In 1985, Commodore followed up on its best selling Commodore 64 with a model with twice as much memory. The design of the 128 offered the possibility of following up with a Commodore 256. It never happened. Instead, Commodore moved on from 8-bit machines after the 128. Here’s why.
The Commodore 256 successor to the C-128
Commodore did design a Commodore 256 in the 1986-87 timeframe. It had twice as much memory as the 128, and the Z-80 CPU was less crippled, so CP/M ran faster on the 256 than it had on the 128. The prototypes had a built-in 3.5-inch disk drive. Their appearance resembled a 128D with a 3.5-inch floppy in place of the built-in 5.25-inch floppy drive. So it kind of looked like the weird offspring of a 128D and an Amiga 1000.
The biggest problem with the Commodore 256 was that it sacrificed C-64 compatibility. The 64 had around 6,000 titles available for it in the mid 1980s versus about 300 C-128 titles. The 128’s key selling point was being able to run existing 64 titles. Sacrificing 64 compatibility to get better CP/M performance was a bad trade. The 256’s target audience was 64/128 owners, not Osborne owners.
The 128 never approached the C-64’s success, but lack of memory wasn’t the reason why. It wasn’t long after the 128’s release that inexpensive IBM compatibles started coming onto the market. Given the choice between spending $600 for a 64 setup, $900 for a 128 setup, or $1500 for an IBM compatible, many people went low with a 64 or high with a PC. The existence of a Commodore 256 wouldn’t have done much to change that. The Amiga 500 and Commodore’s PC clones stood a better chance of succeeding.
The other Commodore 256
Commodore also made another 256K 8-bit machine and actually sold it from 1982 to 1984. Known as the CBM 256, B-256, or CBM 720, it was a PET descendant, rather than a VIC-20 and 64 descendant.
This machine was directly related to the B-128 that Chicago-based Protecto Enterprises sold at closeout in the mid 1980s. Commodore never really gave this line much of a chance, as Commodore didn’t have enough manufacturing capacity left after building C-64s. Since Commodore ended up selling 20 million 64s, it’s hard to argue with that decision. Commodore went from being an also-ran in the business market to owning the home computer market, and at its high point that was good enough to give the company 38% of the computer market in total. Not bad.
This particular series is interesting for several reasons. It had a 6545 video chip in it like a PET, rather than the dual VIC-II/VDC setup like the 128 had. But it also had a SID chip in it, like the popular C-64 and 128. This series also paid a lot more attention to design than most other Commodore machines. Boston-area designer Ira Velinski designed these cases, and Commodore actually patented them. There is a rumor these cases were designed by Porsche, but Commodore ended up not using the Porsche design due to the cost.
These machines had one more interesting thing about them. Commodore built an 8088 CPU card for them, so you they could optionally run MS-DOS. Commodore never released the 8088 card for sale commercially. The result wouldn’t have been completely IBM compatible, so it’s anyone’s guess how commercially successful it could have been. History is littered with MS-DOS machines that were only somewhat IBM compatible, so it’s hard to imagine Commodore succeeding where DEC failed.
All of the machines in this family are rather rare today and highly collectible. The 256K versions are rarer than the 128K versions due to the high cost of memory in the early 1980s. Around 9,800 of them were produced, as opposed to 15,000 of the 128K models. In spite of the small quantity Commodore produced, it still took until 1987 for Protecto to liquidate them all. They were fine machines, but since they pretty much went straight into liquidation, there wasn’t much software available for them. With no software, there wasn’t a lot of market for them.