On Thu, 10 Jan 2002, Marko Mäkelä wrote:
> > Hey - here's an idea. What about a V364 webcam style device - video
> > capture on the S-Video output and some form of telnet interface to the
> > keyboard buffer (perhaps over RS232 using the real ACIA) - log into the
> > linux box controlling the device and send it "SAY" commands of your own?
> > Only one user at a time of course - have to implement some form of
> > concurrency on /dev/v364 :)
>
> Yep, only one user can be allowed to write commands at a time, while
> others can watch and hear the output.
>
> But would you dare to install the machine in continuous operation?
Well - first of all let me make clear, this subject is entirely open for
cbm-hackers discussion. If this auction is won, I will consider the
machine the property of the mailing list (if mailing lists can own
things), not my own. Thus anything that happens to the machine will only
take place with the general consent of the list.
With regard to the machine itself, I would like to see it (or at least,
the more resilient parts of it) in operation. In my third year at
University, I had a plus/4 in the corner of my college room, left running
a Hungarian TED sound demo for what was literally months. The machine ran
without crashing once, which illustrates a point about old semiconductor
reliability. Meanwhile, my x86/Intel things had crashed repeatedly, of
course...
What chips don't like is being turned on and off regularly - the expansion
and contraction with temperature damages the IC because silicon, silicon
dioxide and aluminium metal layers all expand and contract at different
rates - thus you have stresses inside the chip every time it changes
temperature.
If you leave a chip either permanently on, or permanently off, less damage
is done. However, in the climate of most nations, ambient temperature
varies quite a lot between winter and summer, so there is actually *less*
temperature variation if the chip is permanently on, because it sits at a
higher temperature than its surroundings (determined by heat flow
considerations) which is more consistent than having the chip heat and
cool in direct correlation with ambient temperature, which is obviously
the case if it is off.
Now as to the V364, the really "priceless" parts in it are the prototype
TED, 7501, ACIA, EPROMs (to a lesser extent) and of course that 8706
speech logic chip. All of these can be replaced with parts from other 264
series machines, except for the 8706. There is certainly a case for
removing the original chips in the V364 and replacing them with production
parts, at least until the reverse-engineering project is complete and
everyone has decided what should happen to it.
I myself would like to see it in operation, as that is (for me) the best
alternative to being packed away out of sight in a "collection". It would
be great to have webserver and user logs for v364 activity, and to be able
to stream mp3-compressed audio and jpeg-compressed video direct from the
machine, to *anyone* in the world who is interested. That's the next-best
thing to everyone being able to own a V364.
Some form of menu system would be good, with the native 261 voices of the
V364, the phrases from Gorf and Wizard of Wor ("YOUR BONES WILL LIE IN THE
DUNGEONS OF WOR!!") and those from A, Bee, Cs (if they are different from
Magic Voice defaults). I haven't worked out how to grant users the ability
to run arbitrary code on the machine (only one interrupt and one mode of
operation doesn't leave a lot of scope for safety) but if this is at all
possible I would do it.
> I'll assist you with the software (a 6551-based client for cbmlink with
> some additions) and also financially (50 EUR) if you win the bid.
Thanks for the support, Marko - and thankyou to everyone who has sent
mails on this subject. It may be of some concern to you that the highest
bid is still in the name of the other bidder.
Don't worry about this. I am planning to "snipe" this auction right at the
very last "minute" ("second" would be more accurate here) as that is the
surest way to win eBay auctions.
Thanks for reading thus far, and as ever, comments and suggestions are
welcome :)
Richard
Message was sent through the cbm-hackers mailing list
Archive generated by hypermail 2.1.1.