I appreciate that some effort went into coding this benchmark but I don't understand no time was spent trying to optimise each MQ. You could even email the dev

I appreciate that some effort went into coding this benchmark but I don't understand no time was spent trying to optimise each MQ. You could even email the developers and ask for help since you're going to be publishing this as a comparative benchmark.