Make Your Commodore 16 64k, But Not A Commodore 64

The Commodore 16 was a budget home computer from the mid 1980s, the entry-level model in a wider range of machines. As its name suggests it only has 16k of memory in keeping with its budget status, and while it has the rest of the hardware necessary to run software intended for its 64k stablemates, that 16k is impossible to expand without modifying the machine. Should you have a ’16 in your collection this is not a particularly arduous process, and Tynemouth Software have gone into great detail over how it can be achieved.

As was quite common in machines of the period, the address lines for the RAM area above the fitted 16k are not wired to disable it when those addresses are selected, so the same 16k appears mirrored three times in the space between it and the 64k limit. Thus simply plugging in a 64k cartridge would result in the top 48k being unusable, and some means of disabling or supplanting the internal chips was called for. Contemporary upgrades required pin or track snipping, but as they go on to show us there are some less ugly alternatives both permanent and reversible. Whichever you might favor they all at least don’t carry the huge cost hurdle in 2019 that they might have been when the machine was new. Sadly even though their cases may be similar the resulting machine will not be a Commodore 64, not even a new one.

Long-time Hackaday readers will know that the hardware designer for these machines was our Hackaday colleague [Bil Herd], and all followers of Commodore and his work should read his account of the CES trade show at the heady height of Commodore’s  fame.

24 thoughts on “Make Your Commodore 16 64k, But Not A Commodore 64

      1. They are very different machines. All they really have in common is the same style case and keyboard, and a processor based on the 6502. Although they share the same video connector and IEC bus for disk drives and printers, the Datasette and joystick connectors are different. The C16 has the TED chip doing sound and video in place of the SID and VIC in a Commodore 64. The C16 has a better version of BASIC (V4) than the V2 in the C64 and VIC20, and other than simple BASIC programs, they are not software compatible.

    1. The idea of the Plus-4 was a different one, it was made to be a cheap business machine for small businesses that don’t need the power of an IBM. The C16 was created to compete with SInclair Spectrum.
      Both fell flat on their belly.

      1. Yeah. To be honest, the C16’s circuit is really indeed low-cost. Less components than a ZX Spectrum, I think. But then they added the expensive keyboard and I think that ruined the whole low-cost thing. Sinclair was smarter. They added a slightly less-than-total-crap keyboard, and offered a real keyboard as an upgrade. In fact, that was the success of Sinclair. Make a really as low possible-cost device, which does everything necessary, and which has a lot of software for it. That will make people buy the computer. And after a few weeks or months of fiddling, the people would have gotten used to their computer and have some software base, wouldn’t want to replace it and go over that learning-curve again, and so would buy expansions, starting with a keyboard and a drive.

        But Sinclair missed a few things as well. The base station was quite capable. But people wanted to upgrade their computer to a full and serious computer, if only to impress their neighbours. So they wanted a good-quality keyboard, and a real floppy drive. I think that Sinclair should have made the base unit as low-cost as possible (and they did), but should have made it possible to expand the spectrum into a serious-looking machine with serious hardware.

        People will rather expand their platform than replace it altogether. Because a new platform means a new learning-curve, having to start from almost scratch again. The fact that most people who had built up their Altairs and whatnot S100 computers did not massively replace their computers with Apple II’s and TRS-80’s 5 years later, but still kept expanding their computers for years to come, had already proven that.

        Of course this is all hindsight now, the CEO’s at that time had to invent the market as they were going. And they didn’t have Google yet, to keep track of all the new things that were going on. ;)

        1. They did actually have a Sinclair-ish C16, the C116. Also a cruddy keyboard to save on costs. The C116 was discontinued pretty quickly, though, because Commodore lost their way with the 264/TED series after Jack Tramiel left during the development of it. As Commodore engineer Bil Herd put it, marketing wanted to try creating another C64, while the 264 series was meant to supplement the C64, not replace it.

    1. Did the same with my Amiga 1000. One layer of bricks on the internal 256K ram, and one layer on the external 256K ram. :)

      But I also upgraded my C16 to 64KB, and it was easy. No bricklayering necessary. Just replace the 4416’s with 4464’s (taken from an old Hercules card in my case ;)), cut two traces, and solder 2 wires. Wish it was that easy with my Amiga. ;)

      Instead of connecting both A14 and A15 traces of both chips directly to 5V, they were connect to each other, and then connected to 5V. That saved the need to solder 2 extra wires and cut 2 more traces. Not sure if it was accidental or on purpose, but this made it the easiest memory upgrade of a home computer that I ever made. :)

      1. You kids and your easy upgrades! I turned a 16 kB Color Computer into a 32 kB one with the old stacking RAM trick. You had to lift a pin and solder one wire. (Can’t remember if I had to do that x4 or not…)

        512 kB RAM? Who would ever need so much?

    1. I would not recommend this unless you are experienced with desoldering, it’s easy to damage tracks on the C16 boards. There are some pictures in my blog post that is linked in the article of a C16 I had to repair because the owner had tried but it didn’t work. That was the unfortunate conclusion of my blog post, it’s possible but is always going to require removing (or at least disabling) the onboard RAM chips.

  1. The original C16 was a single sided PCB with insertable jumpers to get one trace over another, stuffed in a VIC20 case. Meanwhile we should have been making LCD machines instead but that’s a different story.

    1. Gods bless you, Bil.

      It’d be noice if some rich guy with more money than he knows what to do with would hand you a wad of money to start up a retro computer foundation that preserves all the apocrypha that you and others like you know about that age. Perhaps self-supporting by selling retro-computing kits to enthusiasts, along with fee-based seminars?

  2. Maybe I’m late but the people at Tynemouth didn’t realize the quick and easy way to disable the onboard RAM of a C16 is to install a 48 pin socket under the TED and cut pin 11 then just add a small 10k resistor between the 5V source of the cartridge port and FB40. No soldering required, you can use cheap logic analyzer probes to do it.

  3. When I was a kid, I upgraded my C16 with a small board which I bought from a Commodore 16 and Plus 4 shop based in Dartford. It went in between the Ted chip and it’s socket. You pulled the Ted chip, inserted it into the empty socket on the board and then put the board back into the Ted socket in the main board. There were four chips on the board. Two multiplexes and two ram chips. No soldering and no trace cutting. The only issue I had was that the board would pop out of the Ted chip socket, so I had to tie that in somehow. But apart from that, it worked perfectly. I used to play Ace and Mercenary on it.

Leave a Reply to DocCancel reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.