Re: Help: AVG state machine (BZ / RB / Tempest)

> The ~VGGO comes into the clear pin of a ls74, the data for this flip-flop is
> OP0 (outputs HALT and ~HALT) and it is clocked with ~STROBE3. ~STROBE3
> comes from an address decoder based on the state machine (back to that dang
> sm).

I have a bunch of info on the PROM at home. I dumped it and figured out each state and
made a map of it.
  so it sounds like your state machine is in a state of disarray I guess. ;-)

> ~VGGO also goes to another ff where it is the SET signal, the output of that
> is AVG0. But the data from it is based on STATE2 and I believe AVG0 never
> changed.

I belive that this FF simply toggles the least significant bit of the vector address bus
(every other instruction of course). It is on the 'set' pin because the vector
generator actually executes addresses like this...

  4001
  4000
  4003
  4002 ...etc...

So I actually was wrong when I said the VG starts at VGRAM,0x000. It truly starts at
VGRAM,0x001
 So, STATE2 would then be the next state of AVG0. All this is clocked by the VGCLOCK
right? (im running off memory here, no docs yet).

 When the VG starts running it should feed the byte in at VGRAM,0x001 to the State machine
so that it can start executing that instruction. Can you verify that the VG address bus is
resetting to VGRAM,0x001 and that the data from that location is showing up on the State
Machine PROM inputs? I think only the 3 or 4 most significant bits actually go to the
State machine PROM tho.
  If both those things aren't happening, then the problem should be external to the actual
State Machine PROM and beyond. Im not sure of what functions you can actually perform with
the 9010A in the VG circuit, it is all clock driven once it 'goes'.
  Im blabering now, I had better wait until I get some schems. ;-)

>
> David
>
> ----- Original Message -----
> From: Jess Askey <jess@magenta.com>
> To: <vectorlist@lists.cc.utexas.edu>
> Sent: Monday, August 23, 1999 9:14 AM
> Subject: Re: Help: AVG state machine (BZ / RB / Tempest)
>
> > Doesn' the VGGO line clear the HALT line via a direct flip-flop? If so,
> you should be able
> > to do a lot of this troubleshooting without the 9010A.
> > The basic process of the VGGO is....
> > - clear halt
> > - reset the VG address bus to VGRAM,0x000
> > - start executing VG code at VGRAM,0x000
> >
> > Im not sure what happens if you try and assert VGGO if the VG is not
> > halted tho. Normally the 6502 software needs to check for this before
> > asserting VGGO. Maybe your VG is not halted yet (for some unknown reason)?
> >
> >
> >
> > > David Shoemaker wrote:
> > >
> > > I am going to be dreaming about this damn thing I just know it.
> > >
> > > Working on a BZ boardset. Using my 9010 to download the test program
> from the cat box
> > > manual. When I do the VGGO nothing happens.
> > >
> > > I have read the program back out of memory and that is all ok. (I am
> starting to really
> > > like the 9010).
> > >
> > > As far as I can see the VGGO isn't clearing the HALT line because the
> OP0 and ~STROBE3
> > > lines are not doing the correct thing.
> > >
> > > As far as I can tell this is being caused by the state machine not doing
> anything. But
> > > I am getting pretty confused on its startup state. If I reset the board
> a few times I
> > > can get most anything I want out of the OPx lines. The STROBEx and
> LATCHx all stay
> > > consistently HIGH. Along with the STATEx all being low.
> > >
> > > Has anyone tried to decode the state prom and figure out what in the
> blazes is going on
> > > in here?
> > >
> > > Clay you built an AVG replacement which should be nothing more than the
> BZ / Tempest AVG
> > > right? So you MUST have figured it all out to make sure it worked?
> PLEASE tell me you
> > > know how to debug this stinking AVG.
> > >
> > >
> > > (low wimpering voice) Help me please.........
> > >
> > > I have 6 boardsets and they all have problems in the AVG. URG.
> > >
> > > I will collect all the stuff I have done with the 9010 and pass it on to
> Al for spies
> > > tonight. I have the AVG test app along with BZ memory map stuff etc
> plus some notes.
> > >
> > > David
> > >
> >
Received on Mon Aug 23 14:36:03 1999