Hardware Meeting 3/21/89
                                 J.Ray 3/22/89

         Attendees: Mike Albaugh, Chris Downend, Tim Hubberstey, Pat
              McCarthy, Glenn McNamara, Rich Moore, John Ray, Byron
              Sheppard, Doug Snyder, Gary Stempler.

         Byron Sheppard was introduced as a new hardware engineer in
         Brad Fuller's audio group. Welcome Byron!

         The latest 1 Meg EPROM info from TI is: They are not in
         production yet. Parts will be available in the summer for
         $18-20. They will not be cost-effective for us until early
         1990 when they should cost about $14.

         Pat McCarthy reported that he is 90% done with the schematic.
         He will be implementing most of the chip in TTL to keep the
         Thunderjaws project progressing. The expected size of the
         chip is 3000 gates. He is hoping to get the Priority and
         Shifter chips done for Tjaws as well, but is not sure yet if
         this is possible.

         Doug and Mike reported that they will probably be visiting
         Mississippi next week to check on the design of the ASAP
         chip. Proto chips are expected late May or June. The first
         two boards for the development system (Mother and 68000)
         should be ready in May. The next two boards will be the RAM
         board and the Pod board. The Analyzer will be wire-wrapped
         before it goes to PC. The ASAP development system will be
         packaged in a "baby AT" case due to its cost-effectiveness
         and readily available components (eg. power supplies). Mike
         is working on the software, and will need an AMS for the
         68000 development. The ASAP development system software
         committee has written the first pass report on what they
         would like to see in the system.

         Pat will be writing up a purchase req. for Data I/Os ABEL
         software for the VAX. This will be an upgrade from the PALASM
         software currently used.

         Pat will arrange a demo of Viewlogic schematic capture and
         simulation software for the PC.

         We then discussed moving towards the JAMMA standard edge
         connector. Chris gave a synopsis of the reasons Atari needs
         to move towards this standard: The only games that are
         selling now are blockbuster games (Hard Drivin') and kits.
         Upright games in the $2000 price range are just not selling.
         This change in the market demand requires us to adapt our
         games to the kit market. The kit market is dominated by the
         JAMMA standard.

         JAMMA PROS

              Increased kit sales
              Possibly increased whole game sales
              Economies of harness standardization:
                   Engineering
                   Manufacturing
                   Field service (test fixtures)

         JAMMA CONS

              Increased PCB costs
              Increased connection costs
              Increased FCC problems
              Reduced power supply capabilities
              Inflexible edge connections
              Reduced audio volumes
              Changes "Standard Audio" methods (increased engineering)

         Doug Snyder agreed to coordinate documenting the Atari JAMMA
         specification. This specification will cover:

              Edge connector pinouts
              Video timing, polarity, levels
              Audio levels
              Power supply requirements

         Byron Sheppard agreed to come up with a standard audio amp
         configuration (for use on JAMMA games) that is cheap and uses
         readily available components.

         We discussed monitor usage on JAMMA games. The JAMMA edge
         connector only provides for -CSYNC. However, Doug said that
         the kits he has seen also provide taps somewhere on the PCB
         for separate positive syncs. He suggested we do the same (as
         a minimum) on our boards (on Tetris, he provided a separate
         Atari standard monitor header on the PCB). We should also
         make sure that the monitors we will be using will accept
         -CSYNC. (Some monitors accept -CSYNC on both -HSYNC and
         -VSYNC terminals). We should NOT use (on our JAMMA compatible
         cabinets) monitors that only accept positive sync signals.

         Regarding current and future games:

              TETRIS and TETRIS cocktail are both JAMMA compatible.

              ESCAPE upright will not be a JAMMA cabinet.
              ESCAPE kit will have an adaptor board and minor mods to
              the SA audio board to make it JAMMA compatible.

              SKULL & CROSSBONES main board will be re-designed to
              incorporate the JAMMA connector. The SA II board (with
              minor mods) will mount on top of the main board and have
              connections only to the main board.

              Games beyond these were not discussed. I would presume
              that all games currently in development except for
              S.T.U.N. Runner will be converted to JAMMA.
              Incorporating the audio board onto the main board should
              be seriously considered for cost savings (keeping the
              software drivers as compatible as possible, however).

         Please let me know if you have any additions or corrections.

         John