==========================================================================

       The Evolving Convergent Technologies "Miniframe" FAQ

     Now expanded to include information about the "Mightyframe"

                           Version 0.003
    (still many questions without answers, but it's getting better!)

                   Begun: November 13, 1996
                   Completed: ?

                   Version 0.003: November 30, 1996.

               Maintained by R. D. Davis, rdd@digex.net

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Table of contents

0a.  What is a Miniframe?
0b.  What is a Mightyframe?
1.  What Miniframe/Mightyframe models other than the CG-1000 ever existed?
2.  Were any hard drive interfaces, other than the ST-506, used with the
    Miniframe?
3.  What are the following LEDs for?
4.  What are the following jumpers for?
5.  Did an expansion chasis exist?
6.  What operating systems are known to run on the Miniframe?  
7.  Other than the Archive 56637 tape drive and the ST-506 hard drive,
    what other mass storage devices were supported, or known to be hacked,
    for this machine?
8.  What are the maximum serial port speeds on the main board, serial I/O
    board?
9. Aside from the main board, 8-port serial I/O board, 512KB memory
   boards and Archive (model?) tape controller board, what other I/O
   boards are known to exist, or to have existed?
10. What were the Miniframe's connections to the AT&T UnixPC machines?
11. What are the, if any, interesting, or unique, details about this
    machine?
12. Approximately how many of these machines were produced?
13. Are schematics for this machine available from any known sources?
14. What hardware and software documentation exist for this machine?
15. What interesting, or otherwise, hardware or software, hacks are
    known to have been performed on this machine?
16. What is the 9-pin connector labeled "cluster" for?
17. What type of PSU was used? What are ratings for the PSU?  Was more 
    than one type of PSU used?
18. What is the male IDC connector, between the ST-506 data and control
    cables, on the main board, for? 
19. Are the PALs in the Miniframe copy-protected?
20. Was the Miniframe sold under any other manufacturer's labels
    (badge engineered)?
21. Was Ethernet networking ever supported in any way, or was support
    for it ever considered?  Can an AT&T UnixPC ethernet board be hacked 
    into this machine without too much trouble?
22. In what year did the manufacture of these machines begin?  In what 
    year did production cease?
23. Comparing the Miniframe's processing speed to other machines.
24. Is Convergent Technologies still in business?  Do they still
    provide parts and documentation for these machines?
25. I've just obtined a Miniframe.  What's the first thing that I
    should do?
26. Where can one obtain various bits of source code written
    specifically for the Miniframe?
27. Miniframe diagnostics tape(s).
28. CTIX information and hard drive formatting.
29. What are some hardware problems that may be encountered?
30. What resources for CT Miniframe information be found from the Internet?

Note: throughout the FAQ, the words main board and motherboard are
used interchangeably.  

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

0. What is a Miniframe?

   A Convergent Technologies "Miniframe" is a small computer system in a
   tower-type case (slightly smaller than a MiroVAX II in a BA23 tower
   case or an IBM PC-RT 6150), which uses a Motorola MC68010-10 
   microprocessor for the CPU.  

   This is a rather interestingly constructed machine.  One accesses the
   insides by splitting the cabinet in half.  The left side (viewed from
   the front) opens (two heavy hinges at the bottom of the side which
   opens out and down).  On the side that doesn't open are the PSU and
   tape drive, at the top of the cabinet, and the hard disk (I'll need to
   figure out how to mount it, as it was sitting loose in the cabinet) at
   the bottom.

     +--+
     |  |
     |  |
     |  |
     +--+
        -+      
  +      |
   \     |
    \    | 
     \+--+

   On the side which opens are the following circuit boards: main board,
   serial I/O, 3 memory boards and a tape controller.  What's interesting
   is the way that these boards are attached to each other and to the
   bus; They're stacked on top of each other, and the colons in the
   diagram below represent the interlocking 100-pin IDC connectors on
   each board.  The o's represent the threaded studs which hold the
   boards to the side of the cabinet (the bottom board is also attached
   by screws), secured at the top by thumb-nuts.  The bottom, and
   largest (15 by 18 inches!), board is the main board.  Next is the
   narrower serial I/O board, the three memory boards and then the tape
   controller board (as wide as the memory boards, but about half as long.  

       Top view:

       +------------------+ <---- main board
       |                  |  
       |------------------| <---- serial I/O board
       |------------------| <---- memory & tape drive boards
       |    o   |o    o   | 
       |        |       : | 
       |        |       : |
       |    o   |o    o   |
       |------------------|
       |                  |
       |                  |
       +------------------+

   Each memory board appears to contain four banks of 18 64K (I think --
   they're marked as 8431, HM4864AP-12) RAM ICs, and this being a
   16-bit bus machine, I would guess that this means that each memory 
   board is 512KB of RAM, with another 512KB on the main board, for a 
   total of 2MB.  The CPU is an MC68010.  This machine is somewhat simimlar 
   (although not physically!) to the AT&T 3B1/UNIX-PC, but lacks many of 
   the UNIX-PC's features, such as graphics, a built-in monitor, 
   built in modem, etc. 
   
   The main board has two serial ports, a parallel port and a port
   marked "cluster".  Other connectors on the main board include a
   power connector, hard drive connectors and a floppy drive connector.
   Here's a drawing of how the hard and floppy drives get connected:

                     20   20   34    34   <--IDC drive connectors
                  +--==---==---===---===-------------+
                  |  R    R    R     R         %%    | 
                  |                  |o|      power  |
   reset switch-> {              E2  |o|      conn.  ~ 
                  |                   o
                  | 
                  |
                  |              E1  o o
                  ~                  
              
 
   In the above sketch, 'R' denotes the red (or blue, etc.) stripe
   on the cable.  E1 and E2 are jumpers.

   Here's an explantion of the connectors in the above
   sketch:

   From the writings of Tom Trebisky, tom@as.arizona.edu

	"Start with the big power connector on the main board
	to get our bearings.  Moving from there back along
	the main board we get:

	34 pin hard drive cable (common to both drives)
	34 pin floppy cable (only one floppy supported)
	20 pin cable to HD0
	20 pin cable to HD1 (not used with a single drive)"

                        *   *   * 

   and,

   From the writings of Clarence Dold, dold@rahul.net:

  ">From the rear:
   20 pin control cable to hard disk one mounted in the rear.
   20 pin control cable to hard disk zero mounted in the front.
   34 pin to the floppy drive
   34 pin data cable daisy chain to drive 0 and drive 1.
   Pin 1 (red stripe) is to the rear on each of these."

                        *   *   * 
   
   There are 8 serial ports on the serial port board.

   The hard drive is a 70MB ST506 (Micropolis 1325), but I understand
   that in order to use two drives, an extra PSU should be added.  

   The tape drive controller board is labeled "Mini QICI B-60-375", which
   appears to be a QIC-02 interface board.  The tape drive is a
   full-height Archive model No. 56637 QIC-36 drive, with a ceramic
   read/write head (don't use long, high capacity, tapes with it!).
   There's a QIC-36 <-> QIC02 adapter board in a cage attached to the
   underneath of the tape drive.  Actually, this assembly is two
   half-height units attaced to each other, each with a power connector.

   Looking at the drive from the back, showing jumpers, LEDs, etc. (as
   usual, not drawn to scale):

     tape drive          adapter
   +------------------+-----------------+
   |   jumpers:       |                 |
   |     o |o|        |            [J2] |   J1 = jumper block B2 on PCB
   |    |o||o|        |                 |        ___
   |    |o| o         |          o LEDs |      o o o
   |                  |          o      |        --- 
   |     +==+         |     [J1] o | |  |      
   |     |  |         |          o | |  |   J2 = other jumper block on PCB
   |     |  |         |   conn.  +-+ |  |
   |     |  |         |    to    |   |  |    ---to back of unit-->
   |        |  |  red    | tape dr. |   |  |    o o|o|o o o o o|o|o|
   |     |  |  stripe |    PCB   |   |  |    o o|o|o o o o o|o|o|
   |     +==+ <----   |          +---+  |
   +------------------+-----------------+
    
   This machine is binary compatible with the AT&T UnixPC (7300 and 3B1)
   computers, to a certain extent.  Obviously, it can't use software
   which takes advantage of some of the UnixPC's hardare such as
   graphics, etc. and also can't use a version of GNU's gcc compiler
   built using the shared libraries.

   Correction to the above from Clarence Dold, dold@rahul.net:

     "You just need the shared library from the 3B1.  One file, 
     I forget what it's called, not the same as later SVR 
     versions.  The 3b1 was the first shared library implementation."

   The operating system software native to this machine is a System V,
   relese 2, flavor of UNIX called CTIX.

   ICs used in the Miniframe:

   Main board:

      Assorted 74F, 74S and 74LS TTL

      MC4044P
      MC1488 (2)
      MC1489 

      Intel P8274         +--------------------------------------------+
                          | * This is a guess, as the paper label with |
      HAL16 (12)          |   the convergent part number on it is not  |
      PAL16 (2)           |   very easy to remove without scraping it  |
      PAL20 (2)           |   off -- more difficult than the labels    | 
      HAL20 (3)           |   often used by other manufacturers.  I'll |
      PAL or HAL 16 (8) * |   eventually get around to removing and    |  
      PAL or HAL 20 *     |   checking these.                          | 
                          +--------------------------------------------+
      WD1010AL-00 ST-506 disk cotroller

      VTI 537 VD 8600 VL2797-PC (what's this?)
   
      From the writings of Clarence Dold, dold@rahul.net:

         "The Intel 2797 is a clone of a WD chip, I forget which, 
         as the standard floppy controller prior to the IBM-PC 
         arrival in 1981.  Convergent, founded by an Intel salesman 
         in 1978, favored Intel chips wherever possible." 

                               *   *   * 

      AMD 8253 (2)  - serial controllers
      AMD 8259                  +-------------------------------------+
      AMD 8435                  | ** Some sort of PAL/HAL type ICs, I |
      AMD AM2149-35DC (3)       |    think.                           |
      AMD AM26LS31              +-------------------------------------+

      MMI 8516, 8520, 8506, 8439  **

      (Signetics?) K8532, SCN2652A, C2N40F (40-pin) - what's this?

      DL15C301 (delay line?)

      2764 EPROMs (2)

      Motorola MC68010L10 microprocessor, 16-bit, 10MHz  68010.

      M5K4164 (64K RAM?) (72)

      Also two crystal oscillators: 19.660800 and 40.0 MHz


   IC's etc. on Serial IO board:

      74F, 74S and 74LS TTL
      PAL16 and PAL20 ICs (not counted)
      AMD P8253 (3)
      Intel P8274 (3)
      MC1489 (16)
      MC1488 (10) 
      
      19.660800 MHz crystal osc.
 
      Traces & outlines for the following unused components (approx ct); 
      I would guess that these could be fitted and this board could
      have more serial ports:
 
      48-pin DIP (1)
      10 pin SIP resistor packs (3)
      20-pin DIP (21)
      24-pin DIP, narrow (3)
      24-pin DIP, wide (1)
      resistors (10)
      capacitors (16) 
      CR1
      Y1
      15-pin leads for unused socket marked J1 -- what's this for?

   ICs on memory boards:
   
      74F and 74S TTL
      PAL16 and PAL20 (uncounted)
      MB8254A-15 (64K RAM?) (72)

   ICs on tape drive controller:

      74S, 74F, 74LS TTL 
      PALs (16 & 20)


   From the writings of Tom Trebisky, ttrebisky@as.arizona.edu:

       "A lot of what I had done is actually for the Miniframe
       (aka Motorola 6300, NOT the ATT 6300 which is a x86 box).  
       The miniframe is a sister machine to the 7300, same mmu 
       and much the same i/o structure.  The 7300 has a memory 
       mapped video.  The miniframe does not, but has a sort of 
       network port (2Mb/sec SDLC/HDLC) and a second hard drive."

                        *   *   * 
   
   From the writings of Hannu Aronsson, haa@cs.hut.fi:
   
      "I wanted to get UNIX at home too, and thus got into buying
       a surplus Convergent Miniframe computer (10MHz 68010, 512k
       RAM, 40M MFM disk with very plain vanilla System V) for 
       something like $lots with several terminals. After fixing a
       512k memory board to be a 2M memory board, I got the RAM in 
       the machine to the maximum, 2 megs. Anyway, this little 
       machine was a masterpiece of engineering and I learned a
       lot about UNIX too while hacking with it.
   
       Even if the machine was slow, it gave the single user (me) 
       more CPU than the faster time-sharing machines at the 
       university. Anyway, now looking back, it seems odd that 
       you were really comfortably able to run a GNU Emacs, GCC, 
       a few shells and nethack at the same time on a machine that 
       small."

                        *   *   * 

   From the writings of Petteri Jantti, jantti@juhani.decus.fi: 

      "Actually the Convergents I was using were sold under Motorla 
      label. I had Motorola 6350 (MiniFrame) and 6600 (MegaFrame). 
      Entertaining is a good word. The 6350 (and a 6300, too) was 
      actually quite good. It worked more or less as expected. Funny
      thing about the BBS thing. I was required to set up a BBS on 
      the beast. If my memory served I used UNaXcess as the software.
      Had uucp connection only. Later the machine was replaced with 
      the 6600. Nowm, that was an interesting machine. Only problem 
      with it was that the hardware was so flaky that it was more 
      down than up.

                        *   *   * 

0b.  What is a Mightyframe?

--------------------------------------------------------------------------

1. What Miniframe/Mightyframe models other than the CG-1000 ever existed?

From the writings of Tom Seabury, fd83@cityscape.co.uk:

"The other models in the dedicated CTIX admin booklets we have refer
to models S80, mini, S120, S220, S320 and S640.  We own an S320.  The
mini has been known as the "old" S80, its spec seems to be a reduced
version of the S80, the S80 offering 86020, 16.7MHz, 400Mb max
storage, 6Mb max ram, max 16 rs232 ports."

                        *   *   * 

--------------------------------------------------------------------------

2. Were any hard drive interfaces, other than the ST-506, used with the
   Miniframe?

From the writings of Clarence Dold, dold@rahul.net:

"No.  There was a driver for the removable SyQuest 5MB cartridge,
which was mounted as drive one, leaving the odd quirk that the machine
wants to boot drive one first, if one is available."

                        *   *   * 

--------------------------------------------------------------------------

3. What are the following LEDs for?

   A. LEDs on main board:
        red (nearest cluster conn.), yellow, green, yellow, red -- these are
        numbered, on the outside of the chasis, as 0 through 4.

        I've discovered that the following disply sequence occurs after
        the Miniframe is powered up:


                       Event, t-->          
         LED  -------------------------            
         ---  A  B  C  D  E  F  G  H  I
          4   *  *  *  *  *     *     
          3   *     *  *        *    (B)
          2   *  *  *  *  *  *  *  *  *
          1   *     *  *        *  
          0   *        *  *     *

      Where:

          A = power-up
          C = "seek-complete" (?) signal from hard drive
          E = tape drive begins moving tape
          I = I'm guessing that this is after all self-tests have completed
          
          * = LED is on
         (B)= LED is blinking  
          

        From the writings of Clarence Dold, dold@rahul.net:

      	"Pretty useless boot self test.  After that it was merely a
	keep-alive indicator.  On a machine this slow, it was nice to 
        know it was still working ;-)"

                        *   *   * 

    B. Green LED on serial I/O board  (60-00325, 64-00729BA)
    C. Green LED on memory boards (64-00702, B-60-00142)
    D. LEDs on tape drive board (MiniQICI B-50-375, 64-00954A);
         red - CR6, CR7, CR8, CR9
         green - CR2, CR3, CR4, CR5

--------------------------------------------------------------------------

4. What are the following jumpers for?
    - E1 and E2 on main board

    These jumpers are fitted as shown in the main board drawing in
    question 0a.

    - MA19 and MA20 on (64-00702, B-60-00142) memory boards

        From the writings of Clarence Dold, dold@rahul.net:

        "The memory could be remapped, putting expansion memory at
        address 00, so that defective motherboard memory could be
        diagnosed.  The other jumpers set the address range of each 
        of the memory boards in the stack... MA19 MA20, are address 
        bits 19 and 20."

                        *   *   * 

    - B2 and other two jumpers on tape drive and attached adapter board

--------------------------------------------------------------------------

5. Did an expansion chasis exist?

From the writings of Tom Seabury, fd83@cityscape.co.uk:

"Our S320 has an expansion chasis as well as an ethernet link card in
one of the standard slots."  

                        *   *   * 

--------------------------------------------------------------------------

6. What operating systems are known to run on the Miniframe?  

CTIX, CTOS (?), XINU and ?.  Does anyone know where any OS sources, or
parts thereof, for this machine can be found?

--------------------------------------------------------------------------

7. Other than the Archive 56637 tape drive and the ST-506 hard drive,
   what other mass storage devices were supported, or known to be hacked,
   for this machine?

From the writings of Clarence Dold, dold@rahul.net:

"The SyQuest 5MB hard drive, removable.  With "mtools", available 
from Emmett P. Gray, I was able to transport MSDOS files on 720K 
5-1/4" diskettes, formatted on a PC using fdformat.exe."

                        *   *   * 

--------------------------------------------------------------------------

8. What are the maximum serial port speeds on the main board, serial I/O
   board?

--------------------------------------------------------------------------

9. Aside from the main board, 8-port serial I/O board, 512KB memory
   boards and Archive (model?) tape controller board, what other I/O
   boards are known to exist, or to have existed?

From the writings of Tom Seabury, fd83@cityscape.co.uk:

"Our literature (CTIX basic admin course) suggests 9600 as a setting
for the terminal speeds, though this may not be the limit "For the S
Series (not entirely sur if this includes the mini) the macines may
also include space for expansion boards, those available are: memory
boards, RS232 boards, RS422 cluster boards includign a paralell
printer port, i/o processing boards includes a paralell printer port,
ethernet boards.  your machine has no vme interface, ours which does
allows us to have a smd disk drive, ours is a big heavy whirry thing
in its 19" rack.... totally over the top."

                        *   *   * 

From the writings of Clarence Dold, dold@rahul.net:

"Ethernet is the only one that comes to mind."

                        *   *   * 

--------------------------------------------------------------------------

10. What were the Miniframe's connections to the AT&T UnixPC machines?

From the writings of Clarence Dold, dold@rahul.net:

"Quite a few of the same design team.
At the time, Motorola made 68010 chips, AT&T owned UNIX, and they
both marketed machines made by Convergent ;-)

The MiniFrame was already on the market about June of 1984.
AT&T thought hat it could bring Unix to the desktop, and might
have, if the PC-AT hadn't come out at about the same time.
The UnixPC came out in the first quarter of 85.

The 68020 came out around Sep of 85, with a MightyFrame release in
early 86.  The MiniFrame was phased out in the second half of '86,
with some price pressure to try to get the cheaper machine to die
before the start of '87."

                        *   *   * 

--------------------------------------------------------------------------

11. What are the, if any, interesting, or unique, details about this
   machine?

--------------------------------------------------------------------------

12. Approximately how many of these machines were produced?

From the writings of Clarence Dold, dold@rahul.net:

"I believe about 10,000 were produced.  The last couple hundred
were given to the University alma mater of some of the engineering
team."

                        *   *   * 

--------------------------------------------------------------------------

13. Are schematics for this machine available from any known sources?

--------------------------------------------------------------------------

14. What hardware and software documentation exist for this machine?

From the writings of Tom Seabury, fd83@cityscape.co.uk:

"Apart from our basic administration course booklet we have;
  - ctix operating system version B volumes 1,2
  - ctix administartion tools manual
  - mightyframe series adminisrators reference"

                        *   *   * 

--------------------------------------------------------------------------

15. What interesting, or otherwise, hardware or software, hacks are
    known to have been performed on this machine?

From the writings of Jens-Uwe Mager,jum@anubis.han.de

The Unix used on that machine was derived from ATT System V.2, but unlike 
stock System V.2 Convergent did add virtual memory derived from BSD 4.1. 
Unfortunately they did not also add job control, so I remember using a 
locally hacked emacs and csh that did simulate "job control" with the 
ptrace system call (what a hack!).

                        *   *   * 

--------------------------------------------------------------------------

16. What is the 9-pin connector labeled "cluster" for?

From the writings of Tom Seabury, fd83@cityscape.co.uk:

"The 9 pin labeled 'cluster' is serial rs422 cluster ports, "used to
connect CT's PT & GT terminals" (?? not sure what these are).  "These
connections are known as cluster lines since several terminals may be
connected to each port"
...we never used this facility
"used ONLY for Convergent's PT or GT terminals only, daisychained to
multiple terminals"
on miniframe these are "tty020 onwards""

                        *   *   * 

From the writings of Clarence Dold, dold@rahul.net:

"Convergent Technologies included 307Kbps RS422 ports on all of their
machines.  In the proprietary CTOS-Intel product line, this was used for
peer-to-peer networking, as well as distributed client-server processing as
early as 1979.

In the Unix arena, it was never used for networking, except for a
diagnostic ability to clone a disk drive.  It was normally used to drive
the PT/GT series of terminals, up to 8 or 16 on a cluster connector.
The PT was an 80x25 character terminal.  The GT was an 80x25 or a 640x480
graphics terminal, although I don't know of any graphics other than demos
that ever ran on it.

With one or two terminals, the performance on a 307k line was sparkling,
compared to a 9600 baud terminal.  With more users, you were multiplexing
that 307k, so some tasks slowed down, but then there was only a 10MHz CPU,
driving 2MB of RAM, so multi-use was slow anyway.

Motorola only supported 4 terminals on the MiniFrames they sold.
Convergent supported 8 in the early days, 16 in later days, with an
additional cluster port on an expansion card.
In fact, the cluster line would talk to 16 devices."

                        *   *   * 

From the writings of Jens-Uwe Mager,jum@anubis.han.de:

"This was not used as generic network, it was used as a terminal 
concentrator. I did work on a Motorola labeled 6300 with TM30 Terminals 
(I do not know the original Convergent name for these) as my first Unix 
experience in 1984, and you could daisy chain these terminals to this 
single cluster connector (each terminal had an in and one out connector). 
Each of these TM30 terminals was capable to split the screen into up to 
four windows, each window was full width and could be scrolled to display 
the hidden lines. The terminals had 26 lines but did only display 25 at a 
time (scrolled depending upon cursor position), somewhat unusual and 
often annoying. The traffic on the cluster wire was really packetized, so 
each of the windows and the local serial port (bidirectinally if I 
remember right) could be addressed seperately without fearing of 
intermingling the data streams. I remember that there was an API to 
control these windows and that I had a few funny programs that could mess 
up anyone else's screen. Unfortunately I did not keep these hacks. These 
terminals did run mich faster than serial terminals, which usually had 
9600 baud these days, or 19200 if you used the termio EXTA speed. EXTB 
did make use of the external RS232 clock signal if I remember right."

                        *   *   * 

--------------------------------------------------------------------------

17. What type of PSU was used? What are ratings for the PSU?  Was more 
    than one type of PSU used?

The PSU (at least the one in my Miniframe) contains a label on one of
the capacitors which bears the following information:

   Manufacturer: Modular Devices
   Model: PS 2085 Rev. B
   P/N: 65-00041
   IN: 115/220VAC, 50/60HZ, 41./2.3A
   OUT: +12VDC@10A, +5VDC@27A, -12VDC@0.5A

   The internal fuse (P1) is rated at: 110V 7AFB, 220V 4AFB

   There's a paper sticker on the PCB which reads:
   "COMPONENT TYPE CUSTOM RECTIFIER LR38395"

If it makes any difference, the quality control inspection sticker
in the PSU shows the date of manufacture as July 23, 1986.

I'm not sure if this is the original PSU for this system or not.
Here's what one Miniframe owner, "The Color and the Surge",
(mute@sidehack.gweep.net), had to say about his system: 

    "I used to run a BBS from a MiniFrame with a hole chopped 
    in the side and an extra hard disk and power supply grafted 
    onto it... You see, the original power supply wasn't big 
    enough to run two of the bastard-huge MFM disks the MiniFrame 
    shipped with... If you tried it, they would burn right up."

                        *   *   * 

--------------------------------------------------------------------------

18. What is the male IDC connector, between the ST-506 data and control
   cables, on the main board, for? 

   The floppy drive connector (see question 0a).  

--------------------------------------------------------------------------

19. Are the PALs in the Miniframe copy-protected?

--------------------------------------------------------------------------

20. Was the Miniframe sold under any other manufacturer's labels
    (badge engineered)?

Yes, CT allowed such marketing practices to be used with their
machines.  This machine was also marketed as the Motorola 6300 and
6350, as well as possibly some model(s) from Calcomp.

Information for this topic was obtained from the writings of:
Philip M. Burton, pmb@netcom.com
   
--------------------------------------------------------------------------

21. Was Ethernet networking ever supported in any way, or was support
    for it ever considered?  Can an AT&T UnixPC ethernet board be hacked 
    into this machine without too much trouble?  

Yes, Ethernet boards exist for the Miniframe; here's some information
about them:

From the writings of Clarence Dold, dold@rahul.net:

"There is one I/O board above the CPU, below the memory cards.
Mine is one Ether Port, and eight 25 pin serial ports."

                        *   *   * 

From the writings of Tom Seabury, fd83@cityscape.co.uk:

"We have an ethernet card fitted to our S320 as a standard option." 

                        *   *   * 

--------------------------------------------------------------------------

22. In what year did the manufacture of these machines begin?  In what 
    year did production cease?

--------------------------------------------------------------------------

23. Comparing the Miniframe's processing speed to other machines.

The following benchmarks were found in a file on the 'net, but no
author's name was present in the text.  If anyone knows who performed
these benchmark tests and compiled them, please let me know so that I
may properly credit that person. 


                DHRYSTONE 1.1 BENCHMARK SUMMARY
                 Mon Sep 21 18:05:24 EDT 1987
                    SORTED BY MANUFACTURER

MANUF      MODEL      PROC     CLOCK NOREG   REG OS,COMPILER,NOTES
-----      -----      ----     ----- -----   --- -----------------
AT&T       3B2/300    32000     7.20   409   410 UNIX SVR2.0,cc  ,
AT&T       3B5        WE32000  10.00   578   573 UNIX SVR2 5.2.0.1 V2,cc?large,
AT&T       3B2/400    32100    10.00   672   683 UNIX SVR3.0,fpcc Release 1.0 ,
AT&T       3B1        68010    10.00   973  1033 UNIX System V Release 1,cc 1.5
AT&T       UNIX PC    68010    10.00   973  1034 UNIX 5.0.3,cc  ,
AT&T       3B2/400    32100    10.00  1108  1120 UNIX SVR2.0.4,cc  ,
AT&T       3B5        WE320xx   0.00  1212  1207 UNIX Sys V 2.0.1,cc 2.0.11.0
AT&T       3B15       32100    14.00  1797  1798 UNIX 5.2.1,cc  ,
Acorn      RISC       3 micron  0.00  2081  2147 vn 1.1,  ,
Apollo     DN3000     68020/68 12.50  2481  2481 Domain/IX SR9.5.Bl12,CC 4.58 w
Convergent MegaFrame  68010    10.00   747   783 CTIX 3.3,cc  -O,CTIX 3.3 
Convergent MiniFrame  68010    10.00   919   965 CTIX 3.2,cc  ,
DEC        PDP-11/45  ?         0.00   454   506 UNIX V7M,cc ? 256KB,
DEC        PDP-11/73  J-ll,w/F  0.00   772   875 UNIX 2.9BSD,cc  ,
DEC        PDP 11/44            0.00   884   951 UNIX Sys III,cc  ,
DEC        VAX-11/750           0.00   961   957 VMS 4.3,VAX C 2.0 ,
DEC        PDP 11/70            0.00  1019  1181 UNIX V7,cc  -O,
DEC        VAX 11/780           0.00  1243  1307 UNIX 4.2BSD,cc  ,
DEC        MicroVAX I           0.00  1361  1385 Mach 4.3,cc  ,
DEC        Micro VAX            0.00  1379  1394 Ultrix 1.1,cc  ,
DEC        MicroVAX I           0.00  1385  1399 Ultrix-32m 1.1,cc  ,
DEC        VAX 11/780           0.00  1417  1441 UNIX 4.2BSD,cc  ,
DEC        VAX-11/780           0.00  1526  1523 VMS 4.3,VAX C 2.0 ,
DEC        MicroVAX-2           0.00  1584  1584 Ultrix 1.1,GNU cc 0.9 -O,
IBM        PC/XT      8088      4.77   259   275 COHERENT 2.3.43,Mark Williams
IBM        PC/AT      80286     7.50   991   991 Venix/286 SVr2.3,lmccc AT&T 
IBM        PC/RT      (6150)w/  0.00  1537  1660 AIX SVR1,cc  ,
IBM        4381-2               0.00  4504  4901  ,c/370  ,
IBM        RT-PC      ROMP-C   10.00     0  6329 AIX 2.1,Advanced C 2.1 ,
IBM        3090/200             0.00 31250 31250  ,  ,
Plexus     P35        68000    12.50   835   894 UNIX SYS III,cc  ,
Sun        2/120      68010    10.00  1058  1142 UNIX Sun 2.2,cc  ,
Sun        3/50       68020    15.00  2280  2540 UNIX Sun 3.0,cc  ,
Sun        3/160      68020    16.67  2843  3134 UNIX Sun 3.0,cc  ,
Whitechape MG1        32016     8.00   636   675 UNIX 4.2BSD,cc  ,


--------------------------------------------------------------------------

24. Is Convergent Technologies still in business?  Do they still
    provide parts and documentation for these machines?  Where can
    these machines, and parts for them, be obtained?

The following is based upon the writings of Philip Burton, pmb@netcom.com:
   
A "trade rag" called "Unisys World" contains (contained?)  information
pertaining to products which were made by Convergent Technologies.
Information about this tabloid can be obtained by contacting the PR
department of Unisys in Blue Bell, Pennsylvania.  A used-equipment
tabloid called "The Processor" may contain ads for used CT equipment.

--------------------------------------------------------------------------

25. I've just obtined a Miniframe.  What's the first thing that I
    should do?

Thoroughly disassemble it and test the PSU (with dummy load
attached), removing any spiderwebs, dead, or live, arachnids 
or bugs or parts thereof from the system, etc.  

By thorough disassembly, I mean open the side panel, remove all
of the boards, remove the side panel (unscrew the hinges), remove
the power supply, remove the tape drive, open the power supply and 
disassemble it, open up the tape drive and disassemble it (not
an easy task -- sort of like a Chineese puzzle box, remove the 
RFI shielding, etc.  This is all important, as spiders (which
can cause short circuits), etc. could be lurking in many little 
places, particularly if your system has been sitting around unusued 
for a long time.  Besides, you'll want to do this anway just to 
learn some interesting things about the Miniframe.

I'm not going to tell you to be careful, lecture you on static
electricity, capacitors which may need discharging or spider
bites, or remind you to take detailed notes as you disassemble
the machine, as you should already know about such things if you 
use computers and have a sufficient number of properly functioning 
brain cells. :-)

Once you get the system apart and carefully cleaned, removing as
much oxidation, dust, rust, etc. as possible, and have checked
for burnt or frayed wires, etc., remove, back-up and re-insert
the EEPROMs and PALs (there are lots of these!) in case they
go bad at some point in future and need replacing.

Test the PSU with a dummy load.   Does the PSU fan spin freely?
If not, remove it, disassemble it, lubricate it, reassemble it
and reinstall it.  Reassemble the system.  After you've powered  
it up, give it the sniff test to make sure that nothing's burning.
You might also want to look out for smoke and sparks.

Aside from the EEPROMs and PALs, this appears to be a system for 
which repairs should be rather simple, as no difficult to find
custom ASICs, VLSI devices, etc. are used in it, and replacement 
parts should be easy to obtain.  

--------------------------------------------------------------------------

26.  Where can one obtain various bits of source code written
     specifically for the Miniframe?

     Once upon a time...

   |Subject:      [9]Unix Sources on ftp://ftp.uwo.ca/pub/unix
   |From:         [10]news@julian.uwo.ca (USENET News System)
   |Date:         1996/07/15
   |Organization: The University of Western Ontario, London, Ont. Canada
   |
   |                    pub/unix Directory Documentation
   |
   |                         1995-09-19 12:45:14
   |
   |This directory tree represents the CCS local source tree.  Programs
   |under this tree can be in any of three states.  The first is in an
   |archived format with the sources archived using tar and compress
   |(.tar.Z).  The second is in an unarchived state which will consist of
   |the sources in a subdirectory named after the program.  The last state
   |will be a subdirectory containing executables that have been compiled
   |for installation on local machines. The executable's subdirectory will
   |be named after the program and contain a .exe extension.
   |
   |
   |File            Description
   |==========================================================================
   |[...]
   |convergent      Various sources modified for a Convergent miniframe system

                        *   *   * 

   This part of the archive doesn't appear to exist any longer. :-(  Does
   anyone have copies of the files which were in it?  Do backup tapes of 
   these files exist?

--------------------------------------------------------------------------

27. Miniframe diagnostics tape.

Two dignostics tapes came with my Miniframe; both bearing the
same CT part number:

One is labeled "Miniframe Diagnostic Rev. 3.20" and had "Raw install"
hand-written on the tape box.  The other tape is labeled "Miniframe
Diagnostic 3.20 Dist. Tape".  As far as I can tell, both tapes appear
to be identical -- however, I seem to recall the person who gave my my
miniframe mentioning that there was some difference (if I recall
correctly -- I'll have to inquire about this).  When booting from the
diagnostic tapes, the following appears on the screen:

   MINIFRAME MC68010 DIAGNOSTIC DEBUGGER V1.0
   COPYRIGHT 1983 BY CONVERGENT TECHNOLOGIES, INC.

   MINIFRAME DIAGNOSTICS - V3.2(46)
   Parity interrupts disabled
   Memory tests will begin at 47000, end at 1FFFFC.

   command>

Entering a '?' at the "command>" prompt resulted in a menu of
diagnostic commands, for various tests, being displayed.  Some of the
tests include: clock test, map RAM tests, page protection tests, page
fault tests, memory read, write & refresh, page map tests, parity
tests, user I/O error test, RS-232 (8274) test, WD1010 drive tests.
There are also tests for a printer connected to the parallel port, an
RS-422 test, ethernet test and a tape drive test.  In addition, one of
the commands is for the formatting of a hard drive and the
installation of an OS.

Note -- the following error messages from the WD1010 tests will result
if the hard drive isn't connected properly:

Here are the results that I got for the 'w' and 'y' WD1010 tests:

Test: 7, Subtest: 1: Drive 1 (Disk): Can't Recal: Response = 1 

Test: 6, Subtest: 1: Drive 0 (Disk): Can't Recal: Response = 4

Note that a "Can't Recal" error may also appear if the hard drive
isn't connected properly, although it's possibly not to get this
error message if it isn't connected properly, but the result will
be that all of the sectors will appear to be bad during surface
analysis.


When I got these errors, I was certain that the hard drive which I was
using was good, and I knew that I had the red stripes of the cables
going the right way.  My suspicion was that the cables were somehow
connected wrong.

During the first try, I had the first data cable, from the back of the
system, and the end connector of the daisy-chain control cable,
connected to the hard drive.  The drive select jumper on the hard
drive was set for the first drive.  Something made me think that this
wasn't quite right.  As this is a control cable with twisted wires in
it, I recalled that this could mean that the drive select jumpers need
to be set up as for the second drive, with the cable-connectors used
being the determining factor of the actual drive numbering.  However,
this doesn't appear to be correct, as I learned:

From the writings of Tom Trebisky, tom@as.arizona.edu	

"Doesn't sound right, see above.  Jumper the drive as DS1, not DS2 as
the PC world does (since they like twisted cables, you know all this
eh?)."
 
                        *   *   * 

I also took a guess that the end connector of the data cable was for
drive 1 and the connector in the middle of the cable was for drive 0
-- this turned out to be correct.  Another guess, which turns out to
have been erroneous, was that the data cable closest to the back of
the system is for drive 0 and the other is for drive 1 -- the data
cables turned out to be just the opposite of my guess.

Now then, here's how the cables should be connected with one hard
drive attached and jumpered for DS1: The Drive 0 cable (see question
0a for more information on these cables), which is the second 20-pin
cable from the back of the system, and the middle connector of the
control cable, get attached to the hard drive.  Unless you've got a
defective hard drive, this should eliminate any "Can't Recal" messages
from the diagnostics, as well as the problem of the defects being for
for every cylinder, sector and track during surface analysis.

A bit more information about the cause of the "Can't Recal" message:

From the writings of Jens-Uwe Mager,jum@anubis.han.de

"Yep, the WD1010 is probably missing the track zero signal."

                        *   *   *             

---------------------------------------------------------------------------

28. CTIX installation and hard drive formatting.

a. Hard drive formatting

When one selects the 'i' command, for hard drive formatting and system
installation, using the diagnostics tape (or diskette?), one is
prompted for quite a bit of information.  How to answer some of the
following questions is obvious, but there are some that aren't so
obvious -- some which one can guess (right or wrong) at, and still end
up formatting the drive.  If you get confused, don't let this stop you
from experimenting.

One hard drive that I attempted to install in the Miniframe was a 43.5
MB Miniscribe 6053, which has the following parameters:

Cylinders: 1024
Heads: 5 
Reduced write current cylinder: 1025
Write precompensation cylinder: 512
Sectors per track: 17
Landing zone: 0

When 'i' is selected, the following prompts appear on the screen; my
answers to them are also shown.  I took a few guesses at some of the
following prompts (denoted by an '*').  Notes pertaining to a prompt
are a number enclosed by brackets.  Lines bracketed by "--->" and
"<---" are comments.

(Subtest 2) Formata Disk.
Do you want to format Drive0 (Winchester) (erasing contents)? y <CR>
Give # of Cylinders (RETURN = default of 645): 1024 <CR>
Give # of Tracks per Cylinder (RETURN = default of 7): 5 <CR>
Give # of Sectors per Track (RETURN = default of 17): <CR>

Cylinders=1024, Tracks=5, Sectors=17.
Give Pack Name (RETURN = default of Winche): <CR> 
Does the drive use head select bit 3 (hitech) (RETURN=Y): <CR> * [1]
Does the drive need write precompensation (RETURN=Y): <CR>
Which cylinder does write precompensation start at? Specify drive type:
[1]CMI         13MB   :Cyl   104  :
[2]RODIME      21MB   :Cyl     4  :
[3]MICROPOLIS  52MB   :Cyl   400  :
[4]SYQUEST      5MB   :Cyl   180  :
[5]ATASI       50MB   :Cyl   320  :
[6]HITCHI      50MB   :Cyl   256  :
[7]Other                     :7 <CR>
Please specify cylinder: 512 <CR>
Give step rate (RETURN = default of 0): <CR> *
Give size of partition 0 (RETURN = default of 40960): <CR> [2] 
Give size of BBT in blocks (Default = 1): <CR> [3] *
Do you want  a loader (Answer [Y|y] or [N|n]): y <CR> * 
Give start Logical Block (RETURN = default of 2): <CR> *
Give size in Blocks (Default = 12): <CR> *
Do you want a Dump Area (Answer [Y|y] or [N|n]): n <CR> *
Do you want a Down Load File (Answer [Y|y] or [N|n]): n <CR> *
---> Note: a delay of about 3-4 minutes occurs at this point       <--- 
Beginning Surface Test
---> If defects are found, they're listed here [4]                 <---
---> After a few more minutes, items such as the following appear: <---
The Bad Block Table contains 0 entries. [5]
          DISK TEST Drive0 (Winche)

No longer supported [6]
          DISK TEST Drive0 (Winche)           

No longer supported [6]

command>


Notes:

[1] - if this is answered with a 'n', the following prompts are
      substituted for some of the above prompts as follows:

      Does the drive need reduce write current (RETURN = Y): <CR> *
      Give Step Rate (RETURN = default of 0): <CR> *
      Give Size of partition 0...

[2] - selecting smaller number results in a prompt for the size of
      partition 2, selecting a number smaller than the default for
      the second partition results in a prompt for the size of 
      partition 3, etc.

[3] - BBT = Bad Block Table

[4] - The defects are listed in the form:

      Defect found t Cyl n  Trk n  Sec n (Physical Sector n)

[5] - Correct, according to the "zero defects" label on the hard drive.

[6] - I haven't figured out what these messges mean yet.  Just what is
      it that isn't supported?


b. Operating system installation

Now then, how does one load the operating system (e.g. CTIX) after
formatting the hard drive?  The CTIX tape is not bootable (at least
mine doesn't appear to be).  In addition to trying to boot from this
tape, I tried inserting it just after entering the 'i' for
installation during the diagnostics, so it would be there after the HD
was formatted, but that didn't work either.  Any suggestions?  Should
those "No longer supported" messages be appearing?  Just what is it
that isn't supported?

CTIX was distributed on floppies as well as on a tape:

From the writings of Jens-Uwe Mager,jum@anubis.han.de

"Hmm, we never had a CTIX on tape for that machine, we had a set of
floppies (I believe around 30 to 40 in 720k format) which we used for
installing."

                        *   *   *             

---------------------------------------------------------------------------

29. What are some hardware problems that may be encountered?

From the writings of Tom Trebisky, tom@as.arizona.edu	

"The second miniframe was actually Ill when I got it, but I succeeded
in fixing it.  It had a dead LS646 chip in the floppy dma circuit, and
would read a sector of bytes ABCDEFG as AACCEEGG until that chip was
replaced.  My little monitor was indispensible in figuring this all
out (not to mention the schematics and a logic analyzer)."

                        *   *   *             

---------------------------------------------------------------------------

30. What resources for CT Miniframe information be found from the Internet?

This FAQ, and possibly, eventually, other things, can be obtained via
the world wide web or FTP from:

http://www.access.digex.net/~rdd/miniframe
ftp.digex.net in /pub/access/rdd/miniframe/miniframe.v0.003.faq
                              (or whatever the most recent version is)

There may also be another Miniframe archive/web page appearing.
Watch this space for further details! 

==========================================================================
                             End of FAQ
==========================================================================