> I don't think this is a real VME bus. Maybe based on it, maybe 2 of the 4
> backplane connectors are VME-based. I don't know. What i found out for now:

Any pictures of this available?

> The power supply is capable of a load of 5V/180A, 12V/26A, 12V/3,5A. nice
> values. This is the reason for the very strange power connector of these
> machines. The other thing: It hat one 88100/25MHz and 4 PMMUs. The RAM (32MB)
> consists of several hundred chips, which are really needed completely, so RAM
> control will be a very big performance gain here. Every I/O card has it's own
> 68020 on it, so also here, the performance will be very nice.

180A for 5V, neat. :-) What strikes me the most is 4 PMMUs. They must be
doing
some sort of black magic. Having a 68020 driving a VMEbus card is
common.

> But maybe... we need to write kernels for the I/O hardware, as i saw this on
> the PCS Cadmus machines i have standing around.

I'm not exactly sure what you mean here. Actually booting *BSD on an I/O
card
or writing modules to drive the I/O hardware? In either case everything
needs
to be done. I have one VMEbus 88k system but I am missing the piece that
allows
me to power the system up. (Gnar!)

> For the Triton: I heard, only 250 of these were made, i have 2, and i just
> checked it... they have the serial numbers 10124 and 10125, but i got them from
> different sources... Seems to be very rare, really.

If thats the case then it may be an only you thing. If you want it to
boot *BSD
you're gonna have to write the code.

> Currently... hacking the pinout of the RJ45 console connector :) But the beast
> boots and talks to me, it just doesn't listen to me on the console...

Good luck. I'm guessing you didn't get much documentation with this?

-Matt