Another Superboard II Rev B repair

[Sideburn]

Hi all, I’ve been trying to fix a Superboard II my friend gave me (his first computer) for the last two days and could use some help.

I have made a some good progress so far.

From the get go the board would power up and show the random characters on the screen but hitting break (cpu reset) would do nothing. I then probed the address and data lines on the cpu, ram, roms, etc and noticed no activity.

This is when I realized that the character generator chip was doing its own thing.

I put the 6502 into my Asteroids arcade board and powered up and all was good. CPU functioning.

I pulled all the 2114 RAMS out and tested all of them and they checked out good.

On to the ROMs...

I pulled the ROMS out and tested them on my EPROM programmer.

I was not able to read the monitor chip (a 2708) so I suspected it may be bad.

So I grabbed a newly erased 27C16 EPROM and programmed the SYN600 ROM into it from https://osi.marks-lab.com/software/roms.html

Once I put that in as soon as I hit the BREAK key the cpu and data bus comes alive with signals on the scope, I see the OE pin cycling on the monitor rom and data moving through the ram BUT the screen is clearing out all of the random characters and going BLANK.

Now the BASIC ROMs...

I WAS able to read all of the factory BASIC rooms 1,2, and 4 but 3 was giving me a read error.

And the ones I could read, the data does not verify against the roms I have on file that I downloaded from here: https://github.com/sjgray/OSI-monitor-roms and here: https://github.com/sjgray/OSI-monitor-roms but I am guessing that this just may be because I am selecting the wrong EPROM model number when trying to read the original masked roms, but the fact that ONLY rom #3 is failing to read makes me suspicious of that one being bad.

I tried programming BASIC rom #3 from the basic-roms.zip at osi.marks-lab.com but no luck however I am programming it into a 2716 so maybe I need to do some mods to the pins as mentioned here under the section "Using 2716 (or 2816) ROMs" : https://markcsele.ca/hobbies/my-vintage ... rboard-ii/

** Today I tried again with the pinout changes and still no luck. I have attached photos oh the pinouts if someone can check my work and confirm I have it correct.

In the meantime I figured I would hunt down exactly what is making the screen go from a page full of random characters (seemingly coming straight out of the character generator itself since I can pull all the roms and the cpu out), to a blank screen but I am having difficulty figuring out what's going on from he schematics.

Questions:

- Do I need the 4 BASIC ROMS installed in order to get anything on the screen or should I be able to jump into the monitor without them and prove that the monitor ROM is working?

- What controls whether the character generator auto generates random characters on the screen vs characters and graphics form the BASIC roms or monitor? I don't quite get how this is working. I know that the 8T28 buffers are links to the two 2114's (in my case 4 because this board has been modded to do 40 columns or more) that go to the character generator but I am struggling with sorting out how all this is working.

- What is U20 - U23 doing ?

- The SYNMON1 seems to be working but I tried a C1S monitor and it does nothing. Is this to be expected?

I think I am getting very close to a working board. There are a lot of bodge wires on the bottom and I see cut traces and piggy backed ram on the video ram so there's been some mods done to it. I have chips in all the PROTO areas and lots of wires coming off the U20-U23 chips.. I suspect a lot of this is for the extra columns mod done to the video.

Any help would be awesome.

[Sideburn]

The plot thickens.... Just poking around I pulled U65 and powered up and the video was scrambled and rolling diagonally but when I hit the break key the "C/W/M ?" showed up so now I know for sure the monitor is indeed running. Hitting a key does nothing though and i see no chip selects occurring on any of the BASIC roms.

So now I know I must have some sort of video issue and well as BASIC not running.

IMG_7481.jpeg (1.08 MiB) Viewed 817 times

IMG_7477.jpeg (2.17 MiB) Viewed 817 times

[bxdanny]

You need to have at least the BASIC 4 ROM working in order to get the D/C/W/M ? on the screen, because the Monitor ROM calls the routine at $BF2D in the BASIC 4 ROM to display that. But the screen will clear to blank if only the Monitor ROM is good.

This board seems to have been modified to have 2k of video RAM ($D000-$D7FF), rather than just the 1k ($D000-$D3FF) of the stock Superboard II. I count 51 characters visible horizontally by 30 vertically. That means the number of characters visible horizontally has been doubled, while the number vertically is normal.

The standard SYN600 ROM (working with BASIC 4) puts "D/C/W/M ?" on the screen starting at $D365, which should be not quite halfway down the screen. But your image shows "C/W/M ?" over halfway down. And I wonder what happened to the "D/". Actually, it may be that what is getting executed is really the code for the 4P and 8P machines (C2-4P and the later C4P, C2-8P and the later C8P), which used the 502 CPU board and 540 video board. The 540 had 2k of video RAM and showed a full 64 characters across the screen, and the ROM code for it (which is also contained in the SYN600 ROM) displayed just "C/W/M ?" (no D/), as the disk versions of those systems used a different CPU board without BASIC in ROM. It would also clear the full 2k of video RAM, whereas the stock code for the 600 would only clear 1k. So probably that is what is getting executed. And if the 4P/8P version of the keyboard-scanning ROM code is what's in place at $FDxx, then there will be no response to the keyboard, as it looks for non-inverted values from the keyboard (key pressed = logic 1), rather than the inverted values (key pressed = logic 0) used by the 600 board. But it writes "C/W/M ?" at $D740, which would be invisible to the left of the displayed 48 to 51 characters on the system you seem to have (and near the very bottom of the screen). Anyway, the video display on all these systems always operates even if the CPU is not running.

[bxdanny]

One quick follow-up to the previous message: I think the two halves of data in that 27C16 used for the Monitor ROM are swapped. Specifically, it is seeing what should be the first half ($F800-$FBFF) when it reads the second half ($FC00-$FFFF). (The contents of the first half are never actually used by the system, they are just there so that the same ROM chips can be used in the 4P/8P systems by wiring them up differently.) Another possibility might be if the A10 signal is not reaching the chip, and it is reacting as if that line were low.

[Sideburn]

You need to have at least the BASIC 4 ROM working in order to get the D/C/W/M ? on the screen, because the Monitor ROM calls the routine at $BF2D in the BASIC 4 ROM to display that. But the screen will clear to blank if only the Monitor ROM is good.

Yes I discovered I must have the ROM 4 to even get a blank screen. Well I thought it was blank but then I pulled U65 and make the screen roll and I see a C/W/M prompt but no D.

This board seems to have been modified to have 2k of video RAM ($D000-$D7FF), rather than just the 1k ($D000-$D3FF) of the stock Superboard II. I count 51 characters visible horizontally by 30 vertically. That means the number of characters visible horizontally has been doubled, while the number vertically is normal.

Yes if you look carefully at the board, the video RAMs have two more piggy backed on top of the originals. Also the PROTO areas are populated and I think that's what all the bodge wires are on the bottom of the board.

The standard SYN600 ROM (working with BASIC 4) puts "D/C/W/M ?" on the screen starting at $D365, which should be not quite halfway down the screen. But your image shows "C/W/M ?" over halfway down. And I wonder what happened to the "D/". Actually, it may be that what is getting executed is really the code for the 4P and 8P machines (C2-4P and the later C4P, C2-8P and the later C8P), which used the 502 CPU board and 540 video board. The 540 had 2k of video RAM and showed a full 64 characters across the screen, and the ROM code for it (which is also contained in the SYN600 ROM) displayed just "C/W/M ?" (no D/), as the disk versions of those systems used a different CPU board without BASIC in ROM. It would also clear the full 2k of video RAM, whereas the stock code for the 600 would only clear 1k. So probably that is what is getting executed. And if the 4P/8P version of the keyboard-scanning ROM code is what's in place at $FDxx, then there will be no response to the keyboard, as it looks for non-inverted values from the keyboard (key pressed = logic 1), rather than the inverted values (key pressed = logic 0) used by the 600 board. But it writes "C/W/M ?" at $D740, which would be invisible to the left of the displayed 48 to 51 characters on the system you seem to have (and near the very bottom of the screen). Anyway, the video display on all these systems always operates even if the CPU is not running.

Yeah it is odd because I have burned the SYN600 ROM and that's putting up "D/W/M ?"..
The original MON ROM was a 2708 EPROM (with the window) and it was bad.

So at this stage the board has the following issues:

After hitting BREAK, the screen goes blank (if i pull U65 and make the screen go wonky and scrolling diagonally I can see the C/W/M ? prompt wizzing by...

BASIC is not running. The chip selects on all of the BASIC ROMS is stuck low.
when i try to read the BASIC roms on my EPROM programmer by selecting a 2716 chip model all of them do pass the READ process but the data doesn't match against a verify of official ROMS and ROM#3 fails to do a read.

[Sideburn]

One quick follow-up to the previous message: I think the two halves of data in that 27C16 used for the Monitor ROM are swapped. Specifically, it is seeing what should be the first half ($F800-$FBFF) when it reads the second half ($FC00-$FFFF). (The contents of the first half are never actually used by the system, they are just there so that the same ROM chips can be used in the 4P/8P systems by wiring them up differently.) Another possibility might be if the A10 signal is not reaching the chip, and it is reacting as if that line were low.

Regarding my monitor ROM.. so the original EPROM this board had in it was a 2708 eprom. I am using a 2716... There are some bodge wires on the bottom of for this chip location I am assuming were done to get a 2708 to work. Maybe I need to change something here?

I am programming my 2716 with the SYN600 rom found here->https://osi.marks-lab.com/software/roms.html

[Sideburn]

Now that I know I do not need BASIC (except for #4) I will pull the others (since I suspect #3 is bad) and put #4 only in and confirm I can still enter a blank screen and see if the prompt is there when u65 is removed.

My board will not enter a blank screen if I do not have ROM4 in there..

[Mark]

If you are seeing C/W/M? then the wrong ROM page is located at $FF00. The SYN600 contains multiple ROM pages, some for other systems.

SYN600 - OSI multiboard Monitor ROM
page 000 'H/D/M' maps to $FF00 for a C2/C4 540Vid disk system
page 100 keypoller maps to $FD00 for a C2/C4 540Vid system
page 200 monitor maps to $FE00 for a C2/C4 540Vid system
page 300 'C/W/M' maps to $FF00 for a C2/C4 540Vid tape system
page 400 disk boot maps to $FC00 for a C1 system
page 500 keypoller maps to $FD00 for a C1 system
page 600 monitor maps to $FE00 for a C1 system
page 700 'D/C/W/M' maps to $FF00 for a C1 system

2708

2708-eprom.gif (1.67 KiB) Viewed 577 times

2716

2716-eprom.gif (1.65 KiB) Viewed 577 times

Pin 19 - should be wired to A10 or pulled up to cause ROM page 7 to show up instead of ROM page 3
C1P uses ROM pages 4 to 7 mapped to FC00-FFFF

[bxdanny]

My board will not enter a blank screen if I do not have ROM4 in there..

That actually confirms that you are running the reset code for the 4P/8P computers, as there is a call to the BASIC 4 ROM (for ACIA initialization) before the screenclear in that version.

Try bending up pin 19 (A10) of the monitor ROM and jumpering it high. I bet you will then see D/C/W/M ?, and even be able to enter the 65V monitor if you respond with M.

[Sideburn]

I tried with 19 pulled and same..

BUT I just did have a one time interesting occurrence.

I have now pulled pin 21, and 18 pins (i am using a socket and just bending those pins out) and connected 21 to +5 and 18 to U18 pin 3.

I hit break and the top half of the screen cleared and showed me the correct prompt.
I have not been able to reproduce it yet but luckily i snapped a photo.

IMG_7501.jpeg (1.74 MiB) Viewed 564 times

IMG_7502.jpeg (3.64 MiB) Viewed 564 times