The original /// had a problem with the RAM module which plugged into the motherboard with the cheap tin-plated pins and sockets common at the time oxidizing, as well as chips walking up out of sockets. It did get pretty hot inside the case, since there was no airflow, and that aggravated the problems. Switching to higher quality sockets and connectors which could take a bit of heat solved the reliability problems, and Apple replaced the early boards for free.
The main reason the /// failed is Apple deliberately made it incompatible with software for the Apple II except for the most basic, low-end stuff. They didn't want customers running the Apple II Visicalc and accounting packages on the new machine. They believed people would replace their software just as Apple I owners had when the II came out. The difference is Apple I owners' libraries consisted of a half-dozen cassettes of utilities and a game of "Hunt the Wumpus". Apple II owners had large libraries which they were reluctant to discard, and there was very little software for the ///. The /// was also quite expensive compared to a II with an 80-column card and an accelerator which was functionally equivalent and in fact about twice as fast.
When the /// came out, I was very disappointed at its limits. Same for the IBM PC. It took the Mac and a year of software development and hardware hacking before my main computer shifted away from the Apple II. In fact, my early Mac programs were written and debugged on the Apple II with the S-C 68K cross assembler.
The main reason the /// failed is Apple deliberately made it incompatible with software for the Apple II except for the most basic, low-end stuff. They didn't want customers running the Apple II Visicalc and accounting packages on the new machine. They believed people would replace their software just as Apple I owners had when the II came out. The difference is Apple I owners' libraries consisted of a half-dozen cassettes of utilities and a game of "Hunt the Wumpus". Apple II owners had large libraries which they were reluctant to discard, and there was very little software for the ///. The /// was also quite expensive compared to a II with an 80-column card and an accelerator which was functionally equivalent and in fact about twice as fast.
When the /// came out, I was very disappointed at its limits. Same for the IBM PC. It took the Mac and a year of software development and hardware hacking before my main computer shifted away from the Apple II. In fact, my early Mac programs were written and debugged on the Apple II with the S-C 68K cross assembler.