• SynchroStream spotted in Olivetti M6-5x0

    From Tomas Slavotinek@21:1/5 to All on Sat Mar 5 11:56:48 2022
    Looks like the SSC was not exclusive to the Type 4 complex and the Reply TurboBoard 60/80 planar:

    https://ardent-tool.com/Olivetti/M6-5x0_Planar_Photo_Front.jpg

    This board was used in the MCA-based Olivetti M6-540/560 machines. The
    SSC P/N is the same as on the TurboBoard 60/80 planar. Interestingly
    enough, the SSC is on a daughterboard. I assume this was done to meet
    the CCGA-625 soldering and PCB requirements (process quality, number of
    layers, etc.). With this approach, the OEM could use a cheaper PCB with
    a lower layer count and put it together using the common soldering
    technology.

    The M6-5x0 Service Guide calls it the "Mad River Module": https://www.ardent-tool.com/Olivetti/Docs/service_guide/systems2/cap3.pdf#page=4
    (page 4 physical)
    Is that the SSC codename?

    I've moved the SynchroStream content to a separate page and updated it
    with the new info:
    https://ardent-tool.com/tech/SynchroStream.html

    The board also uses the IBM 10G4672 I/O Controller we know from late
    PS/2 machines. The PDF linked above refers to it as "Galaxy" and lists
    some previously unknown features. Like the 82077 support logic. It also confirms what I've figured out earlier by reverse-engineering the 9595
    2S2P planar. See here:
    https://ardent-tool.com/tech/ASIC_Info.html#10G4672

    All relevant system-specific pages were updated with links to the new pages.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Louis Ohland@21:1/5 to Tomas Slavotinek on Sat Mar 5 08:52:48 2022
    Huh, NCR 53C700-66
    MACH 32, Bt RAMDAC, 2MB VRAM?
    Dallas DS1387 Real Time RAMified 9351
    L2 COAST
    ICS PLL? Clock gen?

    On 3/5/2022 04:56, Tomas Slavotinek wrote:
    Looks like the SSC was not exclusive to the Type 4 complex and the Reply TurboBoard 60/80 planar:

    https://ardent-tool.com/Olivetti/M6-5x0_Planar_Photo_Front.jpg

    This board was used in the MCA-based Olivetti M6-540/560 machines. The
    SSC P/N is the same as on the TurboBoard 60/80 planar. Interestingly
    enough, the SSC is on a daughterboard. I assume this was done to meet
    the CCGA-625 soldering and PCB requirements (process quality, number of layers, etc.). With this approach, the OEM could use a cheaper PCB with
    a lower layer count and put it together using the common soldering technology.

    The M6-5x0 Service Guide calls it the "Mad River Module": https://www.ardent-tool.com/Olivetti/Docs/service_guide/systems2/cap3.pdf#page=4
    (page 4 physical)
    Is that the SSC codename?

    I've moved the SynchroStream content to a separate page and updated it
    with the new info:
    https://ardent-tool.com/tech/SynchroStream.html

    The board also uses the IBM 10G4672 I/O Controller we know from late
    PS/2 machines. The PDF linked above refers to it as "Galaxy" and lists
    some previously unknown features. Like the 82077 support logic. It also confirms what I've figured out earlier by reverse-engineering the 9595
    2S2P planar. See here:
    https://ardent-tool.com/tech/ASIC_Info.html#10G4672

    All relevant system-specific pages were updated with links to the new
    pages.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Tomas Slavotinek@21:1/5 to Louis Ohland on Sat Mar 5 21:43:38 2022
    Yeah, it's a relatively well equipped board. Only 1 MB VRAM but that's
    still decent. No clue how good the NCR SCSI controller is. ICS is a
    clock gen. chip - there are two of them actually (system and video).

    And it was made by Reply.

    Working on the outline...

    On 05.03.2022 15:52, Louis Ohland wrote:
    Huh, NCR 53C700-66
    MACH 32, Bt RAMDAC, 2MB VRAM?
    Dallas DS1387 Real Time RAMified 9351
    L2 COAST
    ICS PLL? Clock gen?

    On 3/5/2022 04:56, Tomas Slavotinek wrote:
    Looks like the SSC was not exclusive to the Type 4 complex and the
    Reply TurboBoard 60/80 planar:

    https://ardent-tool.com/Olivetti/M6-5x0_Planar_Photo_Front.jpg

    This board was used in the MCA-based Olivetti M6-540/560 machines. The
    SSC P/N is the same as on the TurboBoard 60/80 planar. Interestingly
    enough, the SSC is on a daughterboard. I assume this was done to meet
    the CCGA-625 soldering and PCB requirements (process quality, number
    of layers, etc.). With this approach, the OEM could use a cheaper PCB
    with a lower layer count and put it together using the common
    soldering technology.

    The M6-5x0 Service Guide calls it the "Mad River Module":
    https://www.ardent-tool.com/Olivetti/Docs/service_guide/systems2/cap3.pdf#page=4
    (page 4 physical)
    Is that the SSC codename?

    I've moved the SynchroStream content to a separate page and updated it
    with the new info:
    https://ardent-tool.com/tech/SynchroStream.html

    The board also uses the IBM 10G4672 I/O Controller we know from late
    PS/2 machines. The PDF linked above refers to it as "Galaxy" and lists
    some previously unknown features. Like the 82077 support logic. It
    also confirms what I've figured out earlier by reverse-engineering the
    9595 2S2P planar. See here:
    https://ardent-tool.com/tech/ASIC_Info.html#10G4672

    All relevant system-specific pages were updated with links to the new
    pages.


    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)