Yes I think that the formulas question type could do the job if you only use one subpart with multiple coordinates.

I will try to post an example using the equation you provided.

Thanks for making that Daniel, and sharing it. It is really neat. I nice example of what you can do with with a new behaviour plugin, and a very small amount of code.

If I might offer some unsolicited code-review, I think there is a slightly better way to do it. Instead of copying the whole method from qbehaviour_deferredfeedback to modify it, you could also do

class qbehaviour_deferredallnothing extends qbehaviour_deferredfeedback {
    public function process_finish(question_attempt_pending_step $pendingstep) {
        $keep = parent::process_finish($pendingstep);
        if ($keep == question_attempt::KEEP &&
                $pendingstep->get_fraction($fraction) != 1) {
            $pendingstep->set_fraction(0);
        }
        return $keep;
    }
}

The reason for preferring this method, is that it is more likely to go on working correctly even if the deferred feedback method is changed a bit (e.g. to fix a bug).

I am also failing to work out what the disregarded_info method in the renderer is for. Please can you put my out of my misery by explaining.