You're probably right about that and it would be interesting experiment. However, if you make it adjustable it would be complex to get it to remain stable at that percentage. Of course you could use some other dividing methods to get different percentages. The single ff is simple and holds the percentage to a precise stable duty cycle no matter what the temperature.

One possibility is to use delay lines like Data Delay Devices makes, where you can digitally send instructions for how much delay to give. One example is their 16-pin 3D7418-0.25 which has 256 steps, 250ps apart, from 12ns to 75.75ns. Thermal coefficient is equivalent to a variation, over the 0C-70°C operating range, of ±3% from the room-temperature delay settings. Output rise and fall time are 2ns.

One of the ones that is mechanically adjustable like a 15-turn trimmer is the 16-pin DDU39F, variable from 7 to 25ns, spec'ed for 0-70°C, and has an output rise and fall time of 4ns.

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?