It both does and does not depend on the control, what the tests...

It both does and does not depend on the control, what the tests are trying to show is does it increase or not (1 sided tests are easier to prove) the relevant metric. If a certain application uses a mix that has a given metric, will adding X to it increase the performance or not. It's a matter of speculation whether the additive will have the same % effect on a different mix. Thus, would edencrete added to a mix that starts with a compressive strength at 48N/mm2 still increase this by 32%? At the end of the day you'd really need to add the stuffs to the same starting materials in order to compare them directly. What you shouldnt believe is that because the mix used by TLG ended with a much higher compressive strength that this means it's therefore better, that's not how it works, it also does not mean that because EDE had a higher % increase that it is better, as that also is not how it works. There is probably a relationship between base strength and effect of additive, but that could be positively or negatively correlated - there is no info that I can find on that. It's as Archer said, another hit.