Thanks a lot for the explanation!! Good to know that there's probably a lot of delay caused by the internal route to the macro. Is the design of the test chip available to see?