>But so much of the Apple II design was focused around minimizing chip usage and counted on exact timings that I don’t know that it would have been possible to change the clock speed.
A recent Adrian's Digital Basement video <https://www.youtube.com/watch?v=dt1eSXpo1SA> discusses this topic. While showing how a 80286 system runs at 1MHz, he discusses how the PC architecture allows (most) software to run at that clock speed while the Apple II architecture and software are inherently tied into the 1MHz clock speed.
In retrospect it seems so sensible to have the IIe and IIc in 1983 and 1984 move to, say, 2MHz, that I'm sure that fears of breaking software compatibility contributed to that from happening. (That almost certainly would have been a short-term problem. Given how quickly the Apple II software moved en masse to 128K/80 columns by the mid-1980s, developers would have accounted for a faster clock speed too.)
The III's II compatibility mode ramps that down to 1MHz, and has other restrictions to make sure that II software cannot use any III-only features.
Without III sucking up all of Apple's R&D budget and attention c. 1979-1980, the Apple II would surely have seen earlier enhancements. The II+ (1979) would likely have had lowercase and better keyboard (which did not occur until IIe in 1983), and a new model in, say, 1981 might have shipped with an optional Apple 80-column card (again, with the IIe in actuality). Built-in 128K RAM probably would not have occurred until 1984, akin to the IIc's introduction, but earlier support for RAM expansion alongside 80 columns is possible. One of these models would likely have had the 2MHz clock, too, while no II in actuality shipped with a faster clock until IIgs in 1986.
I never felt the IIgs to be an actual Apple II. It feels like a different computer that can almost accurately emulate a //e or //c, but has that inelegant impedance mismatch between. I also get the same impression from the Commodore 128, but feels even worse that it needs to drop to 1MHz to use the C64 video modes, and don’t even get me started about its Z-80 side.
A recent Adrian's Digital Basement video <https://www.youtube.com/watch?v=dt1eSXpo1SA> discusses this topic. While showing how a 80286 system runs at 1MHz, he discusses how the PC architecture allows (most) software to run at that clock speed while the Apple II architecture and software are inherently tied into the 1MHz clock speed.
In retrospect it seems so sensible to have the IIe and IIc in 1983 and 1984 move to, say, 2MHz, that I'm sure that fears of breaking software compatibility contributed to that from happening. (That almost certainly would have been a short-term problem. Given how quickly the Apple II software moved en masse to 128K/80 columns by the mid-1980s, developers would have accounted for a faster clock speed too.)