• Please review our updated Terms and Rules here

ADC Super Six software

I will look into mounting the IBM WD12 disk image in linux and kpartx and see if there is anything useful.
Currently I am using a gotek with a CPM 2.2 image to boot the SuperSix. I am trying to format/copy a new 8" floppy disk to boot from but the machine keeps having problems. After a few minutes the machine locks up or I get 'Cold Boot error' or things like that and have to reset the system. I have to do some more testing to determine if it is a memory problems or another bad chip. I also need to replace the capacitors that I snipped. Also still have the possibility of a short on the 5volt rail. I used an earlier suggestion about checking for hot chips. I have a thermal camera that showed a few PAL chips getting to 130 degrees F. I pulled them and for each chip I pull the current goes down about 20mA. The card idles at about 2 amps.
I am thinking Turbo Dos might be interesting since The Super six has all of the ram to play with. I am assuming I need a ADC SuperBios for TurboDos? I am still in search of more ADC SuperSix software. Right now I have 1 image to boot from.
More work ahead. I need to figure out this lock up problem and I will update my findings.
Thank you to everyone for the help.
 
Some news to report on the ADC SuperSix. I replaced the capacitors I had clipped and that did not change the problems. The SBC still freezes up.
first off if anyone knows of a good memtest program for CPM 2.2 that would be appreciated. I have done simple tests to determine why the machine likes to lock up.
My first test was typing DIR over and over again. After a few minutes and prompt became corrupt, what I typed in was not what was showing up on screen. eventually the computer would lock and have to reset. It seems like if I keep using the computer that problem stays away longer, keep running command and keep the CPU thinking. If I reset/boot and leave the prompt, the computer locks within a minute.
Where could this problem be arising from? To me it comes down to 2 things: Either there is a bad ram chip (I tested all chips individual in a ram tester) or a problem with the DRAM refresh.
Maybe if I am lucky someone has some ideas about where to look or what to test.
Thanks for all of the help!
 
Are your cpu and dma both set to 4MHz ? I would get lockups if either was set to 6MHz. Try jumpers C and D both on pins 2 and 3 for 4MHz

Larry G
 
Just checked it. Jumper D is on 2-3 and jumper C was on 1-2, switching it to 2-3 made no difference. I was excited that it was going to be an easy fix, but didn't change anything. Thanks for the post.
 
I fired up my ADC S6 system then booted cpm 2.2 then ran memtest.com from Sorcim which tests the 1st 64k bank continuous between the program end and bios start.
Each test is displayed along the bottom. I would assume it would stop and report an error. Rename memtest.txt to memtest.com. Hopefully it sneaks thru. If you want a bootable disk image let me know. .IMD imagedsk works best.
BTW my cpm 2.2 boot disks are 8" SSSD type A1 in 22DISK.

SORCIM MEMTEST.JPG
 

Attachments

Attached is an imagedsk of cpm 2.2 which I just booted on my system. It is 8" DSDD except track 0 side 0 is SSSD. The image is 22DISK definition ADV4.
Rename it to ADCPM22.IMD I had to use .TXT to get past the data police ...
 

Attachments

Thank you so much. I will try these out and report back. I started brute forcing my approach and started swapping out ICs on the board. I swapped out everything on the DRAM refresh side, U6, U18, U24, U28, U32, U33, U39, U40, U42, U44, U44. Did not make a difference. Hopefully a memtest program can shed some light.
 
I am trying to copy MEMTEST.COM onto a bootable disk but I don't have the 22disk definition for ADV4. If someone could post it that would be awesome. I don't have enough knowledge to create a definition.
Otherwise I need a bootable image with MEMTEST.COM already on it.
The machine locks up before I can use xmodem to do the transfer.
Thanks.
 
Thank you for the disk images and memtest.com and all of the help. The image I received booted nicely off a gotek. I could run memtest in loops with no problems. I tested from the defaults so it leaves everything below 100h alone and above EAOOh alone. As soon as I exited the program, the command prompt was non responsive. This still makes me think is a DRAM refresh problem. The memory seems fine as long as you use it. But if it sits it starts to change.
Using the Monitor ROM and I have been doing some testing. I will write a value into memory and let it sit and see what changes.
This first picture is what the memory looks like on a fresh boot before initialization
mem_unitit.jpg
After filling memory with A5
mem_a5_init.jpg
Let it sit a few minutes and look at the memory again
mem_a5_broke.jpg
It changes every other column to A4 and in groups of rows.
Here is the same memory filled with 00 and then read after sitting for 5 minutes
mem_00_broke.jpg
I am hoping someone could look at this and maybe guess where they think the memory is going wrong. Is it a fresh problem?
It is interesting how it is creating patterns in the memory. Even the uninitialized memory has a pattern to it.
Thanks for all the help.
 
I played with your Image and Track 37, Side 1 has 3 UNDEFINED Sectors:
IMAGE has a problem of undefined sectors:

37/1 D D D U U U D D

Code: 
# IMDA ADCPM22.IMD > ADCPM22.TXT
#
IMageDisk Analyzer 1.18 / Mar 12 2012
IMD 1.17:  2/06/2011 14:59:46

advanced digital

cpm 2.2

ser# 615-00127

dsdd

Required cylinders: 77
Required heads    : 2
Data rate         : 500kbps
Est. maximum track: 8957 bytes

Possible drives/options to write ADCPM22.IMD :

 3.5" HD 80-track   NOTE: *1 *2
   Double-step: OFF

 5.25" HD 80-track   NOTE: *1
   Double-step: OFF

 8" 77-track
   Double-step: OFF

*1 77 track image likely from 8" drive.
*2 Should fit on 360rpm drive, 300rpm drive will leave long end gap.



# IMDU ADCPM22.IMD ADCPM22.RAW /B /E /D >> ADCPM22.TXT
#
IMageDisk Utility 1.18 / Mar 07 2012
IMD 1.17:  2/06/2011 14:59:46

advanced digital

cpm 2.2

ser# 615-00127

dsdd

Assuming 1:1 for Binary output
 0/0 500 kbps SD  26x128 = 3328 Bytes  # Will DELETE these Bytes to get cpmls working
     1  2  3  4  5  6  7  8  9  10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
     D  D  D  D  D  D  D  D  D  D  D  D  D  D  D  D  D  D  D  D  D  D  D  D  D  D 
 0/1 500 kbps DD  8x1024 = 8192 Bytes
      1   2   3   4   5   6   7   8 
      D   D   D   D   D   D   D   D 
 1/0  D   D   D   D   D   D   D   D 
 1/1  D   D   D   D   D   D   D   D 
 2/0  D   D   D   D   D   D   D   D 
 2/1  D   D   D   D   D   D   D   D 
...
...

36/0  D   D   D   D   D   D   D   D 
36/1  D   D   D   D   D   D   D   D 
37/0  D   D   D   D   D   D   D   D 
37/1  D   D   D   U   U   U   D   D        <<-------------THREE UNDEFINED SECTORS
38/0  D   D   D   D   D   D   D   D 
...
...
75/1  D   D   D   D   D   D   D   D 
76/0  D   D   D   D   D   D   D   D 
76/1  D   D   D   D   D   D   D   D 
154 tracks(77/77), 1250 sectors (3 Unavail)



BEGIN COM9X  Compupro (Viasyn) 8/16 - DSDD 8" - 1024 x 8
DENSITY MFM, HIGH
CYLINDERS 77 SIDES 2 SECTORS 8,1024
SIDE1 0 1,4,7,2,5,8,3,6
SIDE2 1 1,4,7,2,5,8,3,6
ORDER SIDES
BSH 4 BLM 15 EXM 0 DSM 599 DRM 255 AL0 0F0H AL1 0 OFS 4
END

# COM9X  Compupro (Viasyn) 8/16 - DSDD 8" - 1024 x 8
diskdef com9x
  seclen 1024
  tracks 154
  sectrk 8
  blocksize 2048
  maxdir 256
  skew 3
#  DELETED 3328 Bytes using dd in linux - Now, get Directory Location of 0x2000
#
  offset 8192
  boottrk 0
#  boottrk 4
  os 2.2
end

# libdsk data below
[com9x]
description = COM9X  Compupro (Viasyn) 8/16 - DSDD 8" - 1024 x 8
sides = alt
cylinders = 154
heads = 2
secsize = 1024
sectors = 8
secbase = 1
datarate = ED

[com9x]
cyls = 154
heads = 2
secs = 8
interleave = 3
bps = 1024
id = 1
rpm = 360
#rpm = 300
rate = 250
mode = mfm
iam = no



Directory is at 0x2000 = 8192 Decimal
00002000   00 4D 4F 56  43 50 4D 20  20 43 4F 4D  00 00 00 4C  .MOVCPM  COM...L
00002010   04 00 05 00  06 00 07 00  08 00 00 00  00 00 00 00  ................
00002020   00 50 49 50  20 20 20 20  20 43 4F 4D  00 00 00 3A  .PIP     COM...:
00002030   09 00 0A 00  0B 00 0C 00  00 00 00 00  00 00 00 00  ................
00002040   00 53 55 42  4D 49 54 20  20 43 4F 4D  00 00 00 0A  .SUBMIT  COM....
00002050   0D 00 00 00  00 00 00 00  00 00 00 00  00 00 00 00  ................
00002060   00 58 53 55  42 20 20 20  20 43 4F 4D  00 00 00 06  .XSUB    COM....
00002070   0E 00 00 00  00 00 00 00  00 00 00 00  00 00 00 00  ................
00002080   00 45 44 20  20 20 20 20  20 43 4F 4D  00 00 00 34  .ED      COM...4
00002090   0F 00 10 00  11 00 12 00  00 00 00 00  00 00 00 00  ................
000020A0   00 41 53 4D  20 20 20 20  20 43 4F 4D  00 00 00 40  .ASM     COM...@



$ dd if=ADCPM22.RAW bs=1 skip=3328 of=TEST22.RAW



$ cpmls -f com9x -D 22.RAW
$ cpmls -f com9x -T raw,com9x -D TEST22.RAW

     Name    Bytes   Recs  Attr     update             create
------------ ------ ------ ---- -----------------  -----------------
ASM     .COM     8K     64   
CPM     .SYS    10K     72   
CPMINSTA.DOC     2K     13   
DDT     .COM     6K     38   
DEBLOCK .ASM    10K     80   
DISKDEF .LIB     8K     49   
DUMP    .ASM     6K     33   
DUMP    .COM     2K      4   
ED      .COM     8K     52   
FMTHD   .COM     2K     14   
FORMAT  .ASM    10K     65   
FORMAT  .COM     2K     10   
FORMAT  .DOC     2K     16   
INTRFACE.DOC     6K     44   
LDRGEN  .COM     2K      8   
LOAD    .COM     2K     14   
LOADER  .DOC     4K     21   
MODEM   .COM     8K     59   
MODEM   .DOC    12K     96   
MOVCPM  .COM    10K     76   
NEWSIZE .DOC     2K     15   
PIP     .COM     8K     58   
RAMDISK .MAC     4K     28   
READFRST.DOC     4K     17   
STAT    .COM     6K     41   
SUBMIT  .COM     2K     10   
SYSGEN  .COM     2K      8   
TRK0    .LDR     4K     20   
USER    .DOC     4K     26   
XSUB    .COM     2K      6   
Z80     .LIB    10K     78   
   31 Files occupying    168K,    1056K Free.



IMAGE has a problem of undefined sectors:
37/1  D   D   D   U   U   U   D   D


My File is attached. Download, rename to ADCPM22.ZIP, EXTRACT


Larry
 

Attachments

Interesting, the source of that image was from the Don Maslin archive which was done with Teledisk adc-cpm.td0. I don't know if the undefined sectors were from that image ?
I had imaged a floppy using Teledisk then pulled it off with Imagedsk but probably had used the disk for awhile 5 or 6 years ago. I added memtest.com using 22disk and gave
the OP an image of that in a private message. Is there a cp/m utility that will scan a disk and report which files have the bad sectors ? Thanks Larry K for checking that.

Larry G
 
Thanks for catching that. I will add the new image to my collection.

I have been thinking about this memory problem and I hope someone can expand on my thinking or correct me if I am wrong.
I believe that this is a refresh problem but I still can't wrap my head around how this could be happening. A refresh happens row by row. There are sections of memory that have alternating columns of good and altered data. A read or refresh copies the full row into the sense amplifier and if CAS is used it will output a single bit, either way it rewrites the row back. The memory seems to stay intact on direct reads but the refresh is not operating right.
It is also curious that the rows being altered in groups of 8. The columns will be groups of 8 also, 8 bits good, 8 bits altered.
Would the first bit on the address line being sticky cause this? if the first bit is also 1, it would skip every other address. Same with the CAS address.
Does this make sense? It is more of a rambling than a coherent statement but hopefully someone could expand on this idea.
I will hook a logic analyzer up to the addresses A0-A7 and see if it is incrementing through memory correctly.

Also, I have no idea how the bank selection is working on this. I know the card has 128K.. CPM only sees 64K (I think). Could this be a problem? maybe it is refreshing the wrong bank half the time?
 
cpm 2.2 will only see 64k. There is a cpm3.imd image on Bitsavers that requires 128k. That would be a different definition for the gotek.

Larry G
 
The ADC Super Six is running. It was a fault of my own. The board should have 4164 dram chips in it. I swapped out all of the gold ceramic chips for plastic style dips and ran them all threw my 4164 memory tester. Yesterday I noticed that there was a 41256 chip on the board. With further investigation I found another and also two 4264's. After replacing those with the right chips the board seems to run fine. I guess now I know what memory does with wrong sized chips. I assume the 41256's were not the problem but the 4264 has a different memory structure. I feel stupid for the mistake and thank everyone for the support while correcting self-induced problem.
To get back to more on topic of the original title, I have to figure out a well suited OS for this machine. Right now I have CPM 2.2 bootable but does not support the full 128K ram. I am thinking I need to move to CPM 3.x to make this happen. Any suggestions on what pairs the best with this board? I think TurboDos and MPM and other choices. I would like to use the machine as a single user environment and I would really like something that supports sub directories on the file system. Adding more than 128K of ram in the future would be nice too.
I tried formatting the disk drive and that did not work. I have a IBM WD12 disk drive and an MFM emulator with an image of the IBM drive. Using the FMTHD.COM command I set it to Type 1 ST-506 and it would give me an instant unrecoverable error at head 3 track 98. From what I have been reading it sounds like I need to configure CPM for that disk drive type, rebuild the kernel(?) and then I should be able to format the drive. Does this sound right? Maybe I need to see if I have a real ST-506/412 and image that and try it.
Is there a CPM 3 superbios bootable image somewhere? I found some turbodos stuff for the super six but not much about CPM 3.
Thanks for the help everyone.
 
Try this one. It's a bootable cp/m 3.0 banked 8" disk. track 0 heads 0 and 1 are SD 26x128 then the rest is DSDD 8x1024. There was a couple errors reading the boot track but I wrote this image back to a fresh
8" floppy and it booted fine. Maybe @ldkraemer can look it over. Change .TXT to .IMD for imagedsk.
 

Attachments

I bet that's where it loads ccp.com. Try this one. Rename to .IMD
I'll post a photo of the boot. CPM3 has multiple USER directories.
 

Attachments

Here are the Results for ADCCPM3.*

If you can get the Software from USER 4, I would like a copy of MON*.MAC extracted so I can look at those files.
For some reason cpmtools (ver 2.23) doesn't extract them properly.

Thanks.


Larry
 

Attachments

Last edited:
You must log in or register to reply here.
Back
Top