Date : Sun, 28 Sep 2008 12:20:23 -0500
From : jules.richardson99@... (Jules Richardson)
Subject: Domesday video problems
Adam Sampson wrote:
> Hi all,
>
> I'm currently trying to resurrect a Domesday system:
> http://offog.org/stuff/domesday/25092008119.jpg
>
> This belongs to the Computing Lab at the University of Kent, where I
> work.
Heh - are Mark Wheadon & co still there? I was at ukc in the early 90s, and
it's their fault I got into collecting old Acorn systems ;-) (I unearthed an
ARM Eval Kit and Teletext unit there; before that I'd been into Sinclair stuff)
> The system now mostly works: it'll boot off a Laserdisc, and you can
> navigate around happily. However, as you can see in the pictures,
> there's something wrong with the video output: the overlay graphics from
> the BBC are fine, but the genlocked video from the Laserdisc doesn't
> appear correctly:
> http://offog.org/stuff/domesday/25092008116.jpg
> http://offog.org/stuff/domesday/25092008118.jpg
Hmm, I don't have my service manual to hand (I'm in the US these days, and
it's still stuck in the UK), but you say you have a copy now which should
prove useful...
I seem to remember that there's a single board in these machines responsible
for video mixing - it'd be worth poking around there (I can't remember the
module number, but I think it's one of the three accessible from the underside
of the player in the bottom 'slice'). The VP-415 players are *horrible* for
suffering bad solder joints, and I've replaced a fair few capacitors in them,
too. Fantastic design for the time, but the PCB and component quality tends to
let them down.
Check in the service manual where the various video output taps come from, too
- it may be you can tap into a sensible video signal 'upstream' of the video
mix board which you can use to validate the video coming from the player.
Also check with the player's on-board diags. Details on how to kick the player
into diag mode (and the list of fault codes) should be in the service manual.
The built-in diags check the player functionality out pretty well.
Hazy memory (I don't have my AIV hardware with me, either) is telling me that
there's an adjustment for video signal on the rear of the machine too (a pot
accessed with a screwdriver via a tiny hole) - I think that's for final video
output though (so you wouldn't see any sensible data coming from the BBC
either) but it'd be worth a check.
> I believe the Laserdisc player was used as a video source for other
> experiments in the early 90s -- the SCSI ID was set differently, two of
> the Laserdiscs are marked "stills" and "video", and I've found
> framegrabs from them on other old machines.
Hmm, on a personal note, can you track down anyone who was involved? I collect
all sorts of Domesday system info (design details, marketing stuff etc.), so
it'd be really interesting to hear more there.
> It's therefore plausible
> that someone might have fiddled with other settings on the player -- or,
> since it's been sitting around for at least ten years unused, that a
> capacitor or belt needs replacing...
There's about twenty separate PCBs in the players, some of them connected via
direct socket connections and some of them via multi-way cables, so there's
plenty of scope for a dirty/intermittent connection. See comments above too
about dry joints and failing caps.
Taking the machines apart is an art - I actually wrote down some notes along
with photos of various stages of disassembly last time I did it which I'm
going to put online, but the text is all in a notebook back in England right
now :-(
If you really need to get your hands dirty, it's worth making up some
multi-way cables/plugs - then you can hang individual boards out of the player
and poke around with a 'scope whilst the player's running. It's really hard to
get at everything otherwise.
I can't remember for sure now what happens if the player can't spin up to
speed - my fuzzy memory's telling me that I did once try slowing the disc down
to see what happened and it gave a loss of sync. It looks like your player is
syncing properly, but is losing colour information somewhere in the mix stage.
cheers
Jules