Osborne 1 Double Image Repair Advice Needed

[iamdarkyoshi]

So I managed to nab an OCC 1 at the local thrift store for 70$, which is quite a bargain, given how much these things go for...

I got an image at the thrift store and bought it, and brought it home. I powered it on and just left it, while I watched some videos about it.



After the rifa caps blew their shit out and scared the shit out of me, I replaced them with modern caps and continued to play with it.

Upon removing my head from my ass, I noticed that the screen is doubling up what's supposed to be on it,  but in a weird way.



My guess is maybe there's a bit on a data bus that's shorted/open, so stuff going on the first half of memory also ends up in the second half, or it reads the first and second halves the same, idk. My gut feeling tells me there's something wrong on a data line

The screen is quite strobey, and the normal and duplicate images seem to be alternating. I was able to capture an image of just one of the duplicate osborne splash screens by messing with the shutter on my phone, so it's almost like it draws one screen of data, then the same data again shifted halfway down. But the fact the boot error shows up 4 times normally and then on both duplicates the 5th time is what's really perplexing me. I don't have any disks for it yet, so I can't do much but make it throw boot errors.

Any advice on where to start on this thing? It's actually not in terrible condition, so I plan to restore it as best I can. First I'm starting with the electrical...

EDIT: I've ruled out the CRT. Composite video out does the same thing:

[dnwheeler]

Maybe a stuck address line (high or low) causing the memory to "wrap around".

[Kleinstein]

It could be a problem in the screen buffer / memory. In this case the later lines, beyond maybe line 16 should never show up on the screen. So a screen contend with maybe numbered line should show this.

Another possible problem could be trouble with the vertical sync. Old computers sometimes used odd screen formats (e.g. other display rates) and video signals that are not correctly detected by the modern monitor. This could cause such a double picture, a little like a scope picture with poor trigger.

[wilfred]

Although it can't be seen in the photo the OP says it is quite "strobey". I think it is not likely to be a RAM issue but a horizontal hold issue. Is there an adjustment for it around the back somewhere?

[iamdarkyoshi]

Maybe a stuck address line (high or low) causing the memory to "wrap around".

That's what I was thinking, but now I'm even more perplexed than when I started... Check out the video below

It could be a problem in the screen buffer / memory. In this case the later lines, beyond maybe line 16 should never show up on the screen. So a screen contend with maybe numbered line should show this.

Another possible problem could be trouble with the vertical sync. Old computers sometimes used odd screen formats (e.g. other display rates) and video signals that are not correctly detected by the modern monitor. This could cause such a double picture, a little like a scope picture with poor trigger.

The composite video out going to my LCD monitor has the same alternating between normal and shifted image effect the CRT does. See below video for more details on what effect I'm talking about

Although it can't be seen in the photo the OP says it is quite "strobey". I think it is not likely to be a RAM issue but a horizontal hold issue. Is there an adjustment for it around the back somewhere?

By strobey, I meant the normal image is only on screen 50% of the time, and the shifted image is there when the normal one isn't

I've captured some video including 240fps slow motion of what the screen is doing. There's some points where it stops double imaging, which perplexes me greatly.

Please read the video description to better understand what I've managed to capture on video,



I'm going to order some test/diagnostic/software disks for this thing soon so I can display items other than boot error.

[orbanp]

Nice toy!
I still have my O1 stashed away somewhere in the basement, got it in the early '80s.

The video display on the O1 only shows 24 rows by 52 characters, so it is possible that there is nothing wrong with the video output, just a "regular monitor" is not able to synch correctly to the non-standard video output.
(The video buffer is 32x128, the 24x52 "window" on the screen can be moved around the buffer with keystrokes.)

I did expand the video circuit to output the standard display (24x80?) on an external monitor, it did work, though the image quality was not good, it was flickering, if I remember correctly.

There is/was a technical manual floating around the web "o1techm.pdf", 590 pages, 18MB, you should find it, it describes everything in details.

One of the problems was with the disk drives, Osborne decided to make his own drive electronics, and it was rather flaky. I remember getting up in the middle of the night trying to retrieve a file from a floppy when the HW was cold, the drive just failed after a while when in regular use.
I started to re-engineer the floppy interface for a standard floppy drive. I completed the electronics, but probably it needed some change in the firmware as well.  I never completed that, I got a Mac SE-30 instead. The firmware listing is/was also around the net. It was not the source code listing, but the compiled firmware listing.

I did find an extra mobo in a surplus store on the way back, I did all the experimentation on that one.

Peter

EDIT:
Extracted and enclosing the video specs for the O1.
The horizontal sync is 64usec, the vertical is 16.6msec.
See if in the set up of your monitor you can select sync timings and if you can get close to these values.
You can also check with a scope if the video output sync signals are correct from the O1.

[Renate]

Well, it's either lost its vertical sync or else the reset circuit on the line divider is going past 260.
It's hard to tell from a static picture.

As it is 260 line non-interlaced I wouldn't put too much weight into the fact that a new digital monitor hates it.
If you have an old CRT monitor, they don't actually care about the relationship between the H & V sweeps.

If you have a scope, the video signal and the V sync will tell us.

If you don't have a scope, attach a big wire to the video signal and see if you can inject some hum into the signal.
The hum should run slowly through the picture.
If it runs quickly the line counter is screwed up.

Of course, if you have a DVM that does Hz you can just check the vertical sync.

If you have a good ear, just connect an earphone to the vertical sync.
The tone should be halfway between Bb and B.