Sunday, July 19, 2015

A CIA Mystery

My olde Trusty Commodore 128, a constant companion since the mid-1980s, recently developed a little problem.

The first sign was a locked-up IEC (disk drive) bus when booting into C128 mode, which locked up the computer as well.  If you don't know, this is possible because the C128 kernal checks for a bootable disk in device #8 at boot-time.  Anyway, the bus lock would prevent the computer from giving a prompt.  C64 mode, since it did not perform this initial check, was fine. Restarting the computer usually put things right, but it was still a concern.

Then, the other day I tried to boot the computer into Z80 (CP/M) mode.  After CP/M finished booting, I tried entering some commands to discover that the keyboard was unresponsive.  A quick check revealed that the keyboard still worked absolutely fine if the computer was booted into either C128 or C64 mode.

So, intermittent IEC bus problems, keyboard problems -- sounds like a bad I/O chip.  And in the Commodore 128, that means it has something to do with one of the two MOS 6526 CIA (Complex Interface Adapter) chips.

You can see the two CIA chips circled above.

Now, the problem could be something internal to the chip (meaning Bad Chip), or it could be that the chip is not properly seated.  There is also the question as to WHICH of the two is responsible for the problems.  Swapping them answers all those questions.  As a stroke of luck would have it, both of the C128 CIA chips are socketed! Woohoo! That means swapping is trivial.

So I swapped them.

And here's the crazy thing.  After the swap, the IEC problems went away, as did the problems with the keyboard when CP/M was booted.  However, now the keyboard is entirely non-functional in both C64 and C128 mode.

Clearly one of the CIA chips was bad, so I made an educated guess that the original bad chip was the one in the back left, replaced it, and now the computer is fine.

But what are the implications of the fact that a different CIA chip handles keyboard I/O depending on whether the Z80 or 8502 is active?

The CBM-Hackers mailing list is mostly unsure.  Bil Herd might know, so I may shoot him and email and post any updates if he has any insights.

Saturday, July 18, 2015

Completing the Set

Between 1997 and 2002, I collected well over a hundred Commodore computers, and dozens of disk drives, printers, monitors, modems, and other accessories.  By the time I stopped, my "Wanted" list had shrunk to almost nothing, and many of the things on it where only unicorn-like objects of rumor.

One of those computers was the Commodore 264, the namesake model of the "264 Series".  A set of computers based around the MOS 7360 "TED" chip, which grants the series the sometimes moniker of "TED Machines".  This series includes the Commodore 116, Commodore 16, Commodore 232, Commodore 264, Commodore Plus/4, and Commodore V364.

Luck and circumstance allowed me to find every machine on the list above, except one: The Commodore 264.  All my searching had only revealed one other owner, and while he was not selling, I did secure a promise that I would get first right of refusal when he does. :)

Thirteen years later, you'd think I'd have forgotten about it.  But a mindlessly OCD collector never forgets, and I finally struck gold on eBay:

This unit left the Commodore U.S. offices for a game company called Infocom, where it played some role in the porting of their famous Zork trilogy to the Plus/4.  A few years later, a fellow from Rhode Island was granted a tour of Infocom, his favorite game company.  He was given this computer, which the company no longer needed, as a gift.  After deciding it was not interesting, it went into storage until the owner decided the storage was a bit too full, and some of it needed to go.  So to eBay went this computer, and finally to my Commodore lab.

After taking it from the box, the first thing I noticed was that it needed a good cleaning, but that it was in remarkably good cosmetic condition.  An external inspection revealed that it did indeed share its case design with the Plus/4, with the exception that the power connector was an old round C64-style DIN connector, instead of the square custom connector common on the Plus/4s.

The case was held closed by some clear tape, which had long rotted away.  There were no screws holding the case together.  I would later surmise, when I put screws in, that it had probably never been screwed shut, like, ever.

After a quick visual inspection of the board, I decided that it was probably safe to attempt to turn it on.  I used a nearby C64c power supply and attempted to give it life.  Upon flipping the switch, the computer perfectly synced with my NTSC monitor, which wasn't surprising given its origin, but still gave me some relief.  The familiar 264 series color screen popped up, and BASIC appeared to initialize by placing its familiar message at the top of the screen.  Under that message was the READY prompt, but under that was a crash message and an immediate drop into the computers built-in machine language monitor.

I filed this information away, powered it down, and planned a more detailed inspection of the board itself.  While that was going on, the space bar and the four directional arrow key heads would be carefully removed for retr0brighting, since they showed signs of yellowing.

During my first board inspection, I noticed that the keyboard cable was not even connected to the board (the owner had been considerate enough to warn me about this), and several of the keyboard cable pens were bent up.  A light application of Superglue under the folded pins corrected this problem, so I turned my attention to the board itself:

Several interesting things popped out at me:

  • Every single IC is socketed! Repair will be simple when the time comes!
  • The TED chip (located upper-middle, in that square casing) had no protection from overheating, which, in the Plus/4, was provided by a metal contact that dips down inside the lid which covers that casing area, along with a dab of heat sink compound.  
  • The RF modulator, silver box in the upper-left, has "No 20" written on it.  Perhaps this was 264 board number 20?  Prototypes rarely had serial numbers, so I guess they needed to keep track somehow.
  • The KERNAL and BASIC ROM chips (bottom left, just to the left of those empty sockets) are, of course, only dated EPROMs.  They were still a work-in-progress at the time this unit was delivered.
  • The board assembly number is the same used in the Plus/4. No difference there at all.  And the empty sockets in the middle are already there, ready to receive the built-in software ROMs that given the Plus/4 its name.
  • The PLA chip (marked 1CCD), is also some sort of pre-production chip.
My next step would be to archive the KERNAL and BASIC ROM chips, which would also reveal which of the two, if either, caused the crash I observed during my initial test.

During removal of the BASIC ROM, I immediately saw a problem: pin 1 on the chip was bent up, and was either not making contact with the socket, or was making intermittent contact.  After removing the chip, I carefully bent it back into shape.

For reading in ROMs, I still use a Promenade C1 EPROM programmer on my Commodore 64c.  For software, I either use PROMOS 2.0 shell, or PROMSHELL 2.0, depending on the type of ROM that needs reading.  I had no trouble reading either chip, and saved them onto a 1581 3.5" floppy, for transfer to my PC and later uploading.

The ROMs were marked 1/19, which I surmised to refer to the snapshot date of the software on them, probably January 19th, 1984, which would be 3-4 months before release.

Comparing them with the ROMs from Jim Brain's Commodore 264 (marked 2/3) revealed that there are differences, though determining what they are and their significance will have to wait.

So, with my internal inspection and archiving done, and all the photographs taken, I carefully re-inserted the newly fixed keyboard cable, and re-connected the case.  Since it had previously been held together with tape, I scavenged some case screws from an old Plus/4 that was missing it's 7501 CPU altogether and used those to button it up properly.  The screws went in snugly, making me think that the case had never been screwed shut in its entire life.

Another power-on test showed that the bent pin on the BASIC ROM was, in the end, the only problem.  The computer was fully operational.  Further tinkering will wait until the differences in the BASIC and KERNAL ROMs are identified, so that I know what there is to tinker with.

Until then, I can bask in the satisfaction of having completed the 264 Series in my Commodore collection, and look forward to whatever secrets it may reveal in the future.

Monday, July 6, 2015

How to bathe a PET

This last weekend was crazy busy, with almost no time to do much of anything fun.

However, I was able to squeeze in a task that's been on my whiteboard for way too long -- giving my beautiful PET 2001-8 a good cleaning and retr0brighting (de-yellowing) the datasette keys. As you can see in the picture below, those keys really need it.

Once it was on the bench I opened it up, compress-air cleaned the board and case a little, cleaned the edge connector contacts with an eraser (yes, I keep unsharped pencils around JUST for this purpose), and removed the datasette for the final step.

The datasette in the original PET computers is a strange beast.  It's obviously some sort of OEM tape player that has been hacked in order to attach the custom cable.

Retr0bright is the greatest discovery of the 21st century for owners of older plastic electronics parts.  Far from being a simple "bleaching", Retr0bright actually de-yellows plastic surfaces, returning them to white if they were white, cream-colored if they were cream-colored, grey if they were grey, etc.

Although the theory behind it is a bit convoluted, and the recipes numerous and hotly disputed, I've found that Salon Care 40 Creme does everything those complex recipes do, and comes ready off the shelf.  All I do is paint the cream onto the yellowed surface, wrap the surface in cellophane to prevent drying, and put it under a pair of powerful UV-A-B lamps.  And while I'm not entirely certain even the lamps are necessary, I've never had the occasion to test it.

After the retr0brighting for 3-4 days, I reassemble the computer and smile.  Back to the PETting Zoo with you -- we'll play later.

Friday, July 3, 2015

Re-Introducing the C64 'Aldi'

When I first heard about the "Commodore 64 'Aldi'", it did not strike me as anything of particular interest to a collector.  The distinguishing feature of the computer, judging by the name at least, was that it was sold in a chain of stores in Europe.  Big Deal! Does that mean there's something interesting about having a Commodore 64 'Sears' or a Commodore 64 'Toys R Us?'.  Of course not.

And then, while perusing the sites of some European collectors, I discovered that the 'Aldi' always had a White 64C keyboard in its brown Breadbox case, and that it actually contained the same "short board" found in the 64C.  Now that's something a C64 'Sears' can't claim! So, while the notion of getting another PAL C64 did not overwhelm me with excitement (they are tricky for me to use here in the 60hz NTSC-dominated U.S.), I got interested enough to get one.  After chatting with the accommodating Stefan Walgenbach in Germany, we reached an agreement, traded some things, and I ended up with a a brown Breadbox PAL C64 with a short-board, in the original box!

Now, a decade goes by, and I end up in a discussion with a fellow on AmiBay about the 'Aldi' computer he has for sale.  "It's Genuine!" he says, because "it's not one of those West German made look-alike C64s that came later!"  So, there's a Look-Alike 'Aldi'?  And how the heck would he even KNOW?  I mean, No One keeps their store receipts when they buy these things, so who could have confirmed that THIS white-keyboard-brown-breadbox-short-board-64 came from the Aldi stores, but that one did not?

This led me to exchanging an email or two with a fellow in Germany named Stefan Vogt, who has apparently made a hobby out of gathering information about the Aldi, and through his Extensive Research, learned more than a few fascinating things about this computer that made it indeed worthy of the attentions of a Commodore enthusiast, especially if the 64 is your thing.

The punch line is  (paraphrasing Stefan's email here):

  • The 'Aldi' C64 was manufactured in the U.S. for the German market in the Summer of 1987, probably because the short-boards were brand new, and assembly was being kept closer to the engineers until it had been proven...
  • It was the First Commodore computer to feature the cost-reduced C64 short-board.  It was before the C64G, it was before the C64-II, it was before the C64C.
  • It had the earliest revision of that motherboard (250469 rev.3), one of the short boards to have a separate color-ram chip.
  • Contrary to early rumors, some of which came from the German 64er Magazine which reviewed these new C64 motherboards from Commodore, the Aldi *does* have 9 volts on the user-port.
  • And contrary to later rumors, the Aldi Look-Alikes made in West Germany are not true Aldis, having later-revision motherboards.
So, needless to say, I took the guy on AmiBay up on his offer.

Now, given all the information I'd been absorbing, I immediately endeavored to inspect and confirm some of the information I received.  Aside from the obvious 'Aldi' appearance, the first thing I confirmed is that, Yes, this German PAL computer was actually made in the U.S.   Also, something I'd never noticed before is that even the rubber "feet" of the computer are 64C white instead of the normal breadbox Black.

Next was to pop open the case and inspect the motherboard.  While attempting this, I confirmed that no previous owner had ever opened this case.  None.  Ever.  How do I know?  Because it took me a dadgum hour to get the case off after removing the screws -- the plastic latches were so tightly gripping the inside of the case that I had to carefully jimmy them lose with a tiny screwdriver to avoid damaging the case.

But once it was open, sure enough, I found it had the earliest known revision of the short-board c64 motherboard.

Further inspection of the motherboard shows that, indeed, there is a 2114 RAM chip present on this board that is  not found on the later Rev B motherboards.  You can see it below, it's between the VIC-II 8569 (dead center), and the larger PLA/integrated chip to the left of it.  As that chip is also seen in Rev.4 and Rev.A motherboards, I guess it wasn't integrated until the (final?) Rev.B boards.

Of course I also made sure the computer itself worked OK.  My PAL->NTSC converter does not exactly give a clean picture, but it was enough to see that the computer was working.  I also retr0brighted the keyboard, though it didn't need much.

In conclusion, I've definitely learned some interesting things about the later history of the Commodore 64.  Stefan makes the case for the extreme rarity of the 'Aldi' C64, and he may be right.  Although white-keyboard-brown-breadbox-short-board C64s seem at best uncommon, and Rev.3 short board motherboards have been found in all kinds of later C64 cases, including the C64-II/C64C., perhaps it's the strange combination of these things, reflecting an important transitional period in C64 manufacturing, that makes the 'Aldi' a special computer to make so much fuss about.