I really do not think that all good ideas can be cooked and published in under r6 months and in the window of time required to do a submission for a top conference.
What I think you're getting at is that there is little to no incentive to produce a high-quality version of a (rushed) conference paper.
There's no such time constraint.
In practice researchers will often have multiple multi-year projects in the pipeline, along with shorter projects, so that something publishable can be produced at regular intervals.