to
correlate positively with it being closer to working.

But last time I turned it on, even after leaving it I was unable to get
it to past EF on the lights.  The power-on before that ended with

KN01 V7.0 

7..6..5..4..3..2..1..0

24Mb..........0
>> boot -f tftp()
1845632
Excptn:    <vtr=NRML>
Excptn pc: 0xbfc0dc6c
Creg:      0x1000001c<CE=1
Excptn:    <vtr=NRML>
Excptn pc: 0xbfc0f7dc
Creg:      0x3000001c<CE=3
Excptn:    <vtr=NRML>
Excptn pc: 0xbfc0f7dc
Creg:      0x3000001c<CE=3
Excptn:    <vtr=NRML>
Excptn pc: 0xbfc0

which I do not consider encouraging.

I even pulled all but two sticks of RAM (after the above, obviously).
I've considered the possibility that one of those sticks is bad; I just
turned it on again and if I can't get it to POST I'll swap them for two
of the other sticks.

>> This is very encouraging!  It's a bit finicky to get started, but
>> once it works it seems to actually work.

Well...briefly.  I'm going to have to see if I can get it to do any
better.

>> Thank you very much, both of you!

> My contribution was minimal, but I'm happy to assist.  It's always
> nice to hear about one of these machines coming back to life.  The
> DECstation line are great little machines.

A couple decades ago, when I was living in Montreal, I had something
like 12 ports across 8 CPU architectures (exact numbers forgotten, but
something like that) running or ready-to-run, needing nothing more than
plugging in and turning on.  From memory: alpha amd64 hp300 i386 mac68k
macppc next68k pmax shark sparc sparc64 sun3 vax, of which four (hp300,
ma68k, next68k, sun3) are 68k - and I've obviously forgotten somthing
because there was at least one more CPU architecture which was multiply
represented.

Now...I recently realized I now have sparc, shark, i386, amd64, and
sparc64 and nothing else.  I've now got a VAX 3100/80 about three days
into a 1.4T diskless build-of-the-world; I'm hoping it survives and
builds a runnable system (I know my 1.4T *can* self-host on VAX; it did
on my emulated MicroVAX-II).  I have two Alphas, but it's been long
enough I don't know whether either is in bootable shape.

So I'd love to have a working MIPS machine.  But I'm having trouble
getting this one to behave.

It's still stuck at ef.  I'll try swapping in two other sticks of RAM.

/~\ The ASCII				  Mouse
\ / Ribbon Campaign
 X  Against HTML		mouse%rodents-montreal.org@localhost
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B